Just picked up an early 8719 Washinton base with similar issue. As it turned out.
During its life someone put a PE Mod with the final volted. 2 watts. They had a 2sc2166 driving a 2sc1969.
The 2sc1969 was bad. Several caps were out of tolerance. The original is a 2sc1306 driving a 2sc1969.
I changed the Main power suppply cap from 3300uf to 10,000 uf. The 33 uf which the schematic showed should be 220 uf to 220 uf.
The 1000uf 10v to 100016v etc.
Still 2 watts dead key AM. My aging brain kicked in.
I have test rig for Bi-polar NPN rf finals.
An old Uniden Pro 525 with a socket in the back and a potentiometer to vary bias.
I installed an NEC 2sc2078 as the final for the Washinton. Eureka !
12w SSB + & 4 watts RMS AM with an upward kick. 30ma bias 2sc1306 & 50ma bias 2sc2078.
The adjustments are noted above. VR8 Driver Bias, VR9 final bias.
Forgot to mention. I changed out the 2sc1675 Pre-Driver.
The old one was still good though its HFE was lower than the new one.
I did not remove the NPE mod.
She sounds good.
The 2sc2166 & 2sc2078 do not require much drive.
Hope this helps someone. The peaks are high on SSB and AM.
The above output transistors I have are excellent fakes. Made by HG. Their 2sc2078
is at spec.
The 2sc1969/2sc2166 are close.
Depending on Radio the Bias has to be upped or lowered 10% for full output.
The 2sc1306 & 2sc1307 seem to be at spec.
73s Best John
