Guy: Without putting too fine an edge on it and if I understand David it seems he has increased the thermal conductivity and thereby decreased the temperature differential across the heatsink assembly. This reduces thermal stress and is a good thing and similar to what you strive to do with your monster fan. The difference is one of finesse and brute force. Because neither surface area nor air path is affected it is unlikely dissipation is changed. He is just making better use of the heatsink. Thermal conductivity, the reciprocal of thermal resistance, has changed. This resistance is part of a thermal time constant affecting how fast the fan turns on and off so we should not be surprised the fan turns on and off at a different rate. 73, Fred AE6QL
-----Original Message----- From: elecraft-boun...@mailman.qth.net [mailto:elecraft-boun...@mailman.qth.net] On Behalf Of Guy Olinger K2AV Sent: Monday, June 18, 2012 10:52 AM To: David Robertson Cc: elecraft@mailman.qth.net Subject: Re: [Elecraft] KPA 500 My question is whether you actually increased the no-fan heat dissipation of the heat sink complex that significantly, or whether what you did is faking out the heat sensor and causing the fan to NOT come on when it actually should. It bothers me also that the increased dissipation is non-symetrical, and will cause one side of the heat sink to be at a different temperature than the other. This is a bad situation that theoretically can damage transistors by putting unequal mechanical stress on the mounting in high heat situations. It also could cause one transistor to be a lot hotter than the other. Or it COULD be that what you did equalized heat distribution. The problem is that YOU DON'T KNOW WHICH ONE. Fan speed, when it comes on, size of heat sink, etc and the transistors used are a very carefully engineered COMBO. I really would not screw around with it UNTIL you have discussed it with the Elecraft engineer and have his concurrence. There is a lot more here than meets the eye. Be very careful what you recommend to other owners. Elecraft please weigh in before this gets around as an ill-advised urban myth, if indeed it is ill-advised. Beyond that, why is everyone so aroused by fan speed? Fans keep things cool. Fans are good. My roaring 100 CFM monster fan on my contest 3-1000Z amp, my "Loudenboomer," keeps me from melting it when I qsy all over the band and don't remember to retune it. I get pretty stupid late into a contest. Roaring fan makes 3-1000Z last long time. 3-1000Z getting expensive and hard to find. Roaring fan is my friend. Need to use headset anyway for best diversity on 160. Use noise cancelling headset, can't hear the roar. Fans are your friend. 73, Guy. On Mon, Jun 18, 2012 at 7:42 AM, David Robertson <kd1na...@gmail.com> wrote: > Everyone, > > I have had my KPA500 linear for some time now and have been really > happy with it. I use it with my K3 and they communicate with each > other via the > DB15 AUX cable that I built ( the one furnished by Elecraft was too short). > > The only issue I have encountered is the finals in the linear seem to > heat up rather quickly causing the fan to kick in in a rather short > period of time. it doesn't matter if I use the dummy load or a good > match antenna. I also noticed the finals seem to cool rather quickly > after the transmittion is terminated and the fan turns on after about > 30 seconds to one minute after starting a ssb transmittion then goes > to a higher speed after about > 30 seconds more. If my transmittion time is greater then 2 to 3 > minutes the fans go to high and the final temperature is around 70 > degrees C. I have never had the linear go in to a fault because of heat. > > The fix. > > I removed the top, front, and right side panels of the KPA500. When I > first built the linear I realized the mounting screws that mount the Z > panel to the amplifier module were too long so I used the shorter ones > that would normally go to mount the top panel. On inspecting these > screws they were tight. Looking at the right panel I noticed there are > 4 screws that mount the panel directly to the amplifier's heat sink. > They were not as tight as I would have liked. I wanted to get the most > efficient heat transfer from the module so I carefully applied some > heat sink compound to the side of the heat sink of the amplifier > module before carefully mounting the right sied panel back on the > amplifier. I carefully made sure the 4 screws that mount from the > panel to the heatsink ( which are normally covered by the > handle) were solidly and carefully tightened. After inspecting the > transformer connections and the rest of the inside of the linear I > reassembled it and tested it out. > > Findings. > > Now when I am running ssb typically the fan doesn't come on until well > in to the qso and never has the fan reach high mode. If I brick the > key at 500 watts into a dummy load the fan goes high after about 2 > minutes. I also noted the finals heat up slower and cool slower and > the left panel temperature follows the final temperature.. > > Now the amp seems very happy and I am not bugged by the constant fan > noise during a qso. > > I thought I would pass this on to everyone. > > Thanks and 73 > Dave KD1NA > ______________________________________________________________ > 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 > ______________________________________________________________ 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 ______________________________________________________________ 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
