[600MRG] 20 watts out with a 1625, low plate voltage?

[Dwight Blevins]

I think my problem is with loading. It dramatically affects the plate current. I have a big antenna loading coil, probably 200uh, just outside the shack window. Multiple taps, easily changed, but the best I can do is about 70 ma at about 510 volts under load. The tap point also shows max signal strength when the same antenna is used for receiving.
Maybe I'm getting a bit over 20 watts out, but not much. It will light a 75 watt light bulb at a nice orange glow, maybe 1/3 brilliance. Moving the coil tap very much causes a max load dip to about 50 ma, which is really deep, seeing off resonance the tube draws probably 125 ma. Just a turn or two in the other direction on the loading coil and the plate will not dip/load at all. The schematic I used called for all fixed caps on the Pi network. Boy am I ever glad I decided to at least use a variable plate cap on the Pi input. Fixed caps are just too ballpark. Next time I'm going to also incorporate a variable cap on the 6AG7 output. Right now I have a 470uh inductor paralleled with about 250pf fixed cap. That gets it close, but a variable  would be better for peaking things up to max drive.
Other problem is I'm using the old FT-241 crystals. Out of 8 crystals I found only two that are half way usable. The output changes dramatically, depending on the crystal. Probably be better off with a 7.6 mhz crystal and a divide down chip. Oh, forgot to mention, I have more inductance at the far end of the inverted L. It is a 160m trap, about 200uh, paralleled with a small coax make shift cap, to resonate at or just below 160m. This kicks up the radiation resistance a bit for 630m, rather than trying to have all the loading coil inductance near the feed point at the operating shack. All old school stuff for most, and maybe a boring read for some, but at age 71 I'm still learning :) 

73, IkeKW7T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/600mrg/attachments/20180129/fbe011a0/attachment.html>