Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!cs.utexas.edu!uwm.edu!csd4.csd.uwm.edu!info-high-audio-request
From: jhess@orion.oac.uci.edu (James Hess)
Newsgroups: rec.audio.high-end
Subject: Re: Strange tube amp problem
Message-ID: <8900@uwm.edu>
Date: 14 Jan 91 14:06:22 GMT
Sender: news@uwm.edu
Lines: 46
Approved: tjk@csd4.csd.uwm.edu
Originator: tjk@csd4.csd.uwm.edu

In article <8806@uwm.edu> warner@ecs.umass.edu writes:
>Some bizarre things have been happening to me with a Dyna Mark IV tube amp. 
>
>First of all, I replaced all of the passive components with modern devices and
>purchased matched pairs of 6CA7 output tubes. The amps worked wonderfully for
>two months until I began to notice the 6CA7 in the left hand socket of the
>problem amp was showing a glowing red spot on one side of the plate. 

>and the plate/screen voltage to be 300/225V respectively. This was alarming
>because the tube on the opposite socket was reading a healthy 446/445V.
>pair of EL34s. I plugged it in, the plate and screen voltages measured within
>+/- 2V on both tubes, the bias voltage held steady and the line to the grid 
>from the driver was holding at a nice, steady -33V. 
>
>One day,...
>Upon return, I noticed that the sound had changed significantly during my 
>absence. I glanced at the amps and to my shock I saw the entire plate assembly
>of the left tube in the troublesome amp to be glowing bright orange. 

Red spots mean excessive current flow, which is why the plate voltage is low.

I remember the one time I got to work on an Audio Research preamp.  It was
acting funny, but I couldn't find any defective parts.  Finally, with the help 
of factory techs, I found out that it had been modified--one of those changes 
that was supposed to increase the "effortless sound - improves the 'air'".  It
involved some cap changes that destabilized the circut and caused ultrasonic 
oscillation, sometimes motorboating.  It left the circut in a critical state 
in which slight deviations from nominal in the values of the components would 
upset it.  Tweeks not familiar with the requirements of design for the real 
world failed to consider these problems.  Later I ran into a $5,000 esoteric 
solid state amp designed that way--let the bias adjustment drift 20% and 
poof!  There go a dozen output transistors!

I think your problem is in those new passive components -- a capacitor that is 
mislabled, out-of-spec, or not properly soldered, etc.  Start by redoing your 
solder joints.  Connect a scope to the output and look for oscillations.  Tap 
and spray-cool components to locate sensitive ones.  (Many components are 
temperature sensitive, so look for unusual sensitivity and compare the results
between the amps.)  Switch suspect components from one amp to the other.  (A 
lot of work!)  But first -- check your connections, particularly the inputs.  
It could be a poor ground -- clean the plugs and jacks, replace the cables, 
etc.

Happy Hunting!

Jim Hess