Re: [Elecraft] Heat Sink Compound Question
Hi Tony, I posted this reply last November. This topic appeared on the list previously from last October "sticky thermal pads for K3 LPA Mosfets". One of my LPA Mosfets went bad in my K3s. I saw a comment on the reflector about these devices failing and I'm guessing the heat sink design is subject to production variation. I used to design switching power supplies. Reliability is a very strong function of device temperature so keeping cool is important. I hate to take time to fix stuff, so try to fix it so it never fails again. I replaced both of them but also improved the thermal design. As Ray WA6VAB mentioned, I got mine from RF Parts. The epitome of outstanding thermal design is the Intel I7 processor heat sinking. The processor and the heat sink are polished, shiny, and very flat. The best thermal compound is a silver based paste included with the heat sink. It does 100 watts over a couple of square inches so very low thermal resistance (remarkable!). Any gaps compromise the performance. There is a very good tutorial on this here: https://forum.digikey.com/t/thermal-interface-materials I examined the bottom cover and found the surface of the metal to be uneven and this inhibits heat transfer (raises the thermal resistance). I also found the bottom cover may not be perfectly flat. Moreover, the typical flexible thermal pads are not as good as alternatives. I polished the K3s bottom cover where the heat sink attachments are with a Dremel tool and polishing pad and automotive rubbing compound, a fine abrasive. You can see it becomes more shiny and I verified that with a microscope. I got a larger diameter flat washer for placing under the screw head on the outside of the K3s. This places compression over a larger area to help heat transfer. Digikey part number H734-ND. Since Elecraft has different board revisions you should check that the tabs of the MOSFETs can be grounded. If so, then there is a type of thermal pad that is very good. From the data sheet: High-PERFORMANCE, cost effective thermal interface material Used where electrical isolation is not required, Tgon TM 800 is ideal for where electrical contact and thermal transfer are desired. High thermal conductivity of 5 W/mK in Z axis and 240 W/mK in the X-Y axis. This type not only has low thermal resistance across it, but also a low thermal resistance in the plane of the pad helping to spread the heat out. Digikey part number 926-1471-ND manf part number A15037-112. Use just a very small amount of the silver based thermal compound. I run FT8 so the duty cycle is 50%. It's been several months and no problems (1000s of QSOs)--at least yet! Good luck and 73, Chuck K0MV On 3/29/22 15:24, Tony wrote: All: It seems that thermal compounds are not all the same and I was wondering which type is best to use for transceiver applications? I understand Elecraft uses thermal pads these days, but what type of compound was used prior to the switch? Just purchased a produced called MX-5 from Arctic which is commonly used for CPU applications and was wondering if it can be used for HF transceivers, kits etc. See: https://www.amazon.com/gp/product/B08T64M68V/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&th=1 Thanks, Tony __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to k...@altaeng.com -- Charles K0MV __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to arch...@mail-archive.com
Re: [Elecraft] FW: KPA500 faults
Hi William, " surmise it was heating " Consider as the surface mount current sensing resistor (assuming it is surface mount) is made to withstand solder reflow it seems unlikely to be heat. I've had many surface mount resistors develop hair line cracks due to vibration and board flexing (in breadboards I built). In failures I've seen due to heat, the resistor visibly changes color especially in markings, white to brown. As long as you have the unit opened up, perhaps you should take a lupe or magnifying glass, bright flashlight and just look around for other cracks in components or solder joints. I hope I can work you on 40M FT8, I need VP2E. Good luck Chuck K0MV On 3/3/22 08:40, Dr. William J. Schmidt wrote: Sorry guys, this is not a trivial line voltage problem. The 60V line to the PA does not sag that much (starts at 72V goes to 60V under FULL LOAD as measured at the PA)... it's in the voltage divider for the 60V/32 line or the connectors that carry it to the front panel processor for digitizing. Changes with heat load. This amp is in the Caribbean and there is plenty of excess heat and corrosion. (nb. I just replaced the current sensing resistor in my K3S for the 100W module due to a hair-line crack... surmise it was heating). Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ VP2EHZ email: b...@wjschmidt.com -Original Message- From: elecraft-boun...@mailman.qth.net On Behalf Of Naumann, Robert, W5OV Sent: Thursday, March 3, 2022 9:11 AM To: elecraft@mailman.qth.net Subject: [Elecraft] FW: KPA500 faults What line voltage are you running it on? If you're not on 240v, switching to 240v would be the easiest and most effective improvement. Doing so yields a 4:1 improvement in line voltage regulation / reduction in voltage drop. Double the voltage, at half the current draw. 73, Bob W5OV -Original Message- From: elecraft-boun...@mailman.qth.net On Behalf Of Dr. William J. Schmidt Sent: Thursday, March 3, 2022 00:39 To: elecraft@mailman.qth.net Subject: [Elecraft] KPA500 faults I have a KPA500 that faults on low (60 VDC) voltage. After measuring the power supply voltage at the PA module is perfectly fine. The voltage read by the processor on the front panel via pin 10 of P1 is lower than it should be and hence the fault condition. I suspect there is a voltage divider on the LPF board that drops the 60 VDC to something that the processor can read and digitize (in the 0 to 3-5 volt range). The set of schematics I have do not show the voltage divider anywhere. does anyone have this part of the schematics for the KPA500 so that I can adequately troubleshoot this amp and get it on the air for the contest this weekend? Multiple notes to Elecraft have gone unanswered. Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ VP2EHZ email: <mailto:b...@wjschmidt.com> b...@wjschmidt.com __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to w...@arrl.org __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to b...@wjschmidt.com __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to k...@altaeng.com -- Charles K0MV __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to arch...@mail-archive.com
Re: [Elecraft] New wattmeter plans?
I came across this very interesting presentation that you may find informative: https://www.hamradioworkbench.com/uploads/1/1/9/9/119903037/k6jey_wattmeter_talk.pdf 73, Chuck K0MV On 1/13/22 7:03 PM, jerry wrote: On 2022-01-13 17:50, Josh Fiden wrote: Bird is a classic, I keep one on out of my 6m amp. But they’re not cheap. Comparing cost and features with an LP100A not a hard choice IMHO. *** Room for both. I have a Bird 4410a. Multiple ranges - can measure from QRP up to 10kW. An Ebay deal. Don't know how well it's calibrated, and it doesn't do peak power. So I just ordered an LP100A. With the work I've been doing with linear amplifiers, I am very interested in accurate power readings. - Jerry KF6VB 73, Josh W6XU Sent from my iPad On Jan 13, 2022, at 5:19 PM, Bob McGraw wrote: Why not just buy a nice clean Bird 43P and a new element or two. Not hamfest elements. That works for me and gives total confidence on what I see as power output. Of course if one desires all equipment to match, then so be it. __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to je...@tr2.com __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to k...@altaeng.com -- Charles K0MV __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to arch...@mail-archive.com
Re: [Elecraft] KPA-1500 Power Supply
Mine is powered all the time, but the RF deck is not. On 1/3/22 9:30 AM, Larry Boekeloo wrote: Do most of you leave your KPA-1500 Power Supply powered up all the time even with the RF Deck off? Thanks. Larry, KN8N Kalamazoo, Michigan __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to k...@altaeng.com -- Charles K0MV __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to arch...@mail-archive.com
Re: [Elecraft] Problem Tuning K3: now off centre fed dipole
Hi David, Thanks for the synopsis below. I found it very useful being unfamiliar with amateur practice over the years with OCF antennas. My antennas are all center fed. I replied to this message because it was the most informative. Other, later messages on this thread were also interesting. But generally, the discussion lacks quantitative data and so it is difficult to objectively evaluate the efficacy and drawbacks of different realizations. It is also difficult to project which realization or changes would fit a given set of constraints. The positive aspect is there has been years and years of experimentation and different approaches to this multiband dipole problem. I would like to hear more on these. Could you describe G3TXQ's 2 core Guanella balun solution more completely? I checked the Spiderbeam website and it is not clear to me. I think this is relevant to Elecraft since many applications are portable and QRP. We want that antenna to really work and understanding it better, helps. 73, Chuck K0MV On 11/4/21 9:59 AM, CUTTER DAVID via Elecraft wrote: Hi Dan Good question. I'll answer as a non-expert as best I can. This is essentially a quest for a multi-band dipole that is easy to make and use. If you can put up mono band dipoles and beams for all your bands, you probably will not bother going this route. The popular description of the ocf dipole (going back at least to the 50's) has a feed point of one third/two thirds on the dipole. (BTW this is otherwise a normal half wavelength dipole. This provides approximately 200 ohm feed point impedance for most popular HF bands, eg 80, 40, 20 and 10m, not 15m. Back in the 60's we could use coax into a valve PA, which I preferred to the G5RV which needed an outboard matching unit. In transistor days we used a modest matching unit with an swr meter and were ignorant of common mode current. In the 90s I used ladder line to ground level and an auto-tuning unit into a tent for field day, believing it to be more efficient. With grounded coax I didn't have live chassis syndrome and out in the wilds there was no noise pickup. Later I used ladder line through a balun then into the radio with an on-board matching unit and that was, for me a great step forward with auto-tuning. It was discovered by some users that the common mode current performance using off-the-shelf baluns and chokes was inadequate and could result in live chassis syndrome and noise pickup on receive from home locations. Poor matching on 15m was still a problem as were the WARC bands. Now in the 21st Century, just a few years ago, Rick DJ0IP tackled the problem starting with a 40m ocf dipole using a new balun/choke combination. He read that W8JI and others recommended a 20% feedpoint to bring 15m into the 200 ohm region. He then used a Guanella 2-core balun solution from Steve G3TXQ and others to provide the 50 ohm output with low common mode current. It had to be the dual core version, the single core version simply didn't suppress the cmc. This is now marketed by Spiderbeam. Moving to an 80m version was difficult but with a hybrid balun/choke combination he devised a combination that provided at least as good cmc and this version is also marketed by Spiderbeam. They both work on 15m and some WARC bands. For some layouts a modest on-board matching unit is required for complete coverage which is a far cry from bulky outboard units; good news for portable operation. I would call these *modern* off centre-fed dipoles, ie well into this century. I have no financial connection with anyone in that business and I will say that I have never seen anyone else produce such a large amount of compelling evidence as Rick and I recommend his web site for all things balun and choke related to this task. He takes the practical, non-laboratory approach. This in no way conflicts with any work done by Jim, K9YC and I only wish that he would make his own measurements on these modern devices and let go the old prejudice. 73 David G3UNA On 04 November 2021 at 05:24 Dan Presley wrote: Perhaps you could clarify what you meant by a modern OCF. What’s changed from the traditional model? Thanks. Dan Presley 503-701-3871 danpresley@me. com n7...@arrl.net __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to k...@altaeng.com -- Charles K0MV __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Pleas
