RE: [Tinyos-help] battery voltage micaz
Title: Message FYI - I haven't looked into implementing this, but the CC2420 radio on the MicaZ and Telos does offer battery monitor. From the CC2420 datasheet, page 53: "The on-chip battery monitor enables monitoring the unregulated voltage on the VREG_IN pin. It gives status information on the voltage being above or below a programmable threshold." Basically you enabled the battery monitor through the BATTMON control register, set a 5-bit target voltage in the BATTMON.BATTMON_VOLTAGE register, and then read the BATTMON_OKstatus bit to tell if the input voltage is above or belowyour defined threshold. Seems pretty simple to do. And the CC2420 battery monitor can be turned off when you don't need a reading. It would be nice to have a component that would tell your application when the battery voltage is running low... either by an eventdriven by a periodic timer that samples the CC2420, or through your application explicitly polling the battery monitor component. -David -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of R GartzSent: Friday, May 19, 2006 4:04 PMTo: Michael SchipplingCc: tinyos-help@Millennium.Berkeley.EDUSubject: Re: [Tinyos-help] battery voltage micazYes I know they are very suspicious ;) But I don't have any bit shift anywhere in my program(s)... Those are the actual values I'm gettingMichael, in your robocode you're getting the right values for mica2? But you're not using the VoltageM component. So, I decided to try the same thing you did: not using VoltageM. Although I know the Xbow MPR manual describes something about the mapping from ADC7 to the ADC channel 30, and by using VoltageM in micaz. "RobotM.BVoltage - ADCC.ADC[ROBOADC_BVOLTAGE];" (I tried both ROBOADC_BVOLTAGE = 7 and ROBOADC_BVOLTAGE=30) but the dataReady-event never fires (although like I said before, I was expecting something like that when not using VoltageM)So, there remain 2 options: -the VoltageM component contains an error (e.g. wrong mapping / reading of the ADC channel)-Measuring Voltage in micaz is not possible? (although the xbow MPR manual claims it should be)greetings and thanks, R. On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: I don't remember the solution to the problem in the previous email youreference, but I think they were reading the wrong ADC or gettingwrong values somehow...perhaps Gina Upperman is still online here andcan answer. But your values look suspiciously like they have been divided by two.Do you have an accidental bit shift someplace?MSR Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) Greetings, Robin ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
I don't remember the solution to the problem in the previous email you reference, but I think they were reading the wrong ADC or getting wrong values somehow...perhaps Gina Upperman is still online here and can answer. But your values look suspiciously like they have been divided by two. Do you have an accidental bit shift someplace? MS R Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) Greetings, Robin ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: I don't remember the solution to the problem in the previous email you reference, but I think they were reading the wrong ADC or getting wrong values somehow...perhaps Gina Upperman is still online here and can answer. AFAICT, there's no voltage sensor on the micaz (at least there's no sign of it on the schematic on Xbow's web site). The micaz tinyos 1.x Voltage component tries to read the built-in V_BG (bandgap) voltage. Perusing the atmega128 documentation suggests that that voltage is not stable until 40-70us after the ADC is enabled - so the problem may be that it isn't stable. I'm also unclear on how a measurement of V_BG (as setup in TinyOS 1.x's ADCREFM and HPLADCM modules...) relates to battery voltage. But your values look suspiciously like they have been divided by two. Do you have an accidental bit shift someplace? MS R Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
I assume it's the same trick as the mica2 where the ADC Vref IS the battery, and you get an inverted reading from the internal ref which increases as the battery voltage droops. Is there a micaz schematic on the XBOW site? I keep looking for one at http://www.tinyos.net/scoop/special/hardware and then give up... Also, I just referenced your name in vain on having sent me a link to something that details the nesc include path, which I have lost as usual. Can you resend that (assuming it was you...) thanks MS David Gay wrote: On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: I don't remember the solution to the problem in the previous email you reference, but I think they were reading the wrong ADC or getting wrong values somehow...perhaps Gina Upperman is still online here and can answer. AFAICT, there's no voltage sensor on the micaz (at least there's no sign of it on the schematic on Xbow's web site). The micaz tinyos 1.x Voltage component tries to read the built-in V_BG (bandgap) voltage. Perusing the atmega128 documentation suggests that that voltage is not stable until 40-70us after the ADC is enabled - so the problem may be that it isn't stable. I'm also unclear on how a measurement of V_BG (as setup in TinyOS 1.x's ADCREFM and HPLADCM modules...) relates to battery voltage. But your values look suspiciously like they have been divided by two. Do you have an accidental bit shift someplace? MS R Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
Yes I know they are very suspicious ;) But I don't have any bit shift anywhere in my program(s)... Those are the actual values I'm gettingMichael, in your robocode you're getting the right values for mica2? But you're not using the VoltageM component. So, I decided to try the same thing you did: not using VoltageM. Although I know the Xbow MPR manual describes something about the mapping from ADC7 to the ADC channel 30, and by using VoltageM in micaz. RobotM.BVoltage - ADCC.ADC[ROBOADC_BVOLTAGE]; (I tried both ROBOADC_BVOLTAGE = 7 and ROBOADC_BVOLTAGE=30) but the dataReady-event never fires (although like I said before, I was expecting something like that when not using VoltageM)So, there remain 2 options: -the VoltageM component contains an error (e.g. wrong mapping / reading of the ADC channel)-Measuring Voltage in micaz is not possible? (although the xbow MPR manual claims it should be)greetings and thanks, R.On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: I don't remember the solution to the problem in the previous email youreference, but I think they were reading the wrong ADC or gettingwrong values somehow...perhaps Gina Upperman is still online here andcan answer. But your values look suspiciously like they have been divided by two.Do you have an accidental bit shift someplace?MSR Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) Greetings, Robin ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
I did the excavate thing a while ago and satisfied myself that the mica2 uses the moral equivalent of the Battery as the Vref. Since I couldn't find the micaz schematic I just assume that it's the same. I don't know why the 'Z uses the internal ref and the '2 has an external one. Maybe it's that settling time you mentioned. I was thinking I would find the full 'Z schematic as a separate document. But in the User Manual?...strange...Unfortunately it is not complete, it's missing the controller page. I guess it's the same as the '2...I hope. From a quick look the Vref section seems the same. As for the include path, I meant the implicit path. For instance nothing explicitly references hardware.h (in most apps anyway), but it automagically appears in the (pre-)compiled code, and, various source directories are searched in some (unspecified to me) order when compiling. I mentioned it because there was just a statement made about micaz using mica2/HPLUARTC.nc, which might confuse anyone below the level of prelate in the TOS hierarchy... oh, and my (hopefully vanilla and regular-guy kinda) TOS 1.1.7 install doesn't find a man page for ncc...is it really part of the standard install, and I didn't set my MANPATH right? MS David Gay wrote: On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: I assume it's the same trick as the mica2 where the ADC Vref IS the battery, and you get an inverted reading from the internal ref which increases as the battery voltage droops. Well I was thinking that too, but I didn't want to spend lots of time digging through the code to confirm it... (the 1.x code doesn't very explicitly select which reference voltage to use) Is there a micaz schematic on the XBOW site? I keep looking for one at http://www.tinyos.net/scoop/special/hardware and then give up... http://www.xbow.com/Support/Support_pdf_files/MPR-MIB_Series_Users_Manual.pdf (it is on xbow's micaz web page...) Also, I just referenced your name in vain on having sent me a link to something that details the nesc include path, which I have lost as usual. Can you resend that (assuming it was you...) I don't think it was. But man ncc does tell you (at least with the latest toolchain), and compiling with -v will show you all the -I directives... David Gay ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
Don't use my Robo code for Bvoltage. I have a whole different hardware scheme and it gets different values. However, you should be able to get the raw ADC readings using ADCC the way I do. The mica2 and 'Z hardware are different as well, which is why they read different channels, so your mileage may vary. If you are a nuts'n'bolts type, look at the micaz schematic, or what there is of it, in the UserGuide that David Gay just referenced, and read the ATMEGA section on using the internal ref. It's pretty turgid, but there may be a clue about the apparent x2 readings you are getting. sorry, I can't be much more help without actually doing work myself... MS R Gartz wrote: Yes I know they are very suspicious ;) But I don't have any bit shift anywhere in my program(s)... Those are the actual values I'm getting Michael, in your robocode you're getting the right values for mica2? But you're not using the VoltageM component. So, I decided to try the same thing you did: not using VoltageM. Although I know the Xbow MPR manual describes something about the mapping from ADC7 to the ADC channel 30, and by using VoltageM in micaz. RobotM.BVoltage - ADCC.ADC[ROBOADC_BVOLTAGE]; (I tried both ROBOADC_BVOLTAGE = 7 and ROBOADC_BVOLTAGE=30) but the dataReady-event never fires (although like I said before, I was expecting something like that when not using VoltageM) So, there remain 2 options: -the VoltageM component contains an error (e.g. wrong mapping / reading of the ADC channel) -Measuring Voltage in micaz is not possible? (although the xbow MPR manual claims it should be) greetings and thanks, R. On 5/19/06, *Michael Schippling* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote: I don't remember the solution to the problem in the previous email you reference, but I think they were reading the wrong ADC or getting wrong values somehow...perhaps Gina Upperman is still online here and can answer. But your values look suspiciously like they have been divided by two. Do you have an accidental bit shift someplace? MS R Gartz wrote: Hello, I'm trying to measure the voltage of the batteries of my micaz nodes. When using the VoltageM component, I'm always getting values like Vbat= 3901, 4638, 4892, 5261, 5398, 5351, 5445, 5398 - So, 5,3 Volt for two AA batteries?? Seems very strange... When printing the raw 10-bit ADC value (so without the conversion (in the VoltageM component) Vbat = Vref * 1024/ADC_Count) the ADC_Count = 315, 258, 243, 230, 227, 229, 225,... So in order to get more realistic values like Vbat = 2900mV: the ADC_Count should be around 430... Everything is mapped right, I'm measuring the ADC-channel 30 for micaz... Anyone knows what is wrong? (the same problem has already been posted a month ago, http://mail.millennium.berkeley.edu/pipermail/tinyos-help/2006-April/016227.html) Greetings, Robin ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU mailto:Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
Re: [Tinyos-help] battery voltage micaz
On 5/19/06, Michael Schippling [EMAIL PROTECTED] wrote: As for the include path, I meant the implicit path. For instance nothing explicitly references hardware.h (in most apps anyway), hardware.h is referenced from tos.h, which is magically included when you compile a TinyOS program (but if you pass -v to ncc, you will see somewhere a -fnesc-include=tos option, which is the source of the magic...) but it automagically appears in the (pre-)compiled code, and, various source directories are searched in some (unspecified to me) order when compiling. Well, as I said, make sure ncc gets a -v option, and you will see the full set of include paths searched. oh, and my (hopefully vanilla and regular-guy kinda) TOS 1.1.7 install doesn't find a man page for ncc...is it really part of the standard install, and I didn't set my MANPATH right? I did say the latest toolchain ;-) The TinyOS 2.x toolchain is supposed to work with TinyOS 1.x as of TinyOS 1.1.something recent. But looking at the release date for 1.1.15, I guess my fixes for using the 2.x toolchain in 1.x haven't made it out of CVS and into a minor release :-( David Gay ___ Tinyos-help mailing list Tinyos-help@Millennium.Berkeley.EDU https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
