[Amps] instability issue

[N1BUG]

I'm trying to understand a problem with one of my amps. It is 
homebrew, 160-10 meters, using a 4CX1500B with screen grounded, grid 
RF grounded, cathode drive. It uses a conventional tapped coil pi 
network with vacuum variable capacitors.

The amp was going into oscillation when using it on 12 meters 
(frequency of oscillation unknown, but not in-band). When it did 
this it was also destroying the vacuum relay on the output. The 
relay would fail with contacts stuck in the NO position (contact 
connected to amp output).

While investigating I discovered the amp will always go into 
oscillation as soon as operating bias is applied IF there is no load 
on the output AND the LOAD capacitor is set to anything MORE than 
about 700 pF. Oscillation will stop if the load cap is reduced to 
something under 700 pf. This happens regardless of band switch 
position or plate tune setting.

On most bands the amp does not break into oscillation during normal 
operation. That was only happening on 12 meters for reasons not 
understood by me. I managed to get it to stop doing that by 
adjusting the coil tap so that more output capacitance is needed. 
(That coil tap needed to be optimized anyway, as it was never in the 
ideal spot)

Clearly this amp is not as stable as it should be. I am trying to 
understand what is going on here. I am having trouble grasping why 
the load cap seems to be the critical parameter, with coil tap and 
tune cap having essentially no influence. Any suggestions?

Thanks,
Paul


-- 
Paul Kelley, N1BUG
RFI Committee chair,
Piscataquis Amateur Radio Club
http://www.k1pq.org