Skipp, I disagree with your definition of a "proper fix." The method you describe does not address the fundamental concept of my posting, that of the PA source impedance not matching the load impedance of the duplexer.
You propose to alter the load impedance of the duplexer to (hopefully) match the source impedance of the PA, but that will likely not result in an optimum match. In fact, the range of source impedance over which a solid-state PA may vary far exceeds the input tuning range of a duplexer. When using a Bird wattmeter as you suggest, one has not fixed the problem of impedance mismatch; one has merely lessened its detrimental effect. As soon as the wattmeter is removed from the circuit, the circuit mismatch is worsened. Moreover, the presence of the wattmeter in the circuit alters and invalidates the measurement. Every manufacturer of cavity duplexers warns to NEVER tune a cavity under transmitter power, because sparking will occur as the internal spring fingers make and break contact due to stiction with the silver coating. The pitting is very easy to see, and most cavity makers will void the warranty if the cavity was tuned under power. Let's get back to the fundamental premise of my posting: The maximum amount of power transfer between a source and a load occurs only when the source impedance matches the load impedance. That's a basic law of electricity. Of course, the interconnecting cable must have a characteristic impedance that matches both the source and the load. All reputable cavity duplexer makers tune their products not just to pass or reject certain frequencies, but also to do that at a specified load (50 ohms) and with maximum return loss- that is, with the minimum VSWR. I pointed out that when one attempts to feed a 50-ohm duplexer over 50-ohm cable with a 30 ohm (or 70 ohm) PA, the PA will not be efficient and it will get hot. The answer is not to try to retune the duplexer to 30 or 70 ohms, because you now have some 50 ohm cable whose length becomes reactive. I pointed out that the logical solution is to alter the output impedance of the PA to be 50 ohms. When that is accomplished with the recommended impedance matcher, maximum power transfer occurs. It's logical, and it works. It should be noted that the above problems seldom afflict tube-type PAs, since they are usually provided with many tuning adjustments to match the impedance of following stages or components. Readers who wish to explore the topic of impedance matching of power amplifiers should visit the Web sites of EMR, Decibel Products, RFS/Celwave, Telewave, and others for many applications of impedance matchers. I didn't invent impedance matchers; they wouldn't be in commercial radio catalogs if they weren't extremely useful! 73, Eric Lemmon WB6FLY skipp025 wrote: > > Eric, > > The proper fix would be to insert a Bird watt meter between the PA and the > duplexer input, with proper size 50 ohm terminations on both the antenna and > rx ports. > > The duplexer is quickly adjusted for min reflected power, which should be > under a few watts max. > > One of the last steps most people seem to leave out of a standard duplexer > adjustment is to quickly tune the tx side cavities for min reflected power > (SWR) in/under/at rated tx power. > > The proper tuning of the tx side of duplexer is almost never the static > adjustment values. > > You could easily do without the impedance matching boxes. > > cheers > > skipp > > skipp025 @yahoo.com > > www.radiowrench.com > > > Eric Lemmon <[EMAIL PROTECTED]> wrote: > > If your repeater's solid-state PA is running hot, and the power output > > is lower than you expect, you may have an impedance mismatch. This > > problem cannot be corrected by adding a ferrite isolator. That's > > because the isolator (assuming it's properly tuned) will present a > > nearly perfect 50 ohm load to the PA, regardless of whether the antenna > > is connected or not. But, if the PA's output impedance is not 50 ohms, > > the isolator is no help. > > > > Solid-state PAs, especially the PA modules often found in mobile radios, > > frequently do not have stable output impedances. What's worse is that > > the output impedance changes with the drive level. In order to achieve > > a stable output impedance that is very close to 50 ohms, an impedance > > matcher is needed. > > > > Here's an example: I am fielding a 2m repeater that is based on the > > Motorola R1225 50-watt full-duplex transceiver. When I connected a Bird > > 43 wattmeter and a known-good 50 ohm dummy load to the transmitter port, > > I found that the factory power level was almost exactly 50 watts. When > > I hooked the rig to a duplexer, the power output dropped significantly. > > The radio was running hot, more so than it should when delivering 50 > > watts. > > > > On a hunch, I connected a VHF impedance matcher right at the TX port and > > then tuned it for maximum power output into my dummy load as measured by > > the Bird meter. Surprise! The power climbed up to nearly 70 watts, far > > more than the PA's rating. I lowered the drive level via the service > > menu in RSS until it was 50 watts when the impedance was tuned for > > maximum power transfer. When the duplexer was reconnected, the expected > > power was seen going to the antenna, minus the duplexer insertion loss. > > Even after several minutes of TX, the radio was relatively cool. The > > current draw of the radio was about 15% lower than before. > > > > Besides optimizing the efficiency of a PA, installing an impedance > > matcher also eliminates the need for a "magic" length of cable between > > the PA and the duplexer. When the PA is tuned for a 50 ohm source > > impedance, and the duplexer is tuned for a 50 ohm load impedance, the > > length of the interconnecting cable is irrelevant. > > > > I have used impedance matchers manufactured by Telewave and EMR, with > > similar results. These little gems cost between $80 and $100, but the > > benefits are well worth the cost. Just remember that the benefits are > > realized only at one drive level; if the drive level is reduced during a > > power failure, the impedance will no longer be properly matched. > > > > 73, Eric Lemmon WB6FLY > > > > Yahoo! Groups Links > > To visit your group on the web, go to: > http://groups.yahoo.com/group/Repeater-Builder/ > > To unsubscribe from this group, send an email to: > [EMAIL PROTECTED] > > Your use of Yahoo! Groups is subject to: > http://docs.yahoo.com/info/terms/ Yahoo! Groups Links To visit your group on the web, go to: http://groups.yahoo.com/group/Repeater-Builder/ To unsubscribe from this group, send an email to: [EMAIL PROTECTED] Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
