A bad pot could be the case. Especially true if one varies differently than the other.
I've not found the bias to be "jumpy" or all that sensitive in adjustment for mine
(N=5). The bias does increase with heating but not enough to be a problem.
I assume you have a dummy load at the output and input.
Note Q206/7 are the IRF510 finals. The BS170 bias is independent of the
final bias but share common 12 switching.
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
