Although an infrared temperature meter will provide some very useful information, perhaps a more practical method of determining PA efficiency is the use of a DC ammeter and a accurate RF power meter. Since PA efficiency is the ratio of power out to power in, optimizing that ratio is straightforward. For example, if a PA is drawing 20 amperes at 14 VDC, the power input is 280 watts. If an accurate RF power meter shows that the output power is 90 watts, then the efficiency is 90/280 or about 32%. Since 90 watts is leaving the PA as RF power, the remaining 190 watts is spent heating up the PA. What Jeff recommends is adjusting the impedance matching device and/or the power setting so as to optimize the efficiency. In other words, get more power to the antenna and reduce the amount of internal heating. It may turn out that the point of optimum efficiency is not at the rated power output setting.
It's one thing to perform this efficiency test on the bench while feeding a purely resistive dummy load, and quite another thing to perform it at the site while feeding a reactive load such as an antenna or a duplexer. That's one reason why having an isolator at the PA output is a good thing- it stabilizes the PA load at 50 ohms. Even without an isolator, anything that can be done to reduce the amount of real power expended in the PA transistors will definitely prolong their lives. 73, Eric Lemmon WB6FLY -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Nate Duehr Sent: Thursday, June 25, 2009 11:51 PM To: [email protected] Subject: Re: [Repeater-Builder] Repeater Transmitter matching Jeff DePolo wrote: > The changes that occur are at source end inside the PA. The ideal load Z > for the output devices changes when you vary the output power. If you > "tune" the Z-matcher to present a load Z to the devices that results in them > operating at maximum efficiency at at 100 watts, and then reduce drive (or > collector voltage) to yield only 30 watts, the load Z will likely not yield > max efficiency at the new power level. > > Make more sense now? Yup, got it. Your notes here and Bob's comments about the possible internal reasons why, and I'm all straightened out. Interesting that as I think about it from a VERY high-level/general point of view... it seems that unless you're willing to take the time to do the measurements you've mentioned Jeff to get the maximum efficiency at the desired power level, the old saw of "Just run it flat out at rated power" for the old MASTR II PA, probably yields the best "general" advice for them, especially without a Z-matcher or isolator (!). Turning them down (as we've all noted previously) does NOT always mean that they run cooler due to the aforementioned inefficiency "built in" to them, so to speak. We've all seen that one with an IR temperature measuring device pointed at the heatsink and/or power transistors. The reason I mention all of the above... we were "holding back" a little more each time we swapped a VHF back when we had them popping like lightbulbs. For whatever non-scientific, non-engineering reasons, although I'm sure (until we learned better), we thought they would "run cooler" backed off 20-25W. Now I'm starting to see the light as to why other sites (running flat out at rated power or REAL close to it with isolators and some with internal, some with external Z-matchers, just kept going like the Energizer bunny and the sites we had no antenna system problems (that we could ever find), kept popping VHF PAs. A later beautiful highly-detailed rebuild of the last remaining problem-child site not only saw us installing the cleanest rebuilt-from-mobile-parts MASTR II VHF PA at any of our sites, but we also let it cruise along at a bit higher power. What say you guys to the idea that minus doing the testing "just run it" is good advice for someone incapable or unwilling to do that level of testing? I've been doing it now when I can, but we stopped the tsunami of dead VHF PA's after five (!) and I haven't had to worry about it since then. BTW: The UHF PA's are tanks. ESPECIALLY the 75W version. NEVER seen one of those die, yet. Amazing little beasts. Seen one 100W UHF die in seven/eight years. (And we get a TON of lightning so sometimes damage occurs that we don't know about... maybe failing during the storm, maybe failing much later.) Nate WY0X
