Magnavox 8601 back on bench questions

[Tubeglowpio]

Hey guys, so the Maggie 8601 was the first tube amp I ever rebuilt with only the original iron and all new parts. This was about 3 or 4 years ago now and I quickly moved to building other audio gear but I always felt that the Magnavox should have sounded a little bit better, I did modify the circuit a bit from the original such as removing the tone circuit and such. I finally threw it back on the bench today and put it on the scope for the first time and was surprised to see that I was only getting about 400mv rms across an 8 ohm load. This means I'm getting well under a watt of output power. Something really doesn't seem right here, I first want to say I measured about 142 vpp signal on the plates of the output tubes, does that seem too small of an output swing. Voltages seemed ok on outputs, around 255 on plates, 8.2v across cathode resistor 0 v at pin 2 and around 220 on pin 9. Output transformers resistance measures close to sams specs. Any ideas? let me know.

 

[nj pheonix]

I'm guessing all stock.
If you read Dave's article , the OPTs are 4 ohm.
Guessing that's part of the problem

 

[nj pheonix]

http://audiokarma.org/forums/index....eries-a-lot-more.665735/page-32#post-11368744

I'd start here.
Many happy customers

 

[Tubeglowpio]

I'm guessing all stock.
If you read Dave's article , the OPTs are 4 ohm.
Guessing that's part of the problem

I know they are 4 ohm but even so I should be getting more out than what I am. 8 ohm loads should only cut the power output in about half.

 

[FlaCharlie]

I know they are 4 ohm but even so I should be getting more out than what I am. 8 ohm loads should only cut the power output in about half.

I haven't built Dave's famous mod but it gets rave reviews. It does, however, require different OTs.

Here's what I did with one that allows the use of the stock OTs and is optimized for 8 ohm speakers.

https://www.audiokarma.org/forums/i...0-or-8600-conversion-to-12bh7-outputs.824135/

Not sure of the power output but I think it's probably similar to Dave's mod, maybe a bit less. I don't have a scope so no measurements either, but my ears like it a lot.

 

[Tubeglowpio]

Thanks Charlie, that was a nice read. I'm going to save that schematic for a rainy day build, looks like fun. I'm surprised you don't have a scope yet, worth the investment, I find stuff wrong with amps that I never would have known just by listening. This old build of mine for example has pretty bad ringing that I'm going to have to clean up.

 

[nj pheonix]

If you have oscillation that can be eating up your power right there.

 

[Tubeglowpio]

If you have oscillation that can be eating up your power right there.

I was thinking about that but need to understand something a little bit better first so I can rule it out for sure. So one thing is I see is that I get a pretty good waveform right before the output transformers and I lastly put it in triode mode, actually gained about 10 mV rms out and adding or removing feedback resistor didn't seem to change anything.

 

[nj pheonix]

I'm clearly not the expert here..
I'm pretty sure you should be losing power going from pentode to triode.
Also if you're going into oscillation, you need to tame that before you go further.
Maybe compare what you have to the original schematic to see feedback arrangement and values. Perhaps not ideal stock but functional starting point

 

[Tubeglowpio]

I'm clearly not the expert here..
I'm pretty sure you should be losing power going from pentode to triode.
Also if you're going into oscillation, you need to tame that before you go further.
Maybe compare what you have to the original schematic to see feedback arrangement and values. Perhaps not ideal stock but functional starting point

Yes, power drops about half normally when going to triode because the grid is tied to the plate. How do you verify that its going into oscillation if the waveform looks normal before the output and square wave ring is dampened?

 

[mashaffer]

Could one temporarily put a snubber across the OPT primary to limit open loop high frequency response and see if it clears it up?

 

[dcgillespie]

One way to test if it is oscillating is to replace the resistive load normally used for testing with an actual speaker. The load resistor is a super easy load for the amplifier to operate into, and being purely resistive, is least likely to cause the amplifier to act up. No doubt this is the type of load you have been doing your testing with. A speaker however is an impedance, containing not only a resistive element, but also capacitive and inductive elements, which are much more likely to trigger a marginally stable amplifier into instability (particularly the capacitive element). If your waveform presentations remain stable with the amplifier operating into no load and an 8Ω speaker load, then instability is likely not the cause of your problems. Note however that maximum stability is achieved with the rated (4Ω) load attached to the amplifier. As I recall, the stock design was not particularly stable at all. Use of a higher that rated load and/or long speaker leads could trigger the design into unstable operation.

Dave

 

[Tubeglowpio]

One way to test if it is oscillating is to replace the resistive load normally used for testing with an actual speaker. The load resistor is a super easy load for the amplifier to operate into, and being purely resistive, is least likely to cause the amplifier to act up. No doubt this is the type of load you have been doing your testing with. A speaker however is an impedance, containing not only a resistive element, but also capacitive and inductive elements, which are much more likely to trigger a marginally stable amplifier into instability (particularly the capacitive element). If your waveform presentations remain stable with the amplifier operating into no load and an 8Ω speaker load, then instability is likely not the cause of your problems. Note however that maximum stability is achieved with the rated (4Ω) load attached to the amplifier. As I recall, the stock design was not particularly stable at all. Use of a higher that rated load and/or long speaker leads could trigger the design into unstable operation.

Dave

So far I tested it mostly with a non-inductive 8 ohm dummy load. Doing some calculations though it would seem I should have much more than a 50 Vrms sine wave present on the primary side of the output correct? I will try adding some small value capacitors with the dummy load and see how it reacts. I knew the sound was a little off but it sure sounded like I was getting at least a couple of watts of power out vs the 100mW or so of clean power measured. I think what surprised me the most so far is how little the feedback effected the output I was expecting much more gain and instability with it removed but looks basically the same and can still crank up the audio when playing music and hear only a little bit of noticeable distortion.

 

[dcgillespie]

As I recall, I was able to achieve about 1.3 watts RMS @ 1kHz into an 8Ω load at the onset of clipping with the stock design. It would seem that you are well off this mark, so the next check would be for component values -- since you've already done your basic voltage checks. For example, if the cathode bypass cap has opened, or the cathode resistor has gone way high in value, then the voltage readings that appear to be generally good, may in fact be will over biasing the output stage. One clue is how does the amplifier clip? Does it do so evenly (doubtful), or well lopsided?

Dave