I set aside the previously-posted KG-686 and completed the restoration of the simpler KG-650 Signal Generator. Actually, the kit-built unit I have has no indication of a model number, but is electronically identical to the KG-650.

BUT, here’s my problem. The frequency range is in four bands. (I’m ignoring the fifth band)
A 160-550 KC
B 550-1850 KC
C 1.85-7.0 MC
D 7-27 MC

Checking with a calibrated frequency counter across the entire range for each band:
Band C - RF output matches the dial settings exactly.
Band D - RF output is 5% below dial settings.
Band A - RF output is 33% below dial settings.
Band B - RF output is 50% below dial settings.

In this restoration all questionable components have been replaced. The original wiring instructions have been reviewed step-by-step. Most of the original wiring was replaced. Every connection was re-soldered or at least re-melted.

https://farm2.staticflickr.com/1538/26220417056_182093999b_z.jpg

Can someone give me a clue as to what might be wrong?
Six photos of restoration:
https://flic.kr/s/aHskxHUPeb

Some discussion here about band B operating at 1/2 frequency... perhaps it was designed that way, as it was observed on several units... makes no sense. How are you measuring frequency? The output of these cheap kit generators (like my Heathkit sg 8) is far from being a pure sine wave, which could fool a frequency counter.

http://antiqueradios.com/forums/viewtopic.php?t=153481

jr

is that a new new pot/switch, or did you unstick the old one?

Well, I'm using a Heathkit IM-4100 counter which agrees with a good RF generator. I have a vague understanding of what you're saying. I suppose I could use an aligned radio to see if I'm actually getting 455KC for the IF--or just compare it with the good generator at any frequency.

That is the original pot/switch. It took a lot of oil and some muscle, but it seems to work OK.

Thanks.