I don't know if anyone has mentioned this bug yet - I was playing with Preview 11 and noticed that, in transmit, the Peak Power Reading meter reads *much* too high. Preview 9 also reads much too high, too, when I tried that version.
Version 1.4.4 however, gives me a peak power reading of just over 100 watts, which I've independently confirmed with my oscilloscope. Preview 11's Peak Power display, though, hits 220 watts, and I *know* the PA's not generating this much power (again, verified w/scope). I poked around the code a bit, and noticed that the Preview versions have a "min" calculation (in sdr.c) that limits the reading to be, at most, 2.2 (i.e. 220 watts) if the peak calculation is larger than this value. If I remove this 'min' restriction, the Peak Power reading zooms waaaay up. In fact, if I look at the variable uni.meter.tx.val (in do_tx_meter), it can hit values of 400 or more. Whereas this variable only reaches about 1.0 (i.e. 100 watts) in version 1.4.4. Given that the remaining calculations (after this point) in Preview 11 mimic those in 1.4.4, I'm thinking that uni.meter.tx.val in Preview 11 should max out at around 1.0, too, rather than 500x (or so) this value. The peak calculations in do_tx_meter look reasonable to me, so I'm guessing that the problem appears earlier than this routine, with the result that the data used in these calculations is much larger (by orders of magnitude) than it should be. (Or perhaps the data is supposed to be much larger, and a scaling factor needs to be applied?) Unfortunately, I'm a hardware guy, not software, and given that I'm unfamiliar with the code, with the tools, and with much of the coding nomenclature (through my own ignorance), I'm fast approaching my limits at finding the root of this problem... 73, - Jeff, WA6AHL P.S. By the way - I noticed that uni.meter.tx.val also is very large when the meter is in ALC mode, but I don't know if this is as it should be, or not.
