KI4YAN
Member
Offline
Posts: 161
|
|
« on: October 10, 2017, 02:56:35 AM » |
|
I have built a little tandem match directional coupler, been working on it off and on for a few months, when I had a chance at a VNA for a few hours I tweaked it up to work *very* nicely. After I had it working, I started looking for the associated metering circuits, although I'd peeked at a few during construction I didn't think about the amount of coupling being a problem-KI4YAN tends to do things like that.
Anyway, so I have a nice directional coupler assembly, with negligible insertion loss (couldn't actually measure it), return loss from 1.8 to 54mhz is better than 30db, and the coupling is a very constant -26db over the same range, in both directions. It took some fiddling to get it that good and it's down to the level of torque on the cover screws that it stays that good. I would prefer greatly to NOT have to open that cover up!
Now, when I built it I settled on -26db coupling, because doing so would result in 100W on the main line being shown as 0.25W on the detector input. Now, I read through the Tandem Match article again, and find they used a coupling of -43db! That would mean the tandem match would present a 100W main line signal as 0.005W at the detector! I'm off by a factor of 50 here. Luckily, since the tandem match uses op-amps to correct for the diode characteristics, and op-amps to drive the meters, I should be able to adjust things by a factor of 50 to compensate. (I hope!)
Does anyone have any references or articles similar to the ARRL tandem match article that I can look at to compare designs against? Has anyone here built the Tandem match? I am hoping to get a nice forward power/SWR combo meter that will be able to measure down to 1W forward power, and up to 500W forward power maximum. I'd rather have accuracy on the low end of the scale than the top end. Really, when you're going for maximum smoke, it doesn't matter if you've got 250W or 260W on the line, but when you're setting driving power to 1W, that 0.05W error could make the difference!
|