Re: [PSES] Thermal equilibrium - 10% rule
I do not submit algorithms to the agency, rather my raw data logs from start to end of the test. Also, any derived data which are produced by a spreadsheet to calculate acceptability of component ratings. My algorithms are nothing more that a test aid to estimate duration of extremely long tests and to assist in determining completion of a test (thermal equilibrium). This can be implemented in Excel, LabVIEW or other tools. A notable example is the Sandia National Laboratory efficiency test protocol for UL 1741 certified PV inverters. This requires data logging at 10 second intervals. If I happen to be monitoring several thermocouples this is impossible using traditional methods. Especially after about 8 to 10 hours of continual monitoring in 18 different operational settings. Automation is absolutely mandatory. Doug -- Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 Original Message From: oconne...@tamuracorp.com Sent: January 17, 2017 3:30 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Reply-to: oconne...@tamuracorp.com Subject: Re: [PSES] Thermal equilibrium - 10% rule This discussion thread referenced the automation of tests, not the acceptance of data. This is important where three or four different tests are being run on 10 to 20 units. Test data in 'raw' engineering units from a traceable instrument under controlled Type Test conditions is a different issue from process automation of data acquisition. Test automation had once been a recurring discussion during annual lab audits. My reply is typically that the test conditions met requirements of IEC XYZ, and lab and equipment meet ISO17025 tracibility, and here is the 'raw' test data, so please go away... And any case, methinks Doug is tracking this problem well and that NRTL engineers would be wise to understand what can be accomplished with this test data process. Time to step into the century number 21. Brian From: Brian Gregory [mailto:brian_greg...@netzero.net] Sent: Tuesday, January 17, 2017 1:26 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I worked for a NRTL for years, and no reviewer worth his salt ever accepted an equation for thermal time constant or steady-state temperature. Just the data, Ma'am Brian Gregory 720-450-4933 -- Original Message -- From: Doug Powell To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Date: Sun, 8 Jan 2017 11:38:30 -0700 Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities! are invo! lved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of the product T1 = The initial temperature of the product at the start T2 = The final temperature of the product abs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time. My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is par
Re: [PSES] Thermal equilibrium - 10% rule
This discussion thread referenced the automation of tests, not the acceptance of data. This is important where three or four different tests are being run on 10 to 20 units. Test data in 'raw' engineering units from a traceable instrument under controlled Type Test conditions is a different issue from process automation of data acquisition. Test automation had once been a recurring discussion during annual lab audits. My reply is typically that the test conditions met requirements of IEC XYZ, and lab and equipment meet ISO17025 tracibility, and here is the 'raw' test data, so please go away... And any case, methinks Doug is tracking this problem well and that NRTL engineers would be wise to understand what can be accomplished with this test data process. Time to step into the century number 21. Brian From: Brian Gregory [mailto:brian_greg...@netzero.net] Sent: Tuesday, January 17, 2017 1:26 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I worked for a NRTL for years, and no reviewer worth his salt ever accepted an equation for thermal time constant or steady-state temperature. Just the data, Ma'am Brian Gregory 720-450-4933 -- Original Message -- From: Doug Powell To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Date: Sun, 8 Jan 2017 11:38:30 -0700 Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are invo! lved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of the product T1 = The initial temperature of the product at the start T2 = The final temperature of the product abs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time. My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is part of the big picture. I have not fully tested this method yet, but so far I remain hopeful. If this works, I plan to build a small database of product Cp values. I would be interested to know if anyone else gives this a go and how as yet undiscovered problems are overcome. -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Sat, Jan 7, 2017 at 12:51 PM, Richard Nute wrote: > We have to consider that the temperatures sought are not of > metrological value, but to > to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to > estimate, once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to
Re: [PSES] Thermal equilibrium - 10% rule
I worked for a NRTL for years, and no reviewer worth his salt ever accepted an equation for thermal time constant or steady-state temperature. Just the data, Ma'am Brian Gregory720-450-4933 -- Original Message -- From: Doug Powell To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Date: Sun, 8 Jan 2017 11:38:30 -0700 Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are involved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joulesm = mass of the productCp = specific heat of the productT1 = The initial temperature of the product at the startT2 = The final temperature of the productabs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time.My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is part of the big picture. I have not fully tested this method yet, but so far I remain hopeful. If this works, I plan to build a small database of product Cp values. I would be interested to know if anyone else gives this a go and how as yet undiscovered problems are overcome. -Doug Douglas E PowellLaporte, Colorado USAdougp01@gmail.comhttp://www.linkedin.com/in/dougp01 On Sat, Jan 7, 2017 at 12:51 PM, Richard Nute wrote: > We have to consider that the temperatures sought are not of > metrological value, but to > to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to > estimate, once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: -- Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01- This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ie
Re: [PSES] Thermal equilibrium - 10% rule
We glue the ambient thermocouple to a small block of aluminum. It works well, and Al has a high Specific Heat. It’s something the regulator (agency) can “wrap their head around” without a lot of explanation. Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Saturday, January 07, 2017 12:39 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thermocouples for measuring oven temperatures are welded to a chunk of metal to provide thermal inertia. Without that, a normal thermocouple gives wildly fluctuating output, due to convection currents With best wishes DESIGN IT IN! OOO – Own Opinions Only http://www.jmwa.demon.co.uk/ J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Saturday, January 7, 2017 4:13 AM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Yes, My first experience with that was a 300 lb piece of equipment. I figured the thermal time constant should be many hours. It took me several hours of waiting to finally realize my ambient temps were also cycling. It was the air conditioning in my case as well. The German inspector from LGA suggested I put my ambient thermocouple in a glass vial with few cc's of machine tool oil. This did a nice job of smoothing my readings and drafts no longer had any effect. doug -- Douglas E Powell mailto:doug...@gmail.com http://www.linkedin.com/in/dougp01 From: mailto:peperkin...@cs.com Sent: January 6, 2017 8:36 PM To: mailto:doug...@gmail.com; mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: RE: [PSES] Thermal equilibrium - 10% rule Doug et al, Well, you don’t have to have 100K lbs of equipment to get porpoiseing. I remember a 500W product project where the test house engineer came in to witness the testing and was disturbed by the porpoiseing swings when the product temp seemed stable. He finally understood after the 2nd explanation that the swings were caused by the room AC cycling and he could see the result. Test successfully completed. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 mailto:p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 3:09 PM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA mailto:doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell <mailto:oconne...@tamuracorp.com> wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto:mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 11:11 AM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which
Re: [PSES] Thermal equilibrium - 10% rule
I would be interested to see it, please. With best wishes DESIGN IT IN! OOO - Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: ce-test, qualified testing bv - Gert Gremmen [mailto:g.grem...@cetest.nl] Sent: Monday, January 9, 2017 11:11 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Those interested, i cleaned up my temperature prediction spreadsheet and if you want to play with it, let me know. The math is crude, it might even have small errors. The spreadsheet is prepopulated in column E with data from an calculated RC network with source (green area), that you can erase, or use it to play with base data. The data is rounded to a settable number of digits. Best performance at time sampling of tau/50 and 2 digits minimum of resolution of the temperature ( x.xx degrees). At 3 digits the prediction is spectacular. Accuracy won't impact the results. It's the resolution that counts. With 1 digit of resolution the results are not very usable in terms of calculation, however your experienced engineers eye may draw conclusions anyway. After an arbitrary number of samples, the sheet calculates the final temperature tau , end of test time and creates a graph. Regards, Ing. Gert Gremmen Approvals manager + ce marking of electrical/electronic equipment + Independent Consultancy Services + Compliance Testing and Design for CE marking according to EC-directives: - Electro Magnetic Compatibility 2014/30/EC - Electrical Safety 2014/35/EC - Radio & Telecommunication Terminal Equipment 99/5/EC Web: www.cetest.nl (English) www.ce-test.nl (Dutch) www.cetest.fr (under construction) Phone : +31 10 415 24 26 --- This e-mail and any attachments thereto may contain information that is confidential and/or protected by intellectual property rights and are intended for the sole use of the recipient(s) named above. Any use of the information contained herein (including, but not limited to, total or partial reproduction, communication or distribution in any form) by persons other than the designated recipient(s) is prohibited. If you have received this e-mail in error, please notify the sender either by telephone or by e-mail and delete the material from any computer. Thank you for your co-operation. From: John Allen [mailto:jral...@productsafetyinc.com] Sent: Monday 9 January 2017 05:19 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule The subject itself can easily be a presentation at ISPCE. Doug, what you and Gert have done should be a formal paper. Please consider collaborating and making it happen. John From: Pete Perkins <0061f3f32d0c-dmarc-requ...@ieee.org> Sent: Sunday, January 8, 2017 9:33 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Doug, A great approach; it would make for an interesting ISPCE/PSES presentation and a paper. Go for it. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Sunday, January 8, 2017 10:39 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not co
Re: [PSES] Thermal equilibrium - 10% rule
Those interested, i cleaned up my temperature prediction spreadsheet and if you want to play with it, let me know. The math is crude, it might even have small errors. The spreadsheet is prepopulated in column E with data from an calculated RC network with source (green area), that you can erase, or use it to play with base data. The data is rounded to a settable number of digits. Best performance at time sampling of tau/50 and 2 digits minimum of resolution of the temperature ( x.xx degrees). At 3 digits the prediction is spectacular. Accuracy won't impact the results. It's the resolution that counts. With 1 digit of resolution the results are not very usable in terms of calculation, however your experienced engineers eye may draw conclusions anyway. After an arbitrary number of samples, the sheet calculates the final temperature tau , end of test time and creates a graph. Regards, Ing. Gert Gremmen Approvals manager + ce marking of electrical/electronic equipment + Independent Consultancy Services + Compliance Testing and Design for CE marking according to EC-directives: - Electro Magnetic Compatibility 2014/30/EC - Electrical Safety 2014/35/EC - Radio & Telecommunication Terminal Equipment 99/5/EC Web: www.cetest.nl (English) www.ce-test.nl (Dutch) www.cetest.fr (under construction) Phone : +31 10 415 24 26 --- This e-mail and any attachments thereto may contain information that is confidential and/or protected by intellectual property rights and are intended for the sole use of the recipient(s) named above. Any use of the information contained herein (including, but not limited to, total or partial reproduction, communication or distribution in any form) by persons other than the designated recipient(s) is prohibited. If you have received this e-mail in error, please notify the sender either by telephone or by e-mail and delete the material from any computer. Thank you for your co-operation. From: John Allen [mailto:jral...@productsafetyinc.com] Sent: Monday 9 January 2017 05:19 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule The subject itself can easily be a presentation at ISPCE. Doug, what you and Gert have done should be a formal paper. Please consider collaborating and making it happen. John From: Pete Perkins <0061f3f32d0c-dmarc-requ...@ieee.org> Sent: Sunday, January 8, 2017 9:33 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Doug, A great approach; it would make for an interesting ISPCE/PSES presentation and a paper. Go for it. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Sunday, January 8, 2017 10:39 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are invo! lved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of t
Re: [PSES] Thermal equilibrium - 10% rule
The subject itself can easily be a presentation at ISPCE. Doug, what you and Gert have done should be a formal paper. Please consider collaborating and making it happen. John From: Pete Perkins <0061f3f32d0c-dmarc-requ...@ieee.org> Sent: Sunday, January 8, 2017 9:33 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Doug, A great approach; it would make for an interesting ISPCE/PSES presentation and a paper. Go for it. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 p.perk...@ieee.org<mailto:p.perk...@ieee.org> From: Doug Powell [mailto:doug...@gmail.com] Sent: Sunday, January 8, 2017 10:39 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are involved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of the product T1 = The initial temperature of the product at the start T2 = The final temperature of the product abs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time.My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is part of the big picture. I have not fully tested this method yet, but so far I remain hopeful. If this works, I plan to build a small database of product Cp values. I would be interested to know if anyone else gives this a go and how as yet undiscovered problems are overcome. -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com<mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 On Sat, Jan 7, 2017 at 12:51 PM, Richard Nute mailto:ri...@ieee.org>> wrote: > We have to consider that the temperatures sought are not of > metrological value, but to > to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to > estimate, once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to mailto:emc-p...@ieee.org>> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Ins
Re: [PSES] Thermal equilibrium - 10% rule
Doug, A great approach; it would make for an interesting ISPCE/PSES presentation and a paper. Go for it. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 <mailto:p.perk...@ieee.org> p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Sunday, January 8, 2017 10:39 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe-(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are involved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of the product T1 = The initial temperature of the product at the start T2 = The final temperature of the product abs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time.My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is part of the big picture. I have not fully tested this method yet, but so far I remain hopeful. If this works, I plan to build a small database of product Cp values. I would be interested to know if anyone else gives this a go and how as yet undiscovered problems are overcome. -Doug Douglas E Powell Laporte, Colorado USA <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 On Sat, Jan 7, 2017 at 12:51 PM, Richard Nute mailto:ri...@ieee.org> > wrote: > We have to consider that the temperatures sought are not of > metrological value, but to > to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to > estimate, once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to mailto:emc-p...@ieee.org> > All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas mailto:sdoug...@ieee.org> > Mike Cantwell mailto:mcantw...@ieee.org> > For policy questions, send mail to: Jim Bach
Re: [PSES] Thermal equilibrium - 10% rule
Gert, As with Doug's method it would make for an interesting ISPCE/PSES discussion and paper. Put something together. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 p.perk...@ieee.org -Original Message- From: ce-test, qualified testing bv - Gert Gremmen [mailto:g.grem...@cetest.nl] Sent: Sunday, January 8, 2017 7:26 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Hi Rich, The quotient of subsequent derivatives of equidistant samples of the temperature has a direct relation to the time constant. TC=Timestep/(1-(dtempn+1/dtempn)) Numerically this is easy to calculate in a spreadsheet Once the time constant is known it is easy to find the temperature value of the 63.2% of the final temperature. samples: time1 - Temp1 time2 - Temp2 (Temp2-Temp1) - - time3 - Temp3 (Temp3-Temp2) - - dtemp2/dtemp1 time4 - Temp4 (Temp4-Temp3) - - dtemp3/dtemp2 The quotients in the last column are constant for a true inverse exponential temperature rise curve (same as a RC circuit). Of course , only for simple constant power heating curves that are truly exponential. The third column will quickly show if the curve is exponential or not , if the quotients are not all the same, something more complex is happening. I hope this is a clear explanation Regards, Ing. Gert Gremmen, BSc g.grem...@cetest.nl www.cetest.nl Kiotoweg 363 3047 BG Rotterdam T 31(0)104152426 F 31(0)104154953 Before printing, think about the environment. -Oorspronkelijk bericht- Van: Richard Nute [mailto:ri...@ieee.org] Verzonden: zaterdag 7 januari 2017 20:51 Aan: EMC-PSTC@LISTSERV.IEEE.ORG Onderwerp: Re: [PSES] Thermal equilibrium - 10% rule > We have to consider that the temperatures sought are not of > metrological value, but to to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to estimate, > once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Richard, I have tried a number of approaches in the past. Given that most products are quite complex with regard to all the potential heat sources/sinks and interfaces I decided that anything along the lines of FEA is impractical. I also tried the time constants idea which is analogous to RC time constants. I found this works well enough if you have a good amount of history with the product itself. Otherwise, due to the non-linear nature of the problem, it is difficult to predict end time or temperature until 3 to 4 time constants have already passed. I tried using the slope of ΔT to estimate when the end of the test is pending. The next attempt was to dig in a little following the equations V = Voe -(t/RC) and V = Vo[1-e-(t/RC)] where I substitute V for the the various temperatures (Vo = the absolute value of the temperature delta from start to end), C is analogous to product mass and R is the Rtheta of the product. With a little testing history, you can assume the composition of the product is similar for other products designed by the same company (copper, steel, plastics, air, liquids, etc), I solved for RC and then rearranged the algebra to solve for t which is time. There are a couple of problems in that I am still unable to come up with a general purpose solution. First this is a simultaneous solution of several unknowns which is not conducive to quick on the fly solutions. This is especially true when you are in the early stages of a temperature run when things are still moving quickly. As you know extrapolating outside an existing dataset is risky, especially when nonlinearities are involved. I am now going back to basics. Q = Cp * m * abs(T2-T1) q = heat energy in Joules m = mass of the product Cp = specific heat of the product T1 = The initial temperature of the product at the start T2 = The final temperature of the product abs() is used to correct for heating or cooling With the start/final temperatures and mass taken from prior tests I can extract a Cp for a particular product. Understanding one watt is Joules/second you can factor into the equation time.My thought is that the composition of a product from the same engineering group with have similar ratios of copper, iron, plastics, etc. And then I may be able to solve for total test time or final temperature. Not forgetting that the air mass and equipment of the environmental chamber is part of the big picture. I have not fully tested this method yet, but so far I remain hopeful. If this works, I plan to build a small database of product Cp values. I would be interested to know if anyone else gives this a go and how as yet undiscovered problems are overcome. -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Sat, Jan 7, 2017 at 12:51 PM, Richard Nute wrote: > > We have to consider that the temperatures sought are not of > > metrological value, but to > > to establish a safe/non-safe result. > > Yes! > > > The mathematical limit of an exponential rise is easy to > > estimate, once a few timed samples are available, > > I haven't been able to come up with an equation, even though I have tried > and sought help from folks who are more knowledgeable than me in the field > of thermodynamics. Please tell us your methodology. > > Best wishes for the New Year! > Rich > > - > > This message is from the IEEE Product Safety Engineering Society emc-pstc > discussion list. To post a message to the list, send your e-mail to < > emc-p...@ieee.org> > > All emc-pstc postings are archived and searchable on the web at: > http://www.ieee-pses.org/emc-pstc.html > > Attachments are not permitted but the IEEE PSES Online Communities site at > http://product-compliance.oc.ieee.org/ can be used for graphics (in > well-used formats), large files, etc. > > Website: http://www.ieee-pses.org/ > Instructions: http://www.ieee-pses.org/list.html (including how to > unsubscribe) > List rules: http://www.ieee-pses.org/listrules.html > > For help, send mail to the list administrators: > Scott Douglas > Mike Cantwell > > For policy questions, send mail to: > Jim Bacher: > David Heald: > -- Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules
Re: [PSES] Thermal equilibrium - 10% rule
Hi Rich, The quotient of subsequent derivatives of equidistant samples of the temperature has a direct relation to the time constant. TC=Timestep/(1-(dtempn+1/dtempn)) Numerically this is easy to calculate in a spreadsheet Once the time constant is known it is easy to find the temperature value of the 63.2% of the final temperature. samples: time1 - Temp1 time2 - Temp2 (Temp2-Temp1) - - time3 - Temp3 (Temp3-Temp2) - - dtemp2/dtemp1 time4 - Temp4 (Temp4-Temp3) - - dtemp3/dtemp2 The quotients in the last column are constant for a true inverse exponential temperature rise curve (same as a RC circuit). Of course , only for simple constant power heating curves that are truly exponential. The third column will quickly show if the curve is exponential or not , if the quotients are not all the same, something more complex is happening. I hope this is a clear explanation Regards, Ing. Gert Gremmen, BSc g.grem...@cetest.nl www.cetest.nl Kiotoweg 363 3047 BG Rotterdam T 31(0)104152426 F 31(0)104154953 Before printing, think about the environment. -Oorspronkelijk bericht- Van: Richard Nute [mailto:ri...@ieee.org] Verzonden: zaterdag 7 januari 2017 20:51 Aan: EMC-PSTC@LISTSERV.IEEE.ORG Onderwerp: Re: [PSES] Thermal equilibrium - 10% rule > We have to consider that the temperatures sought are not of > metrological value, but to to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to estimate, > once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
> We have to consider that the temperatures sought are not of > metrological value, but to > to establish a safe/non-safe result. Yes! > The mathematical limit of an exponential rise is easy to > estimate, once a few timed samples are available, I haven't been able to come up with an equation, even though I have tried and sought help from folks who are more knowledgeable than me in the field of thermodynamics. Please tell us your methodology. Best wishes for the New Year! Rich - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Is there any safety standard that requires black-box testing? Normally, the manufacturer has to disclose a lot of internal detail about the product. For EMC, it's a bit different. A disclosure is still required, but the test house may well look for effects that the manufacturer doesn't mention or might not have thought of. Under current European rules, the test house can consider such things, but may carry out tests only if the manufacturer agrees. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: ce-test, qualified testing bv - Gert Gremmen [mailto:g.grem...@cetest.nl] Sent: Saturday, January 7, 2017 11:32 AM To: John Woodgate ; EMC-PSTC@LISTSERV.IEEE.ORG Subject: RE: [PSES] Thermal equilibrium - 10% rule We have to consider that the temperatures sought are not of metrological value, but to to establish a safe/non-safe result. This means that the whole time constant discussion is relevant only when the final expected temperature is close to the standards limit value. The mathematical limit of an exponential rise is easy to estimate, once a few timed samples are available, and if that value is far enough above or below the limit, a conclusion is easy to draw. For cyclic heating equipment, a similar approach is not difficult. If the cycle is unknown , there is no other option than to wait infinitely. Unfortunately, any cyclic information is unknown if the test engineer is not involved with the operating principle of the EUT, so if following a black box testing approach, the measurement will never get to completion ! ;<) Regards, Ing. Gert Gremmen Approvals manager + ce marking of electrical/electronic equipment + Independent Consultancy Services + Compliance Testing and Design for CE marking according to EC-directives: - Electro Magnetic Compatibility 2014/30/EC - Electrical Safety 2014/35/EC - Radio & Telecommunication Terminal Equipment 99/5/EC Web:www.cetest.nl (English) www.ce-test.nl (Dutch) www.cetest.fr (under construction) Phone : +31 10 415 24 26 --- This e-mail and any attachments thereto may contain information that is confidential and/or protected by intellectual property rights and are intended for the sole use of the recipient(s) named above. Any use of the information contained herein (including, but not limited to, total or partial reproduction, communication or distribution in any form) by persons other than the designated recipient(s) is prohibited. If you have received this e-mail in error, please notify the sender either by telephone or by e-mail and delete the material from any computer. Thank you for your co-operation. From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Saturday 7 January 2017 00:32 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts:
Re: [PSES] Thermal equilibrium - 10% rule
We have to consider that the temperatures sought are not of metrological value, but to to establish a safe/non-safe result. This means that the whole time constant discussion is relevant only when the final expected temperature is close to the standards limit value. The mathematical limit of an exponential rise is easy to estimate, once a few timed samples are available, and if that value is far enough above or below the limit, a conclusion is easy to draw. For cyclic heating equipment, a similar approach is not difficult. If the cycle is unknown , there is no other option than to wait infinitely. Unfortunately, any cyclic information is unknown if the test engineer is not involved with the operating principle of the EUT, so if following a black box testing approach, the measurement will never get to completion ! ;<) Regards, Ing. Gert Gremmen Approvals manager + ce marking of electrical/electronic equipment + Independent Consultancy Services + Compliance Testing and Design for CE marking according to EC-directives: - Electro Magnetic Compatibility 2014/30/EC - Electrical Safety 2014/35/EC - Radio & Telecommunication Terminal Equipment 99/5/EC Web: www.cetest.nl (English) www.ce-test.nl (Dutch) www.cetest.fr (under construction) Phone : +31 10 415 24 26 --- This e-mail and any attachments thereto may contain information that is confidential and/or protected by intellectual property rights and are intended for the sole use of the recipient(s) named above. Any use of the information contained herein (including, but not limited to, total or partial reproduction, communication or distribution in any form) by persons other than the designated recipient(s) is prohibited. If you have received this e-mail in error, please notify the sender either by telephone or by e-mail and delete the material from any computer. Thank you for your co-operation. From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Saturday 7 January 2017 00:32 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 11:11 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much t
Re: [PSES] Thermal equilibrium - 10% rule
Thermocouples for measuring oven temperatures are welded to a chunk of metal to provide thermal inertia. Without that, a normal thermocouple gives wildly fluctuating output, due to convection currents With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Saturday, January 7, 2017 4:13 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Yes, My first experience with that was a 300 lb piece of equipment. I figured the thermal time constant should be many hours. It took me several hours of waiting to finally realize my ambient temps were also cycling. It was the air conditioning in my case as well. The German inspector from LGA suggested I put my ambient thermocouple in a glass vial with few cc's of machine tool oil. This did a nice job of smoothing my readings and drafts no longer had any effect. doug -- Douglas E Powell <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 From: peperkin...@cs.com <mailto:peperkin...@cs.com> Sent: January 6, 2017 8:36 PM To: doug...@gmail.com <mailto:doug...@gmail.com> ; EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: RE: [PSES] Thermal equilibrium - 10% rule Doug et al, Well, you don’t have to have 100K lbs of equipment to get porpoiseing. I remember a 500W product project where the test house engineer came in to witness the testing and was disturbed by the porpoiseing swings when the product temp seemed stable. He finally understood after the 2nd explanation that the swings were caused by the room AC cycling and he could see the result. Test successfully completed. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 <mailto:p.perk...@ieee.org> p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 3:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell < <mailto:oconne...@tamuracorp.com> oconne...@tamuracorp.com> wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto: <mailto:doug...@gmail.com> doug...@gmail.com] Sent: Friday, January 06, 2017 11:11 AM To: <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Tho
Re: [PSES] Thermal equilibrium - 10% rule
You would have to be prepared to explain to a regulator how and why you did what you did. So don't make it too clever. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: Brian O'Connell [mailto:oconne...@tamuracorp.com] Sent: Friday, January 6, 2017 11:53 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Correct, so for multiple duty cycles or varying loading intervals, possible solution would be to use windowed average (queue size based on 1/f and dT/dt and previous sample interval) with a 'delayed' least-squares fit thru the maxima to see if flat. [hear the ghosts of my math instructors yelling at me to use polynomial regression]. Brian From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 3:32 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Yes,My first experience with that was a 300 lb piece of equipment. I figured the thermal time constant should be many hours. It took me several hours of waiting to finally realize my ambient temps were also cycling. It was the air conditioning in my case as well. The German inspector from LGA suggested I put my ambient thermocouple in a glass vial with few cc's of machine tool oil. This did a nice job of smoothing my readings and drafts no longer had any effect. doug-- Douglas E Powelldoug...@gmail.comhttp://www.linkedin.com/in/dougp01 From: peperkin...@cs.comSent: January 6, 2017 8:36 PMTo: doug...@gmail.com; EMC-PSTC@LISTSERV.IEEE.ORGSubject: RE: [PSES] Thermal equilibrium - 10% rule Doug et al, Well, you don’t have to have 100K lbs of equipment to get porpoiseing. I remember a 500W product project where the test house engineer came in to witness the testing and was disturbed by the porpoiseing swings when the product temp seemed stable. He finally understood after the 2nd explanation that the swings were caused by the room AC cycling and he could see the result. Test successfully completed. :>) br, Pete Peter E Perkins, PEPrincipal Product Safety & Regulatory Affairs ConsultantPO Box 23427Tigard, ORe 97281-3427 503/452-1201 p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 3:09 PMTo: EMC-PSTC@LISTSERV.IEEE.ORGSubject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel _vbscript_ to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E PowellLaporte, Colorado USAdoug...@gmail.comhttp://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell <oconne...@tamuracorp.com> wrote:Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions:time constant = (material ksp * mass) / (24*60)sample interval = time constant / (mass * material kx)breakpoint flags = falseif sample interval < min interval sample interval = min intervalinterrupts: temperatures to circular buffer log samples and windowed averages to network storageloop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate doneBrianFrom: Doug Powell [mailto:doug...@gmail.com]Sent: Friday, January 06, 2017 11:11 AMTo: EMC-PSTC@LISTSERV.IEEE.ORGSubject: Re: [PSES] Thermal equilibrium - 10% ruleJohn,I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible.On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". -DougDouglas E PowellLaporte, Colorado USAdoug...@gmail.comhttp://www.linkedin.com/in/dougp01On Fri, Jan 6, 2017 at 11:30 AM, john Allen <john_e_al...@blueyonder.co.uk> wrote:Minor comment:Having been the Secretary to a number of BSI committees many years ago, Isympathise with Ralph's last para as the Chairmen can be even worse than theother Members - they "know what they know" and it can take an awful lot of"effort" to "persuade" them that they need to "think again" - once had torefuse to publish a Chairman's version of a new standard because it did notmeet the basic BSI guidelines for how a product st
Re: [PSES] Thermal equilibrium - 10% rule
Doug et al, Well, you don’t have to have 100K lbs of equipment to get porpoiseing. I remember a 500W product project where the test house engineer came in to witness the testing and was disturbed by the porpoiseing swings when the product temp seemed stable. He finally understood after the 2nd explanation that the swings were caused by the room AC cycling and he could see the result. Test successfully completed. :>) br, Pete Peter E Perkins, PE Principal Product Safety & Regulatory Affairs Consultant PO Box 23427 Tigard, ORe 97281-3427 503/452-1201 <mailto:p.perk...@ieee.org> p.perk...@ieee.org From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 3:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell mailto:oconne...@tamuracorp.com> > wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto:doug...@gmail.com <mailto:doug...@gmail.com> ] Sent: Friday, January 06, 2017 11:11 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com <mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 11:30 AM, john Allen mailto:john_e_al...@blueyonder.co.uk> > wrote: Minor comment: Having been the Secretary to a number of BSI committees many years ago, I sympathise with Ralph's last para as the Chairmen can be even worse than the other Members - they "know what they know" and it can take an awful lot of "effort" to "persuade" them that they need to "think again" - once had to refuse to publish a Chairman's version of a new standard because it did not meet the basic BSI guidelines for how a product standard should be written, until I had rewritten a large part of it to make it at least reasonably "testable" for the EMC-related requirements - a little (sometimes a lot!) of subject knowledge is required to set appropriate test and assessment requirements! Unfortunately, nowadays, a large number of Committee Secretariats appear to employ non-SME staff to run their committees and so they can be lead by the nose by the Chairmen. FWIW, in the context of this thread, I used to use "commonsense" in deciding when the temperatures appeared to have stabilized - -taking into account where the probes were located - particularly wh
Re: [PSES] Thermal equilibrium - 10% rule
My experience is that agencies frown on any form of averaging. So my approach is to monitor it or all maxima and apply the +/- 2 degree rule. If it so happens that these cycles was wide spread (greater than 15 minutes, I simply take the last three maxima and check for +/- 2 degrees. If I have a complex cycling problem, I choose the slowest running cycle and apply the +/- 2 degree window to that. Any other effects should be captured in that with no problem. doug Original Message From: oconne...@tamuracorp.com Sent: January 6, 2017 4:56 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Reply-to: oconne...@tamuracorp.com Subject: Re: [PSES] Thermal equilibrium - 10% rule Correct, so for multiple duty cycles or varying loading intervals, possible solution would be to use windowed average (queue size based on 1/f and dT/dt and previous sample interval) with a 'delayed' least-squares fit thru the maxima to see if flat. [hear the ghosts of my math instructors yelling at me to use polynomial regression]. Brian From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 3:32 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Correct, so for multiple duty cycles or varying loading intervals, possible solution would be to use windowed average (queue size based on 1/f and dT/dt and previous sample interval) with a 'delayed' least-squares fit thru the maxima to see if flat. [hear the ghosts of my math instructors yelling at me to use polynomial regression]. Brian From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 3:32 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Standards can't necessarily cover every possibility. For cyclical effects, I would say that conditions are stable if two successive temperature maxima are equal (within a reasonable tolerance). This doesn't necessarily work if more than one cycle frequency is involved, in which case you have to look for the repetition of the whole sequence, which might take a long time. For example, a 10 minute cycle and a 12 minute cycle give a sequence that repeats every 60 minutes. If you are lucky. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 11:09 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell mailto:oconne...@tamuracorp.com> > wrote: Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto:doug...@gmail.com <mailto:doug...@gmail.com> ] Sent: Friday, January 06, 2017 11:11 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com <mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 11:30 AM, john Allen mailto:john_e_al...@blueyonder.co.uk> > wrote: Minor comment: Having been the Secretary to a number of BSI committees many years ago, I sympathise with Ralph's last para as the Chairmen can be even worse than the other Members - they "know what they know" and it can take an awful lot of "effort" to "persuade" them that they need to "think again" - once had to refuse to publish a Chairman's version of a new standard because it did not meet the basic BSI guidelines for how a product standard should be written, until I had rewritten a large part of it to make it at least reasonably "testable" for the EMC-related requirements - a little (sometimes a lot!) of subject knowledge is required to set appropriate test and assessment requirements! Unfortunately, nowadays, a large number of Committee Secretariats appear to employ non-SME staff to run their committees and so they can be lead by the nose by the Chairmen. FWIW, in the context of this thread, I used to use "commonsense" in deciding when the temperatures appeared to have stabilized - -taking into account where the probes were located - particularly
Re: [PSES] Thermal equilibrium - 10% rule
Thanks! I'll take a look. I generally log with LabView or direct into an Excel spreadsheet, maybe I can get an Excel VB Script to post expected times. One of the concerns I am dealing with now is how to determine stability when there are cyclical operations going on. I am using the prescribed stability criteria and using this on the minima/maxima of the temperature variations as it moves up and down. Funny, as I sit starting at thermal data moving in this way, I think of it "porpoise-ing" up and down. All the best, Doug -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 2:10 PM, Brian O'Connell wrote: > Rough pseudo code for my transformer algorithm for logging and monitoring > temperatures for normal operating conditions: > > time constant = (material ksp * mass) / (24*60) > sample interval = time constant / (mass * material kx) > breakpoint flags = false > > if sample interval < min interval >sample interval = min interval > > interrupts: >temperatures to circular buffer >log samples and windowed averages to network storage > > loop: >for each channel > verify exponential and set breakpoint flag for each channel > update thermal lag time > adjust sample interval if time constant > thermal lag/2 >find least dT/dt channel >find largest thermal lag time per ambient time per matching indices of > windowed means >if all breakpoint flags > indicate done > > Brian > > > From: Doug Powell [mailto:doug...@gmail.com] > Sent: Friday, January 06, 2017 11:11 AM > To: EMC-PSTC@LISTSERV.IEEE.ORG > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > John, > > I agree with the common sense approach and use it frequently. It's always > interesting to me how I can look at a screen plot of 60 thermocouples and > in a second or two decide, "yes this is stable". I can even estimate how > much time it will take to become stable as a test nears the end (usually > about the time of a lunch break). More than once I have attempted to write > an algorithm to make the same projection and have failed every time. The > non-linearities and multiple heat sources & sinks makes this nearly > impossible. > > On a side-bar, Voltaire is quoted as saying "Common sense is not so > common", which mean he is man with similar sensibilities as myself. > > Those who know me personally have often heard me say what I consider to be > a corollary, "Common sense is usually neither.. common or sensible". > > > -Doug > > > Douglas E Powell > Laporte, Colorado USA > doug...@gmail.com > http://www.linkedin.com/in/dougp01 > > > > On Fri, Jan 6, 2017 at 11:30 AM, john Allen > wrote: > Minor comment: > Having been the Secretary to a number of BSI committees many years ago, I > sympathise with Ralph's last para as the Chairmen can be even worse than > the > other Members - they "know what they know" and it can take an awful lot of > "effort" to "persuade" them that they need to "think again" - once had to > refuse to publish a Chairman's version of a new standard because it did not > meet the basic BSI guidelines for how a product standard should be written, > until I had rewritten a large part of it to make it at least reasonably > "testable" for the EMC-related requirements - a little (sometimes a lot!) > of > subject knowledge is required to set appropriate test and assessment > requirements! > > Unfortunately, nowadays, a large number of Committee Secretariats appear to > employ non-SME staff to run their committees and so they can be lead by the > nose by the Chairmen. > > FWIW, in the context of this thread, I used to use "commonsense" in > deciding > when the temperatures appeared to have stabilized - -taking into account > where the probes were located - particularly when the observed > temperatures > were substantially below the relevant Insulation Class limits. > > John E Allen > W. London, UK > > -Original Message- > From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] > Sent: 06 January 2017 17:20 > To: EMC-PSTC@LISTSERV.IEEE.ORG > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > For what it's worth, we measure and log temperatures at 1 minute intervals > (sometimes quicker, depending on what is being tested) and graph each data > set in Excel and look at the curves. When they go flat (even with some > ripple), we call it stable. Good enough for all the agencies we work with. > And I like the term steady-state much better, but what it
Re: [PSES] Thermal equilibrium - 10% rule
Rough pseudo code for my transformer algorithm for logging and monitoring temperatures for normal operating conditions: time constant = (material ksp * mass) / (24*60) sample interval = time constant / (mass * material kx) breakpoint flags = false if sample interval < min interval sample interval = min interval interrupts: temperatures to circular buffer log samples and windowed averages to network storage loop: for each channel verify exponential and set breakpoint flag for each channel update thermal lag time adjust sample interval if time constant > thermal lag/2 find least dT/dt channel find largest thermal lag time per ambient time per matching indices of windowed means if all breakpoint flags indicate done Brian From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 11:11 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 11:30 AM, john Allen wrote: Minor comment: Having been the Secretary to a number of BSI committees many years ago, I sympathise with Ralph's last para as the Chairmen can be even worse than the other Members - they "know what they know" and it can take an awful lot of "effort" to "persuade" them that they need to "think again" - once had to refuse to publish a Chairman's version of a new standard because it did not meet the basic BSI guidelines for how a product standard should be written, until I had rewritten a large part of it to make it at least reasonably "testable" for the EMC-related requirements - a little (sometimes a lot!) of subject knowledge is required to set appropriate test and assessment requirements! Unfortunately, nowadays, a large number of Committee Secretariats appear to employ non-SME staff to run their committees and so they can be lead by the nose by the Chairmen. FWIW, in the context of this thread, I used to use "commonsense" in deciding when the temperatures appeared to have stabilized - -taking into account where the probes were located - particularly when the observed temperatures were substantially below the relevant Insulation Class limits. John E Allen W. London, UK -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] Sent: 06 January 2017 17:20 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/li
Re: [PSES] Thermal equilibrium - 10% rule
Thermal time constants deal with thermistors and how it changes from one temperature to another or from one ‘ambient state’ to another. One thermal time constant is approximately 63.2% of the total difference in temperature form the initial temperature to a new temperature. They do not however define the total system thermal reactions. Thus, the need for any system to have a defined ‘wait time’ to allow temperature stabilization for other than thermistors. Thanks Dennis Ward This communication and its attachements contain information from PCTEST Engineering Laboratory, Inc., and is intended for the exclusive use of the recipient(s) named above. It may contain information that is confidential and/or legally privileged. Any unauthorized use that may compromise that confidentiality via distribution or disclosure is prohibited. Please notify the sender immediately if you receive this communication in error, and delete it from your computer system. Usage of PCTEST email addresses for non-business related activities is strictly prohibited. No warranty is made that the e-mail or attachments(s) are free from computer virus or other defect. Thank you. From: Gary Tornquist [mailto:08f1d78b7a2b-dmarc-requ...@ieee.org] Sent: Friday, January 6, 2017 11:12 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule As others have said a fixed number of minutes to wait, doesn’t make sense. How long it takes to reach thermal equilibrium to a given number of degrees depends on the thermal time constant(s) of the system under test and the size of the thermal step imposed on the system. I had some success estimating thermal time constants by curve fitting with Excel. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 10:53 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule Ralph, I have used 1-minute as well . In fact for Solar PV systems the Sandia Efficiency test protocol recommends 5 second and 30 second intervals with a fairly complex averaging algorithm to boot. <http://www.gosolarcalifornia.ca.gov/equipment/documents/Sandia_Guideline_2005.pdf> http://www.gosolarcalifornia.ca.gov/equipment/documents/Sandia_Guideline_2005.pdf Doug On Fri, Jan 6, 2017 at 10:19 AM, Ralph McDiarmid mailto:ralph.mcdiar...@schneider-electric.com> > wrote: For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com <mailto:ken.ja...@emccompliance.com> ] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes mailto:jrbar...@iglou.com> > > Reply-To: mailto:jrbar...@iglou.com> > > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor <mailto:ken.ja...@emccompliance.com> > > Cc: mailto:EMC-PSTC@LISTSERV.IEEE.ORG> > > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40
Re: [PSES] Thermal equilibrium - 10% rule
I’m not sure I would consider IEC 62368-1 as having entirely practical criteria. In particular, I dislike the “10%” clause. Is it 10% of the temperature measured in Kelvin or Centigrade? All of the specific references in that part of B.1.6 are to Kelvin. “With reference to those tests that are to be continued until steady state temperatures are attained, steady state is considered to exist if the temperature rise does not exceed 3 K in 30 min. If the measured temperature is at least 10 % less than the specified temperature limit, steady state is considered to exist if the temperature rise does not exceed 1 K in 5 min.” If I have a circuit board rated 130 C, that is 403 K. There is a big difference between 10%. I presume it is 10% of the temperature in Centigrade. Just for fun, let’s presume I have an IT product that has a section that must be kept at temperatures below freezing. Off hand, I don’t have a specific cryogenic system that becomes unsafe if it gets to warm, but this is just a though experiment. If there is a part that must remain below freezing for safety, “10 % below” that negative number now means a measurement warmer than the safety limit. Admittedly, this is a somewhat outlandish though experiment, but it is a case where “10 % below” in Centigrade becomes even more problematic Ted Eckert Microsoft Corporation The opinions expressed are my own and do not necessarily reflect those of my employer. From: Jim Hulbert [mailto:jim.hulb...@pb.com] Sent: Friday, January 6, 2017 7:30 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule IEC 62368-1 seems to have a fairly practical criteria for thermal “steady state” (not “equilibrium”) in B.1.6. Steady state is considered to exist if the temperature rise does not exceed 3 K in 30 min. If the temperature is at least 10% below the specified limit, then steady state is considered to exist if the temperature rise does not exceed 1 K in 5 min. It is not necessary to control ambient temperature to any specific value, but the value should be monitored and recorded. Jim Hulbert From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 9:32 AM To: EMC-PSTC@LISTSERV.IEEE.ORG<mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug On Fri, Jan 6, 2017 at 12:31 AM, John Woodgate mailto:jmw1...@btinternet.com>> wrote: Only that , even with cooling and conductor resistivity taken into account, the approach to equilibrium is still very nearly exponential (the resistivity change is non-linear) , so in theory takes infinite time. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk<http://www.jmwa.demon.co.uk/> J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Ken Javor [mailto:ken.ja...@emccompliance.com<mailto:ken.ja...@emccompliance.com>] Sent: Friday, January 6, 2017 3:18 AM To: EMC-PSTC@LISTSERV.IEEE.ORG<mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule Understand I’m not arguing with Rich Nute on a topic with which I have little familiarity. I’m just looking at the physics of heat transfer. If I am running current through a conductor that causes its temperature to rise, then as temperature rises, according to Newton’s Law of Cooling, more heat is transferred from the hot body to the cooler surroundings, in direct proportion. So at some point the temperature will equilibrate because the heat energy radiated/convected/conducted to the surroundings will equal the electrical energy dissipated in the conductor. What am I missing? Ken Javor Phone: (256) 650-5261 From: Richard Nute mailto:ri...@ieee.org>> Reply-To: mailto:ri...@ieee.org>> Date: Thu, 5 Jan 2017 17:03:36 -0800 To: mailto:
Re: [PSES] Thermal equilibrium - 10% rule
Hi Doug: My comments are imbedded in your message. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 6:32 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. Agree! I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past Agree! and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Put a tent over it, and put a heater in the tent to simulate the hottest day. Take your temperature measurements just before the outdoor temperature causes the tent and equipment to decrease in temperature. Thanks all, Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
1 minute measurement interval was what I suggested. That has worked well for power converters from 10W to 1MW. Excel is a good way of determining “steady state”, and I’ve used that technique for some time. Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric From: Gary Tornquist [mailto:08f1d78b7a2b-dmarc-requ...@ieee.org] Sent: Friday, January 06, 2017 11:12 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule As others have said a fixed number of minutes to wait, doesn’t make sense. How long it takes to reach thermal equilibrium to a given number of degrees depends on the thermal time constant(s) of the system under test and the size of the thermal step imposed on the system. I had some success estimating thermal time constants by curve fitting with Excel. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 10:53 AM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Ralph, I have used 1-minute as well . In fact for Solar PV systems the Sandia Efficiency test protocol recommends 5 second and 30 second intervals with a fairly complex averaging algorithm to boot. http://www.gosolarcalifornia.ca.gov/equipment/documents/Sandia_Guideline_2005.pdf Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
As others have said a fixed number of minutes to wait, doesn’t make sense. How long it takes to reach thermal equilibrium to a given number of degrees depends on the thermal time constant(s) of the system under test and the size of the thermal step imposed on the system. I had some success estimating thermal time constants by curve fitting with Excel. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 10:53 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Ralph, I have used 1-minute as well . In fact for Solar PV systems the Sandia Efficiency test protocol recommends 5 second and 30 second intervals with a fairly complex averaging algorithm to boot. http://www.gosolarcalifornia.ca.gov/equipment/documents/Sandia_Guideline_2005.pdf Doug On Fri, Jan 6, 2017 at 10:19 AM, Ralph McDiarmid mailto:ralph.mcdiar...@schneider-electric.com>> wrote: For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com<mailto:ken.ja...@emccompliance.com>] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG<mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes mailto:jrbar...@iglou.com>> > Reply-To: mailto:jrbar...@iglou.com>> > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > mailto:ken.ja...@emccompliance.com>> > Cc: mailto:EMC-PSTC@LISTSERV.IEEE.ORG>> > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to mailto:emc-p...@ieee.org>> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas mailto:sdoug...@ieee.org>> Mike Cantwell mailto:mcantw
Re: [PSES] Thermal equilibrium - 10% rule
Please do tell us about the nonsense statement about diodes. I'd like a Friday morning smile. Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 6:51 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards committees will re-use text if no-one tells them not to. It may take many tellings, too. It took me several years to get a total nonsense statement about diodes deleted from IEC 60384 and equally long times to get a typo in the electrochemical table in IEC 60065,60950-1 and 62384-1 fixed. But the longest time was taken in getting the specification of 'petroleum spirit' changed from a misinterpretation to a realistic statement. With best wishes DESIGN IT IN! OOO – Own Opinions Only http://www.jmwa.demon.co.uk/ J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 2:32 PM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
I think: Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". is a bit too cynical or pessimistic. It isn't common enough, for sure, but it is often sensible. For example, does someone born and educated in Britain, with a Doctor's degree, need to apply for citizenship and take a language test because he can't prove that 50 years ago his French mother was in Britain legally? Common sense says 'No', but HM Gov is said to differ. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 7:11 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "Common sense is not so common", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "Common sense is usually neither.. common or sensible". -Doug Douglas E Powell Laporte, Colorado USA <mailto:doug...@gmail.com> doug...@gmail.com <http://www.linkedin.com/in/dougp01> http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 11:30 AM, john Allen mailto:john_e_al...@blueyonder.co.uk> > wrote: Minor comment: Having been the Secretary to a number of BSI committees many years ago, I sympathise with Ralph's last para as the Chairmen can be even worse than the other Members - they "know what they know" and it can take an awful lot of "effort" to "persuade" them that they need to "think again" - once had to refuse to publish a Chairman's version of a new standard because it did not meet the basic BSI guidelines for how a product standard should be written, until I had rewritten a large part of it to make it at least reasonably "testable" for the EMC-related requirements - a little (sometimes a lot!) of subject knowledge is required to set appropriate test and assessment requirements! Unfortunately, nowadays, a large number of Committee Secretariats appear to employ non-SME staff to run their committees and so they can be lead by the nose by the Chairmen. FWIW, in the context of this thread, I used to use "commonsense" in deciding when the temperatures appeared to have stabilized - -taking into account where the probes were located - particularly when the observed temperatures were substantially below the relevant Insulation Class limits. John E Allen W. London, UK -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com <mailto:ralph.mcdiar...@schneider-electric.com> ] Sent: 06 January 2017 17:20 To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com <mailto:ken.ja...@emccompliance.com> ] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results,
Re: [PSES] Thermal equilibrium - 10% rule
John, I agree with the common sense approach and use it frequently. It's always interesting to me how I can look at a screen plot of 60 thermocouples and in a second or two decide, "yes this is stable". I can even estimate how much time it will take to become stable as a test nears the end (usually about the time of a lunch break). More than once I have attempted to write an algorithm to make the same projection and have failed every time. The non-linearities and multiple heat sources & sinks makes this nearly impossible. On a side-bar, Voltaire is quoted as saying "*Common sense is not so common*", which mean he is man with similar sensibilities as myself. Those who know me personally have often heard me say what I consider to be a corollary, "*Common sense is usually neither.. common or sensible*". -Doug Douglas E Powell Laporte, Colorado USA doug...@gmail.com http://www.linkedin.com/in/dougp01 On Fri, Jan 6, 2017 at 11:30 AM, john Allen wrote: > Minor comment: > Having been the Secretary to a number of BSI committees many years ago, I > sympathise with Ralph's last para as the Chairmen can be even worse than > the > other Members - they "know what they know" and it can take an awful lot of > "effort" to "persuade" them that they need to "think again" - once had to > refuse to publish a Chairman's version of a new standard because it did not > meet the basic BSI guidelines for how a product standard should be written, > until I had rewritten a large part of it to make it at least reasonably > "testable" for the EMC-related requirements - a little (sometimes a lot!) > of > subject knowledge is required to set appropriate test and assessment > requirements! > > Unfortunately, nowadays, a large number of Committee Secretariats appear to > employ non-SME staff to run their committees and so they can be lead by the > nose by the Chairmen. > > FWIW, in the context of this thread, I used to use "commonsense" in > deciding > when the temperatures appeared to have stabilized - -taking into account > where the probes were located - particularly when the observed > temperatures > were substantially below the relevant Insulation Class limits. > > John E Allen > W. London, UK > > -Original Message- > From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] > Sent: 06 January 2017 17:20 > To: EMC-PSTC@LISTSERV.IEEE.ORG > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > For what it's worth, we measure and log temperatures at 1 minute intervals > (sometimes quicker, depending on what is being tested) and graph each data > set in Excel and look at the curves. When they go flat (even with some > ripple), we call it stable. Good enough for all the agencies we work with. > And I like the term steady-state much better, but what it is called isn't > terribly important as I see it. The temperature can be stable (no further > increase observable) over a period, even if it's oscillating slightly > around > a mean value. The mean value could be the average of the last 20 or so > reading for instance. It requires some judgement. > > I empathize with John's experience about it taking "many tellings" on a > committee to get something right. I have stopped just short of the Makita > Khrushchev United Nations technique at times! > > Ralph McDiarmid > Product Compliance > Engineering > Solar Business > Schneider Electric > > > > > > -Original Message- > From: Ken Javor [mailto:ken.ja...@emccompliance.com] > Sent: Friday, January 06, 2017 8:21 AM > To: EMC-PSTC@LISTSERV.IEEE.ORG > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > I understand that if the system is poorly designed there can be > catastrophic > results, but the test is written assuming the equipment temperature does > equilibrate, right? > > Ken Javor > Phone: (256) 650-5261 > > > > From: John Barnes > > Reply-To: > > Date: Fri, 06 Jan 2017 00:06:24 -0500 > > To: Ken Javor > > Cc: > > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > > > Ken, > > Most metals have a positive Temperature Coefficient of Resistance (TCR). > > As they get hotter, their resistivity-- and the resistance of a given > > conductor-- increases. If you have a current source (or a voltage > > source with a large series impedance) driving current through a small > > cross-section conductor with poor heat sinking for a long time, the > > conductor will go into thermal runaway-- and eventually melt in two-- > > when the current through it exceeds a certain critical value, > > depen
Re: [PSES] Thermal equilibrium - 10% rule
Oops, sorry Mr. Khrushchev, wherever you are! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 10:40 AM To: Ralph McDiarmid ; EMC-PSTC@LISTSERV.IEEE.ORG Subject: RE: [PSES] Thermal equilibrium - 10% rule That's Nikita, not Makita. And if you look at the clip carefully, he's wearing two shoes, so the one he's using as a gavel must have come from a sputnik. With best wishes DESIGN IT IN! OOO - Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] Sent: Friday, January 6, 2017 5:20 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes > Reply-To: > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > Cc: > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: __ This email has been scanned by the Symantec Email Security.cloud service. __ - ---
Re: [PSES] Thermal equilibrium - 10% rule
Ralph, I have used 1-minute as well . In fact for Solar PV systems the Sandia Efficiency test protocol recommends 5 second and 30 second intervals with a fairly complex averaging algorithm to boot. http://www.gosolarcalifornia.ca.gov/equipment/documents/Sandia_Guideline_2005.pdf Doug On Fri, Jan 6, 2017 at 10:19 AM, Ralph McDiarmid < ralph.mcdiar...@schneider-electric.com> wrote: > For what it's worth, we measure and log temperatures at 1 minute intervals > (sometimes quicker, depending on what is being tested) and graph each data > set in Excel and look at the curves. When they go flat (even with some > ripple), we call it stable. Good enough for all the agencies we work > with. And I like the term steady-state much better, but what it is called > isn't terribly important as I see it. The temperature can be stable (no > further increase observable) over a period, even if it's oscillating > slightly around a mean value. The mean value could be the average of the > last 20 or so reading for instance. It requires some judgement. > > I empathize with John's experience about it taking "many tellings" on a > committee to get something right. I have stopped just short of the Makita > Khrushchev United Nations technique at times! > > Ralph McDiarmid > Product Compliance > Engineering > Solar Business > Schneider Electric > > > > > > -Original Message- > From: Ken Javor [mailto:ken.ja...@emccompliance.com] > Sent: Friday, January 06, 2017 8:21 AM > To: EMC-PSTC@LISTSERV.IEEE.ORG > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > I understand that if the system is poorly designed there can be > catastrophic results, but the test is written assuming the equipment > temperature does equilibrate, right? > > Ken Javor > Phone: (256) 650-5261 > > > > From: John Barnes > > Reply-To: > > Date: Fri, 06 Jan 2017 00:06:24 -0500 > > To: Ken Javor > > Cc: > > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > > > Ken, > > Most metals have a positive Temperature Coefficient of Resistance (TCR). > > As they get hotter, their resistivity-- and the resistance of a given > > conductor-- increases. If you have a current source (or a voltage > > source with a large series impedance) driving current through a small > > cross-section conductor with poor heat sinking for a long time, the > > conductor will go into thermal runaway-- and eventually melt in two-- > > when the current through it exceeds a certain critical value, > > depending on the ambient temperature. > > > > A. J. Rainal wrote two papers about this effect on printed circuit > > boards (PCB's) about 40 years ago: > > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed > >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. > >1375-1388, September 1981. > > * Rainal, A. J., "Temperature Rise at a Constriction in a > >Current-Carrying Printed Conductor," The Bell System Technical > >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > > > Non-resettable fuses use this principle, which is specified by the > > I^2t curves in their datasheets. > > > > You may also want to read up on Preece's Law (Fuse Equations) and > > Onderdonk's Fuse Equation. > > > > > > > > John Barnes KS4GL (retired) > > Lexington, Kentucky > > http://www.dbicorporation.com/ > > > > - > > This message is from the IEEE Product Safety Engineering Society emc-pstc > discussion list. To post a message to the list, send your e-mail to < > emc-p...@ieee.org> > > All emc-pstc postings are archived and searchable on the web at: > http://www.ieee-pses.org/emc-pstc.html > > Attachments are not permitted but the IEEE PSES Online Communities site at > http://product-compliance.oc.ieee.org/ can be used for graphics (in > well-used formats), large files, etc. > > Website: http://www.ieee-pses.org/ > Instructions: http://www.ieee-pses.org/list.html (including how to > unsubscribe) List rules: http://www.ieee-pses.org/listrules.html > > For help, send mail to the list administrators: > Scott Douglas > Mike Cantwell > > For policy questions, send mail to: > Jim Bacher: > David Heald: > > __ > This email has been scanned by the Symantec Email Security.cloud service. > __ > > - >
Re: [PSES] Thermal equilibrium - 10% rule
That's Nikita, not Makita. And if you look at the clip carefully, he's wearing two shoes, so the one he's using as a gavel must have come from a sputnik. With best wishes DESIGN IT IN! OOO Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] Sent: Friday, January 6, 2017 5:20 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes > Reply-To: > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > Cc: > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: __ This email has been scanned by the Symantec Email Security.cloud service. __ - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.
Re: [PSES] Thermal equilibrium - 10% rule
Minor comment: Having been the Secretary to a number of BSI committees many years ago, I sympathise with Ralph's last para as the Chairmen can be even worse than the other Members - they "know what they know" and it can take an awful lot of "effort" to "persuade" them that they need to "think again" - once had to refuse to publish a Chairman's version of a new standard because it did not meet the basic BSI guidelines for how a product standard should be written, until I had rewritten a large part of it to make it at least reasonably "testable" for the EMC-related requirements - a little (sometimes a lot!) of subject knowledge is required to set appropriate test and assessment requirements! Unfortunately, nowadays, a large number of Committee Secretariats appear to employ non-SME staff to run their committees and so they can be lead by the nose by the Chairmen. FWIW, in the context of this thread, I used to use "commonsense" in deciding when the temperatures appeared to have stabilized - -taking into account where the probes were located - particularly when the observed temperatures were substantially below the relevant Insulation Class limits. John E Allen W. London, UK -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] Sent: 06 January 2017 17:20 To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes > Reply-To: > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > Cc: > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-
Re: [PSES] Thermal equilibrium - 10% rule
I'm sorry, I can't remember the details and it was around 30 years ago. If anyone has a copy of the original IEC 386 (not 384), the offending words should be obvious. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. -Original Message- From: Ralph McDiarmid [mailto:ralph.mcdiar...@schneider-electric.com] Sent: Friday, January 6, 2017 5:03 PM To: John Woodgate ; EMC-PSTC@LISTSERV.IEEE.ORG Subject: RE: [PSES] Thermal equilibrium - 10% rule Please do tell us about the nonsense statement about diodes. I'd like a Friday morning smile. Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric From: John Woodgate [mailto:jmw1...@btinternet.com] Sent: Friday, January 06, 2017 6:51 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Standards committees will re-use text if no-one tells them not to. It may take many tellings, too. It took me several years to get a total nonsense statement about diodes deleted from IEC 60384 and equally long times to get a typo in the electrochemical table in IEC 60065,60950-1 and 62384-1 fixed. But the longest time was taken in getting the specification of 'petroleum spirit' changed from a misinterpretation to a realistic statement. With best wishes DESIGN IT IN! OOO – Own Opinions Only http://www.jmwa.demon.co.uk/ J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 2:32 PM To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
For what it's worth, we measure and log temperatures at 1 minute intervals (sometimes quicker, depending on what is being tested) and graph each data set in Excel and look at the curves. When they go flat (even with some ripple), we call it stable. Good enough for all the agencies we work with. And I like the term steady-state much better, but what it is called isn't terribly important as I see it. The temperature can be stable (no further increase observable) over a period, even if it's oscillating slightly around a mean value. The mean value could be the average of the last 20 or so reading for instance. It requires some judgement. I empathize with John's experience about it taking "many tellings" on a committee to get something right. I have stopped just short of the Makita Khrushchev United Nations technique at times! Ralph McDiarmid Product Compliance Engineering Solar Business Schneider Electric -Original Message- From: Ken Javor [mailto:ken.ja...@emccompliance.com] Sent: Friday, January 06, 2017 8:21 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes > Reply-To: > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > Cc: > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, > depending on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald: __ This email has been scanned by the Symantec Email Security.cloud service. __ - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
IEC 62368-1 seems to have a fairly practical criteria for thermal “steady state” (not “equilibrium”) in B.1.6. Steady state is considered to exist if the temperature rise does not exceed 3 K in 30 min. If the temperature is at least 10% below the specified limit, then steady state is considered to exist if the temperature rise does not exceed 1 K in 5 min. It is not necessary to control ambient temperature to any specific value, but the value should be monitored and recorded. Jim Hulbert From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 06, 2017 9:32 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug On Fri, Jan 6, 2017 at 12:31 AM, John Woodgate mailto:jmw1...@btinternet.com>> wrote: Only that , even with cooling and conductor resistivity taken into account, the approach to equilibrium is still very nearly exponential (the resistivity change is non-linear) , so in theory takes infinite time. With best wishes DESIGN IT IN! OOO – Own Opinions Only www.jmwa.demon.co.uk<http://www.jmwa.demon.co.uk/> J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Ken Javor [mailto:ken.ja...@emccompliance.com<mailto:ken.ja...@emccompliance.com>] Sent: Friday, January 6, 2017 3:18 AM To: EMC-PSTC@LISTSERV.IEEE.ORG<mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule Understand I’m not arguing with Rich Nute on a topic with which I have little familiarity. I’m just looking at the physics of heat transfer. If I am running current through a conductor that causes its temperature to rise, then as temperature rises, according to Newton’s Law of Cooling, more heat is transferred from the hot body to the cooler surroundings, in direct proportion. So at some point the temperature will equilibrate because the heat energy radiated/convected/conducted to the surroundings will equal the electrical energy dissipated in the conductor. What am I missing? Ken Javor Phone: (256) 650-5261 From: Richard Nute mailto:ri...@ieee.org>> Reply-To: mailto:ri...@ieee.org>> Date: Thu, 5 Jan 2017 17:03:36 -0800 To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG>> Subject: Re: [PSES] Thermal equilibrium - 10% rule Thermal equilibrium is impossible. Thermal change is an exponential, never achieving the end-point. The test is to determine that some part does not exceed its temperature rating. Plotting the (exponential) curve shows whether the temperature in question is or is not going to exceed its rating. (Curve-fitting would be nice, but impossible.) As soon as this determination is made, the test can be concluded. Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal change is an exponential. And, it is tight. For a small (low thermal mass) part, air currents will change the temperature beyond this limit. The statement does not reflect knowledge of primitive thermodynamics. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG<mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be o
Re: [PSES] Thermal equilibrium - 10% rule
I understand that if the system is poorly designed there can be catastrophic results, but the test is written assuming the equipment temperature does equilibrate, right? Ken Javor Phone: (256) 650-5261 > From: John Barnes > Reply-To: > Date: Fri, 06 Jan 2017 00:06:24 -0500 > To: Ken Javor > Cc: > Subject: Re: [PSES] Thermal equilibrium - 10% rule > > Ken, > Most metals have a positive Temperature Coefficient of Resistance (TCR). > As they get hotter, their resistivity-- and the resistance of a given > conductor-- increases. If you have a current source (or a voltage > source with a large series impedance) driving current through a small > cross-section conductor with poor heat sinking for a long time, the > conductor will go into thermal runaway-- and eventually melt in two-- > when the current through it exceeds a certain critical value, depending > on the ambient temperature. > > A. J. Rainal wrote two papers about this effect on printed circuit > boards (PCB's) about 40 years ago: > * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed >Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. >1375-1388, September 1981. > * Rainal, A. J., "Temperature Rise at a Constriction in a >Current-Carrying Printed Conductor," The Bell System Technical >Journal, vol. 55 no. 2, pp. 233-269, February 1976. > > Non-resettable fuses use this principle, which is specified by the > I^2t curves in their datasheets. > > You may also want to read up on Preece's Law (Fuse Equations) and > Onderdonk's Fuse Equation. > > > > John Barnes KS4GL (retired) > Lexington, Kentucky > http://www.dbicorporation.com/ > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Standards committees will re-use text if no-one tells them not to. It may take many tellings, too. It took me several years to get a total nonsense statement about diodes deleted from IEC 60384 and equally long times to get a typo in the electrochemical table in IEC 60065,60950-1 and 62384-1 fixed. But the longest time was taken in getting the specification of 'petroleum spirit' changed from a misinterpretation to a realistic statement. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Friday, January 6, 2017 2:32 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The 10 percent of the previously elapsed time business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug On Fri, Jan 6, 2017 at 12:31 AM, John Woodgate mailto:jmw1...@btinternet.com> > wrote: Only that , even with cooling and conductor resistivity taken into account, the approach to equilibrium is still very nearly exponential (the resistivity change is non-linear) , so in theory takes infinite time. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Ken Javor [mailto:ken.ja...@emccompliance.com <mailto:ken.ja...@emccompliance.com> ] Sent: Friday, January 6, 2017 3:18 AM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: Re: [PSES] Thermal equilibrium - 10% rule Understand I’m not arguing with Rich Nute on a topic with which I have little familiarity. I’m just looking at the physics of heat transfer. If I am running current through a conductor that causes its temperature to rise, then as temperature rises, according to Newton’s Law of Cooling, more heat is transferred from the hot body to the cooler surroundings, in direct proportion. So at some point the temperature will equilibrate because the heat energy radiated/convected/conducted to the surroundings will equal the electrical energy dissipated in the conductor. What am I missing? Ken Javor Phone: (256) 650-5261 _ From: Richard Nute mailto:ri...@ieee.org> > Reply-To: mailto:ri...@ieee.org> > Date: Thu, 5 Jan 2017 17:03:36 -0800 To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG> > Subject: Re: [PSES] Thermal equilibrium - 10% rule Thermal equilibrium is impossible. Thermal change is an exponential, never achieving the end-point. The test is to determine that some part does not exceed its temperature rating. Plotting the (exponential) curve shows whether the temperature in question is or is not going to exceed its rating. (Curve-fitting would be nice, but impossible.) As soon as this determination is made, the test can be concluded. Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal change is an exponential. And, it is tight. For a small (low thermal mass) part, air currents will change the temperature beyond this limit. The statement does not reflect knowledge of primitive thermodynamics. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of t
Re: [PSES] Thermal equilibrium - 10% rule
I agree that thermal equilibrium can never be achieved especially in light of the zeroth law of thermodynamics, but also for more reasons than the exponential nature of the thing. Possibly a better phrase to use is thermal stability. I actually believe that the standards committees will re-use the text of previous standards and averse to changing it in subsequent revisions with the rationale that it worked in the past and possibly there's someone relying on this bit of information. The *10 percent of the previously elapsed time* business simply does not make sense to me and I have never used it in actual testing. Nevertheless, it has shown up in a couple of recently published standards I use for large format storage batteries. In my case, I have an outdoor product that weighs in at more than 100,000 pounds and thermal equilibrium cannot be achieved in 24 hours, meanwhile the daily cyclical temperature compound the problem of measurement. Obviously, I am going to have to do live compensation of ambient conditions during the test and not afterward. Thanks all, Doug On Fri, Jan 6, 2017 at 12:31 AM, John Woodgate wrote: > Only that , even with cooling and conductor resistivity taken into > account, the approach to equilibrium is still very nearly exponential (the > resistivity change is non-linear) , so in theory takes infinite time. > > > > With best wishes DESIGN IT IN! OOO – Own Opinions Only > > www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England > > > > Sylvae in aeternum manent. > > > > *From:* Ken Javor [mailto:ken.ja...@emccompliance.com] > *Sent:* Friday, January 6, 2017 3:18 AM > > *To:* EMC-PSTC@LISTSERV.IEEE.ORG > *Subject:* Re: [PSES] Thermal equilibrium - 10% rule > > > > Understand I’m not arguing with Rich Nute on a topic with which I have > little familiarity. I’m just looking at the physics of heat transfer. If I > am running current through a conductor that causes its temperature to rise, > then as temperature rises, according to Newton’s Law of Cooling, more heat > is transferred from the hot body to the cooler surroundings, in direct > proportion. So at some point the temperature will equilibrate because the > heat energy radiated/convected/conducted to the surroundings will equal the > electrical energy dissipated in the conductor. > > What am I missing? > > Ken Javor > Phone: (256) 650-5261 > > ------ > > *From: *Richard Nute > *Reply-To: * > *Date: *Thu, 5 Jan 2017 17:03:36 -0800 > *To: * > *Subject: *Re: [PSES] Thermal equilibrium - 10% rule > > > *Thermal equilibrium* is impossible. Thermal change is an exponential, > never achieving the end-point. > > The test is to determine that some part does not exceed its temperature > rating. Plotting the (exponential) curve shows whether the temperature in > question is or is not going to exceed its rating. (Curve-fitting would be > nice, but impossible.) As soon as this determination is made, the test can > be concluded. > > Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal > change is an exponential. And, it is tight. For a small (low thermal > mass) part, air currents will change the temperature beyond this limit. > > The statement does not reflect knowledge of primitive thermodynamics. > > Best wishes for the New Year! > Rich > > > > *From:* Doug Powell [mailto:doug...@gmail.com] > *Sent:* Thursday, January 05, 2017 2:47 PM > *To:* EMC-PSTC@LISTSERV.IEEE.ORG > *Subject:* [PSES] Thermal equilibrium - 10% rule > > > All, > > > > The following has always been confusing for me, > > > > > *For those tests that require the DUT to reach thermal equilibrium, > thermal equilibrium is considered to be achieved if after three consecutive > temperature measurements taken at intervals of 10% of the previously > elapsed duration of the test but not less than 15 min, indicate no change > in temperature greater than ±2°C (3.6°F).* > > > > > The portion of text "*taken at intervals of 10% of the previously elapsed > duration of the test"* has been around for some time and it exists in > several safety standards, the "*not less than 15 min*" can be other > values like 10 minutes, 5 minutes, etc. I believe the intent is to > minimize the number of data points being recorded for extremely long > thermal tests. The idea makes sense when I think back to the days of manual > data logging. Nowadays, with the advent of automated data logging, this > point probably becomes moot. Still, the phrase bugs me and I would like to > understand how this might be interpreted. > > > > First the "*previously elapsed duration*" could indicate either
Re: [PSES] Thermal equilibrium - 10% rule
Only that , even with cooling and conductor resistivity taken into account, the approach to equilibrium is still very nearly exponential (the resistivity change is non-linear) , so in theory takes infinite time. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Ken Javor [mailto:ken.ja...@emccompliance.com] Sent: Friday, January 6, 2017 3:18 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Thermal equilibrium - 10% rule Understand I’m not arguing with Rich Nute on a topic with which I have little familiarity. I’m just looking at the physics of heat transfer. If I am running current through a conductor that causes its temperature to rise, then as temperature rises, according to Newton’s Law of Cooling, more heat is transferred from the hot body to the cooler surroundings, in direct proportion. So at some point the temperature will equilibrate because the heat energy radiated/convected/conducted to the surroundings will equal the electrical energy dissipated in the conductor. What am I missing? Ken Javor Phone: (256) 650-5261 _ From: Richard Nute mailto:ri...@ieee.org> > Reply-To: mailto:ri...@ieee.org> > Date: Thu, 5 Jan 2017 17:03:36 -0800 To: mailto:EMC-PSTC@LISTSERV.IEEE.ORG> > Subject: Re: [PSES] Thermal equilibrium - 10% rule Thermal equilibrium is impossible. Thermal change is an exponential, never achieving the end-point. The test is to determine that some part does not exceed its temperature rating. Plotting the (exponential) curve shows whether the temperature in question is or is not going to exceed its rating. (Curve-fitting would be nice, but impossible.) As soon as this determination is made, the test can be concluded. Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal change is an exponential. And, it is tight. For a small (low thermal mass) part, air currents will change the temperature beyond this limit. The statement does not reflect knowledge of primitive thermodynamics. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG <mailto:EMC-PSTC@LISTSERV.IEEE.ORG> Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "previously elapsed duration" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug -- Douglas E Powell doug...@gmail.com <mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to mailto:emc-p...@ieee.org> > All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) <http://www.ieee-pses.org/list.html> <http://www.ieee-pses.org/list.html> List rules: http
Re: [PSES] Thermal equilibrium - 10% rule
My reading is that the 'previously elapsed duration' is the whole time since the heat run was started. So if you started it 100 min ago, the intervals would be 10 min but for the prescribed 15 min. It could be claimed that the wording implies that 'compound interest' should apply, so that the duration is incremented at each interval, leading to intervals of 10 min, 11 min and 12.1 min, but anyone claiming that is a philosopher, not an engineer. With best wishes DESIGN IT IN! OOO – Own Opinions Only <http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England Sylvae in aeternum manent. From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 5, 2017 10:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "previously elapsed duration" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug -- Douglas E Powell doug...@gmail.com <mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to mailto:emc-p...@ieee.org> > All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) <http://www.ieee-pses.org/list.html> List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas mailto:sdoug...@ieee.org> > Mike Cantwell mailto:mcantw...@ieee.org> > For policy questions, send mail to: Jim Bacher mailto:j.bac...@ieee.org> > David Heald mailto:dhe...@gmail.com> > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Ken, Most metals have a positive Temperature Coefficient of Resistance (TCR). As they get hotter, their resistivity-- and the resistance of a given conductor-- increases. If you have a current source (or a voltage source with a large series impedance) driving current through a small cross-section conductor with poor heat sinking for a long time, the conductor will go into thermal runaway-- and eventually melt in two-- when the current through it exceeds a certain critical value, depending on the ambient temperature. A. J. Rainal wrote two papers about this effect on printed circuit boards (PCB's) about 40 years ago: * Rainal, A. J., "Current-Carrying Capacity of Fine-Line Printed Conductors," The Bell System Technical Journal, vol. 60 no. 7, pp. 1375-1388, September 1981. * Rainal, A. J., "Temperature Rise at a Constriction in a Current-Carrying Printed Conductor," The Bell System Technical Journal, vol. 55 no. 2, pp. 233-269, February 1976. Non-resettable fuses use this principle, which is specified by the I^2t curves in their datasheets. You may also want to read up on Preece's Law (Fuse Equations) and Onderdonk's Fuse Equation. John Barnes KS4GL (retired) Lexington, Kentucky http://www.dbicorporation.com/ - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Understand I’m not arguing with Rich Nute on a topic with which I have little familiarity. I’m just looking at the physics of heat transfer. If I am running current through a conductor that causes its temperature to rise, then as temperature rises, according to Newton’s Law of Cooling, more heat is transferred from the hot body to the cooler surroundings, in direct proportion. So at some point the temperature will equilibrate because the heat energy radiated/convected/conducted to the surroundings will equal the electrical energy dissipated in the conductor. What am I missing? Ken Javor Phone: (256) 650-5261 From: Richard Nute Reply-To: Date: Thu, 5 Jan 2017 17:03:36 -0800 To: Subject: Re: [PSES] Thermal equilibrium - 10% rule Thermal equilibrium is impossible. Thermal change is an exponential, never achieving the end-point. The test is to determine that some part does not exceed its temperature rating. Plotting the (exponential) curve shows whether the temperature in question is or is not going to exceed its rating. (Curve-fitting would be nice, but impossible.) As soon as this determination is made, the test can be concluded. Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal change is an exponential. And, it is tight. For a small (low thermal mass) part, air currents will change the temperature beyond this limit. The statement does not reflect knowledge of primitive thermodynamics. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, > > For those tests that require the DUT to reach thermal equilibrium, thermal > equilibrium is considered to be achieved if after three consecutive > temperature measurements taken at intervals of 10% of the previously elapsed > duration of the test but not less than 15 min, indicate no change in > temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "previously elapsed duration" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug -- Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) <http://www.ieee-pses.org/list.html> List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher David Heald - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, sen
Re: [PSES] Thermal equilibrium - 10% rule
Safety standards, for this subject, are dogmatic and ritualistically procedural and do not pertain to basic principles of thermodynamics. Thermal equilibrium requires a temporal measurements of two or three systems, where it has been verified no heat is being transferred between the systems.. As you indicated, this stuff is done by automated data acquisition systems (where my computer typically logs data at 10mSec to 10 Sec intervals). So very obvious when the system has stabilized when compared to the ambient temperature's dT/dt. None of my lab computers have complained about the number of data points being logged. And humans need to be removed from manual data logging. Decently written instrument-control code will find stuff that the weak and puny humans will never be able to observe. Brian From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "previously elapsed duration" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
Re: [PSES] Thermal equilibrium - 10% rule
Thermal equilibrium is impossible. Thermal change is an exponential, never achieving the end-point. The test is to determine that some part does not exceed its temperature rating. Plotting the (exponential) curve shows whether the temperature in question is or is not going to exceed its rating. (Curve-fitting would be nice, but impossible.) As soon as this determination is made, the test can be concluded. Defining thermal equilibrium as ±2°C (3.6°F) is not achievable as thermal change is an exponential. And, it is tight. For a small (low thermal mass) part, air currents will change the temperature beyond this limit. The statement does not reflect knowledge of primitive thermodynamics. Best wishes for the New Year! Rich From: Doug Powell [mailto:doug...@gmail.com] Sent: Thursday, January 05, 2017 2:47 PM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: [PSES] Thermal equilibrium - 10% rule All, The following has always been confusing for me, For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F). The portion of text "taken at intervals of 10% of the previously elapsed duration of the test" has been around for some time and it exists in several safety standards, the "not less than 15 min" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "previously elapsed duration" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug -- Douglas E Powell doug...@gmail.com <mailto:doug...@gmail.com> http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
[PSES] Thermal equilibrium - 10% rule
All, The following has always been confusing for me, *For those tests that require the DUT to reach thermal equilibrium, thermal equilibrium is considered to be achieved if after three consecutive temperature measurements taken at intervals of 10% of the previously elapsed duration of the test but not less than 15 min, indicate no change in temperature greater than ±2°C (3.6°F).* The portion of text "*taken at intervals of 10% of the previously elapsed duration of the test"* has been around for some time and it exists in several safety standards, the "*not less than 15 min*" can be other values like 10 minutes, 5 minutes, etc. I believe the intent is to minimize the number of data points being recorded for extremely long thermal tests. The idea makes sense when I think back to the days of manual data logging. Nowadays, with the advent of automated data logging, this point probably becomes moot. Still, the phrase bugs me and I would like to understand how this might be interpreted. First the "*previously elapsed duration*" could indicate either the duration of the entire test or the duration of the last logging interval. Second, over a long test 10% of the previous interval very quickly shortens the logging intervals to something ridiculously small. For example, in three data points interval of 1440 minutes (24 hours) will be reduced to 144 minutes (2.4 hours) and then 14.4 minutes, at which point the 15 minute minimum takes over. I generally record at a much shorter intervals, even if for several days and check for equilibrium at a modulus of 15 minutes, as required. Thanks, Doug -- Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas Mike Cantwell For policy questions, send mail to: Jim Bacher: David Heald:
