On Sat, Jan 7, 2012 at 8:39 AM, Peter Gottlieb <n...@verizon.net> wrote: > Ah, thanks, that was enough to get me thinking, to pull me away from my > job's problems and back to the fun side. > > I just replaced my wife's computer, and old obsolete big Dell box, with a > sleek new (and much faster) laptop. I was about to throw out the old box > but now have a use for the large CPU heatsink/fan assembly. I will mount > the FE-5680A to an aluminum plate and then to the heatsink. I don't need > the plate thermally, but it makes the mechanical mounting much easier. I'll > use some thermally conductive pads between things. I will use something > like the circuit you provided (thank you), I have a bunch of those TO-92 > temperature sensors with wires attached, more surplus from work, and then > will mount the whole thing with power supply into a box where I can set up > the air flow like I want. > > So, here's a question. One app is a rack of gear which all needs to get the > 10 MHz. I could just go find some distribution amp, but I would prefer to > build something. Has anyone done this? I was thinking perhaps a good solid > reasonably high power op amp buffer feeding resistors to each output to each > piece of gear? Anyone done this and found any "gotchas" or success stories?
I think this is exactly what you want. An RF distribution amp using video amplifier chips. The kit is no longer available but the schematic is. Look near the end of the user manual and you can get that here. This design is well tested and people way it works well. http://www.tapr.org/kits_tadd-1.html The transformers in the above are not available but I bet you could take some using toroid cores. I guess you could also use an n-way splitter from Mini Circuits, Or maybe multiple splitters with just one RF power amp. I think I will build an analog fan controller too. But in the long term I want a digital fan controller. They cost about the same to build. Get a "tiny avr", that's uP in an 8-pin DIP package for about $2. This will have an A/D converter for the temp sensor and PWM for the motor but the good part about a digital controller is you can read the tachometer output from a three wire fan and now you can control the exact RPM. The loop can be very stable. Actually a digital controller would have two loops, one to server the fan to the set RPM and one to control the RPM based on temperature. I think you get the best result from the largest fan you can fit in there running at a very low RPM. These digital controllers can hold the temperature very close. I happen to like Atmel AVR chips, a PIC could also work. So I Googled "AVR fan controller source code" and found dozens of published projects. Seems it is the project many beginners do right after the blinking LED. > > Peter > > > > > On 1/6/2012 1:51 PM, ewkeh...@aol.com wrote: >> >> I just grabbed something I had around it is a 24 V 0.1 A. I run at 15 V, >> dimensions are 80X80X24 mm I just bought some 80X80X10 mm and I am sure >> they will work as well. There are so many choices I recently bought a new >> one >> with integrated heat sink and tried it on a FRS all for $ 6 shipping >> included. >> Attached are two circuits I use, the top one since I did not have a PC >> board. I now have a board and I used in an other application the two stage >> one >> and if you use a heat sink I recommend replacing the feedback resistor on >> stage two with a capacitor. >> Bert Kehren >> >> >> In a message dated 1/6/2012 1:16:01 P.M. Eastern Standard Time, >> n...@verizon.net writes: >> >> What kind of temperature controlled fan did you use? >> >> >> On 01/06/12, ewkeh...@aol.com wrote: >> >> I do not understand why this is even discussed. Running at lower >> temperature will extend life and using a fan with temperature control >> will >> cost no >> more than $ 12 and I challenge any of you how I can get for so little >> money >> more than one order of magnitude improvement. As I reported before I >> started >> out with heat sink only and quickly realized that I would not be able to >> measure aging because the last 2 digits where all over the place and >> unless >> you have an environment where your lab is within 0.1C you are throwing >> away the real advantage of a Rb. >> I did enclose the Rb cell and the OCXO on a FEI 5962B, its modularity >> lends >> it self for such testing, it was not worth the effort and the power >> saving >> was minimal. >> Once my aging tests are completed I will test for 15 V voltage >> sensitivity. >> Bert Kehren >> >> >> In a message dated 1/6/2012 11:35:02 A.M. Eastern Standard Time, >> n...@verizon.net writes: >> >> A heat sink may not be required, per se, although I would expect that a >> larger thermal mass and/or thermal regulation via a closed loop fan >> controller will help smooth out/stabilize temperature effects. >> >> >> On 01/06/12, Chuck Forsberg WA7KGX N2469R<c...@omen.com> wrote: >> >> The Tech Manual does not call for heat sinking (unless I missed >> something). >> The top has labels over much of the surface. >> The bottom has a plastic sheet between the circuitry and bottom plate. >> It appears the unit was expected to be rather hot when running. >> I have mine mounted on the out side of the box using standoffs. >> On 01/06/2012 07:39 AM, Bob Smither wrote: >>> >>> -----BEGIN PGP SIGNED MESSAGE----- >>> Hash: SHA1 >>> >>> Chris Albertson wrote: >>>> >>>> On Thu, Jan 5, 2012 at 6:11 PM,<[1]time-n...@custodes.info> wrote: >>>> >>>>> l<[2]http://www.freqelec.com/rb_osc_fe5680a.html> says 32W peak, >> >> but then >>>>> >>>>> also 15-18v@700mA, which doesn't make sense. >>>> >>>> It will pull 35W for the first five or so minutes then the current >> >> drops >>>> >>>> rather suddenly to about 700mA. >>>> >>>> I have an analog amp meter on my power supply and I can see a switch >> >> over >>>> >>>> after the unit heats up. They must run an internal oven heater full >> >> tilt >>>> >>>> at first then go into regulated mode. >>>> >>>> Some one else said you can cause the FE5680 to draw more power in >> >> steady >>>> >>>> state mode by adding heat sinking it. Yes that works. Seems the >> >> FE5680 >>>> >>>> wants to be at some set temperature and the heat sink means it takes >> >> more >>>> >>>> power to keep at the set point. I just let the fe5680 rest on a >> >> small >>>> >>>> aluminum plate. >>> >>> Have you measured the case temperature of your FE5680? >>> >>> I put mine on a heat sink and the case temperature stays around 50C. >> >> Without >>> >>> the heat sink it was around 60C. Does anyone know what temperature is >>> recommended? The 50C seems a little hot, but the unit appears to work >> >> well. >>> >>> - -- >>> Bob Smither, PhD Circuit Concepts, Inc. >>> >> ======================================================================= >> == >>> >>> Government is not healthy for children and other living things. >>> -- Jeff Daiell >>> >> ======================================================================= >> == >>> >>> [3]smit...@c-c-i.com [4]http://www.C-C-I.Com 281-331-2744(office) >> >> -4616(fax) >>> >>> -----BEGIN PGP SIGNATURE----- >>> Version: GnuPG v1.4.5 (GNU/Linux) >>> Comment: Using GnuPG with CentOS - [5]http://enigmail.mozdev.org >>> >>> iD8DBQFPBxXFsmY7BY+CYksRAlutAJ9R9STR0oja4ib1CjKXLEfGe3uXpgCfRW8D >>> m78FO0Trn+6bIBPKrNs8PVQ= >>> =FdtA >>> -----END PGP SIGNATURE----- >>> >>> >>> _______________________________________________ >>> time-nuts mailing list -- [6]time-nuts@febo.com >>> To unsubscribe, go to >> >> [7]https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> >>> and follow the instructions there. >> >> -- >> Chuck Forsberg WA7KGX N2469R [8]c...@omen.com [9]www.omen.com >> Developer of Industrial ZMODEM(Tm) for Embedded Applications >> Omen Technology Inc "The High Reliability Software" >> 10255 NW Old Cornelius Pass Portland OR 97231 503-614-0430 >> _______________________________________________ >> time-nuts mailing list -- [10]time-nuts@febo.com >> To unsubscribe, go to >> [11]https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> References >> >> 1. mailto:time-n...@custodes.info >> 2. http://www.freqelec.com/rb_osc_fe5680a.html >> 3. mailto:smit...@c-c-i.com >> 4. http://www.C-C-I.Com/ >> 5. http://enigmail.mozdev.org/ >> 6. mailto:time-nuts@febo.com >> 7. https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> 8. mailto:c...@omen.com >> 9. http://www.omen.com/ >> 10. mailto:time-nuts@febo.com >> 11. https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> >> >> >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. -- Chris Albertson Redondo Beach, California _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.