Hello Everybody,
Hopefully somebody can help me here.
For about ten years now I have had a pair of McIntosh M-194A output
transformers that were originally used in the McIntosh MI-60 Amplifier
(Rack mounted version of the MC60). I have just built some amps using
these transformers (MI-60 schematic), but I have an intermittent
oscillation problem with one of the amps. Both amps are wired correctly
and I didn't make any mistakes with wiring the transformer leads. The
one amp that oscillates only does so when the 250k input level control
pot is either turned all the way down or all the way up. I checked the
pot; it's good also. The oscillation is ultrasonic but I know it's
occuring because the output tubes will start to pass way too much
current.
I was able to stop the oscillation on the problem amp by removing the
750pf capacitor that's wired in parallel with the 1.3k feedback
resistor, coming from the output transformer's feedback winding.
Without this capacitor, the amp's bias current is rock solid and very
stable.
Here is my question:
What exactly does this "feedback" capacitor do? Does it provide more
high frequency feedback? Does it correct phase shift problems? Can I
just omit this capacitor and still have the amps perform to
specification?
The strange thing is that this problem is only with one amplifier; the
other amp has no problems with this capacitor in place. Both amps are
built exactly the same.
Thanks in advance for any help.
Daniel Franklin

Is there a long run of wore involved in the feedback wiring?
I have had this same problem with a Wllamson clone and it turned out to be a
wire dress issue.

--
Alan Gallacher
Born to Tinker!

wrote in message
oups.com...
Hello Everybody,
Hopefully somebody can help me here.
For about ten years now I have had a pair of McIntosh M-194A output
transformers that were originally used in the McIntosh MI-60 Amplifier
(Rack mounted version of the MC60). I have just built some amps using
these transformers (MI-60 schematic), but I have an intermittent
oscillation problem with one of the amps. Both amps are wired correctly
and I didn't make any mistakes with wiring the transformer leads. The
one amp that oscillates only does so when the 250k input level control
pot is either turned all the way down or all the way up. I checked the
pot; it's good also. The oscillation is ultrasonic but I know it's
occuring because the output tubes will start to pass way too much
current.
I was able to stop the oscillation on the problem amp by removing the
750pf capacitor that's wired in parallel with the 1.3k feedback
resistor, coming from the output transformer's feedback winding.
Without this capacitor, the amp's bias current is rock solid and very
stable.
Here is my question:
What exactly does this "feedback" capacitor do? Does it provide more
high frequency feedback? Does it correct phase shift problems? Can I
just omit this capacitor and still have the amps perform to
specification?
The strange thing is that this problem is only with one amplifier; the
other amp has no problems with this capacitor in place. Both amps are
built exactly the same.
Thanks in advance for any help.
Daniel Franklin

On 30 Mar 2005 18:47:26 -0800, wrote:

I was able to stop the oscillation on the problem amp by removing the
750pf capacitor that's wired in parallel with the 1.3k feedback
resistor, coming from the output transformer's feedback winding.
Without this capacitor, the amp's bias current is rock solid and very
stable.
Here is my question:
What exactly does this "feedback" capacitor do? Does it provide more
high frequency feedback? Does it correct phase shift problems? Can I
just omit this capacitor and still have the amps perform to
specification?

To answer your last question first, yes with a qualifier. The
phase lead capacitor's gig is to fudge signal timing all the way
around the loop, all the way from the amplifier's input, all the
way to its output, and back through the feedback path, to within
a close enough guesstimate to time alignment so that the whole
thing doesn't think it's an oscillator.

IOW, it's all a fudge factor. The forward path has *lots* of
delays and if the delays add up to 180 degrees and gain is greater
than one, bingo, you've solved the Zen koan:

"Oscillators don't. Amplifiers do." The question is, what oscillates?


The strange thing is that this problem is only with one amplifier; the
other amp has no problems with this capacitor in place. Both amps are
built exactly the same.

Obvious not *exactly*. God and the Devil dwell in the details.

Good fortune; you've picked an amazingly complex commercial design
to replicate. Quite a baptism of fire-bottles,


Chris Hornbeck
6x9=42
"Right, the DBT is for all intents and purposes vastly superior to
casual listening."

wrote
...one amp that oscillates only does so when the 250k input level
control
pot is either turned all the way down or all the way up....

Is this the right circuit?

http://www.drtube.com/schematics/mcintosh/mc60.gif

Is that the way you have the input wired?

cheers, Ian

i'd try swaping tubes to see if it is related to that

also look for some funny routing of the wires for the input and for the
output xformer primary windings.

check all the small capacitors

also check the resistors in series with the volumne control wiper.


Mark

Is this the right circuit?

http://www.drtube.com/schematics/mcintosh/mc60.gif

Is that the way you have the input wired?

cheers, Ian

Actually, from the input jack, the signal goes through a .47 cap to the
"high" end of the 250k pot. The wiper of the pot then goes to the grid
of the 12AX7. I'm using shielded microphone wire for these connections.
This is the way the early MI60/MC60's were wired.
I have thought about changing the input to the way your schematic is,
but am afraid that doing so will alter other aspects of the circuit.
Daniel

wrote

Actually, from the input jack, the signal goes through a .47 cap
to the
"high" end of the 250k pot. The wiper of the pot then goes to the
grid
of the 12AX7. I'm using shielded microphone wire for these
connections.
This is the way the early MI60/MC60's were wired.
I have thought about changing the input to the way your schematic
is,
but am afraid that doing so will alter other aspects of the
circuit.

Now I am mystified.

You might think that the problem would occur when the pot is midway,
when source impedance appears highest to the input valve. The
opposite is true in your case.

But...why the cap before the pot? Are you sure you are not changing
the DC conditions of the first valve when you vary the pot? Have you
got a bad connection somewhere? Have you measured and compared with
the good side?

cheers, Ian

Well, I fixed the problem! I ended up grounding (to chasis) the common
of the secondary (speaker output) of the output transformer.
Traditionally I have always left the secondary "floating" on McIntosh
amps, because for some reason I thought they sounded better that way. I
guess now with the secondary tied to ground, it is actually wired
exactly like the schematic! Why is this? Logically, it makes sense as
now the secondary "coil" is not "floating" all over the place. Maybe
something in this particular transformer.

As to your input question: The maximum impedence the input sees is 250k
ohms at max "volume", the way it is currently wired. I have noticed on
McIntosh amps (mc30, mc60, etc.) that after every few hundred units
produced, they tinker with the schematic a bit. It's times like this
that I could pick one of their old-timer's brains and ask why this is
so!

Anyway, thanks again to all who gave input on this thread.
Daniel



But...why the cap before the pot? Are you sure you are not changing
the DC conditions of the first valve when you vary the pot? Have you
got a bad connection somewhere? Have you measured and compared with
the good side?

cheers, Ian