Help please! 1973 Marshall Super Lead - EL34 to 6550s bias question

[Greg_L]

Don't use the limiter to bias. You need full power.

I have checked it without the limiter. Still the same.

[Enzo]

The bulb limiter is a very useful tool, but you cannot rely on voltage readings taken while on it.

[Greg_L]

The bulb limiter is a very useful tool, but you cannot rely on voltage readings taken while on it.

Noted, thanks. But I did take my readings without the limiter. I started with a variac and the limiter so nothing would blow up. Once I was confident nothing would go up in smoke, the power came straight from the wall.

[Helmholtz]

I'm gonna put it back to EL34 configuration and get readings with the EL34s.

The resistor values for the EL34 bias circuit (except the 220k grid resistors) seem to be more appropriate for your extra "cold" 6550s as well.

[Greg_L]

The resistor values for the EL34 bias circuit (except the 220k grid resistors) seem to be more appropriate for your extra "cold" 6550s as well.

I was thinking that myself....not that I know for sure, but thinking it was worth a try. I'm going to put it back to EL34 spec and try the 6550s.

[Chuck H]

Something is definitely hinky here. You've measured with and without a current limiter in place. You paralleled some resistances (to 6550 spec as per an old Unicord schematic) in the circuit but weren't specific about what the final resistances are. You said in post #24 that you're going to "put it back to EL34 configuration and get readings with the EL34s". But one of your el34's is blown. Then later you said that you're going to put it back to el34 spec and see how it runs with the 6550's. The 6550's have already exhibited unusual current characteristics and we don't know their provenance. And then there's the unusual increase in current through the amp indicated by the limiter brightness with what you've said was a small increase in bias current. And that seems odd. So it seems the limiter is going in and out of the testing parameters and some current figures for later adjustment ARE being taken with the limiter in place.

This thread is wandering all over the place. I think we all need to slow down and carefully execute some concise and understood measurements with known parameters, then allow time for responses and follow up questions before moving ahead at each step.

My guess is that one or both of the 6550's is questionable. and it's been missed in the jumpy tests. (Was the limiter there when BOTH sockets were tested? Only one of them? Are you sure? did you move ahead with anything and now aren't sure what your actual test conditions were?) And anyway, "I" also wouldn't adjust the amp for an anomalous set of tubes and call it good because I would always question those tubes. That's like adjusting your car's suspension for strangely wrong sized tires when the right tires are available.

[Greg_L]

Something is definitely hinky here. You've measured with and without a current limiter in place. You paralleled some resistances (to 6550 spec as per an old Unicord schematic) in the circuit but weren't specific about what the final resistances are. You said in post #24 that you're going to "put it back to EL34 configuration and get readings with the EL34s". But one of your el34's is blown. Then later you said that you're going to put it back to el34 spec and see how it runs with the 6550's. The 6550's have already exhibited unusual current characteristics and we don't know their provenance. And then there's the unusual increase in current through the amp indicated by the limiter brightness with what you've said was a small increase in bias current. And that seems odd. So it seems the limiter is going in and out of the testing parameters and some current figures for later adjustment ARE being taken with the limiter in place.

The final resistances are perfectly in spec with what the schematic calls for with 6550 tubes. I think I posted the figures in my first post. It is possible I've done something wrong. I am just a human novice. But my voltages elsewhere seem to be on point. It seems, right now, like these 6550s just aren't conducting well.
I can undo what I've done and it will be back to EL34 specs, true. I did have a blown EL34, but I also have a pile of spare EL34s. The amp worked fine with EL34s.
All of my actual meaningful measurements have been without the limiter in place. Amp plugged straight into the wall and on a dummy load.
True, the 6550s are questionable and might be bad. I don't have a tube tester.

This thread is wandering all over the place. I think we all need to slow down and carefully execute some concise and understood measurements with known parameters, then allow time for responses and follow up questions before moving ahead at each step.

I'm trying to be as systematic and logical as I can. I'm trying to give the best info and implement the responses the best I can. Again, I am a novice. I have a rudimentary understanding of what's going on but I can read a schematic and take readings. I have successfully built, modified, and maintained amps before. I'm no expert, but I'm not totally green either. I'm like...green-ish.

My guess is that one or both of the 6550's is questionable. and it's been missed in the jumpy tests. (Was the limiter there when BOTH sockets were tested? Only one of them? Are you sure? did you move ahead with anything and now aren't sure what your actual test conditions were?) And anyway, "I" also wouldn't adjust the amp for an anomalous set of tubes and call it good because I would always question those tubes. That's like adjusting your car's suspension for strangely wrong sized tires when the right tires are available.

I agree. I'm also questioning the tubes, for now. Sadly this is the only set of 6550s I have and honestly I'd rather just put it back to EL34s and have it work than order another set of 6550s. Also, this is a 100 watt amp. Four 6550s.

[vintagekiki]

The value of the bias voltage depends solely on the bias voltage divider (15 kOhm) according to (56 kOhm + 22 kOhm bias trim pot)
Bias feed resistors (100 - 220 kOhm) and grid leaks resistors (1.5 - 5.6 kOhm) does not participate in the formatting value of bias voltage.

Bias voltage value depend from anode supply voltage and specified working classes of service (A, AB, B) for a given value quiet anode current.

In order to avoid an accident when adjusting bias voltage (quiet current), with the bias trim pot set bias voltage to the highest -Ug1 value (quiet current min), and gradually decrease the -Ug1 to set the quiet current to the desired value.
If the bias voltage cannot be adjusted with bias trim pot 22 kOhm, increase or decrease resistor 56 kOhm.

Everything has its price.
The higher value of the quiet current gives a aggressive powerful sound but shortens tube life.
A smaller quiet current value gives a gentle soft tone but a longer tube life.

Useful links

https://music-electronics-forum.com/showthread.php?t=49473

http://www.marshallforum.com/threads/biasing-a-100-watt-superlead.54981/

[Greg_L]

The value of the bias voltage depends solely on the bias voltage divider (15 kOhm) according to (56 kOhm + 22 kOhm bias trim pot)
Bias feed resistors (100 - 220 kOhm) and grid leaks resistors (1.5 - 5.6 kOhm) does not participate in the formatting value of bias voltage.

Bias voltage value depend from anode supply voltage and specified working classes of service (A, AB, B) for a given value quiet anode current.

In order to avoid an accident when adjusting bias voltage (quiet current), with the bias trim pot set bias voltage to the highest -Ug1 value (quiet current min), and gradually decrease the -Ug1 to set the quiet current to the desired value.
If the bias voltage cannot be adjusted with bias trim pot 22 kOhm, increase or decrease resistor 56 kOhm.

Everything has its price.
The higher value of the quiet current gives a aggressive powerful sound but shortens tube life.
A smaller quiet current value gives a gentle soft tone but a longer tube life.

Useful links

https://music-electronics-forum.com/showthread.php?t=49473

http://www.marshallforum.com/threads/biasing-a-100-watt-superlead.54981/

Thanks for the links.

It seems I had a good negative bias voltage at the grids with the 6550s, but they just weren't passing current.

I removed everything I'd done and it's set to run EL34s again. The EL34s are kind of all over the map because they're a hodge podge of unmatched tubes, but they are conducting and my voltages seem very good. I'm getting bias readings from 30-50ma from tube to tube. I know 50ma is high, but that's just one tube. The other three are all around 30ma. The high current moves with the tube. My plate voltage is back up to 455vdc. Grid bias voltage is -44vdc. Screens are around 448vdc. All of this was measured with the amp plugged into the wall- no limiter. So back at EL34 spec, the amp seems fine (not counting the mismatched tubes).

Do I dare put the 6550s back in just to see what happens???? What if they work all of the sudden with everything at EL34 specs?

[vintagekiki]

Do I dare put the 6550s back in just to see what happens???? What if they work all of the sudden with everything at EL34 specs?

To start, put one tube 6550 in socket.
Previously with the bias trim pot set to the highest -Ug1 value (- 55 V min), put one 6550 and gradually decrease the -Ug1 to set the quiet current to the desired value (30 - 40 mA).


https://el34world.com/charts/Schematics/files/Marshall/Marshall_charts.pdf
Marshall Voltages

Edit 191016
http://web.archive.org/web/20110621211356/http://diyguitarist.com/Images/BiasChart-KT88.jpg
KT88/6550 Bias Chart

[Greg_L]

To start, put one tube 6550 in socket.
Previously with the bias trim pot set to the highest -Ug1 value (- 55 V min), put one 6550 and gradually decrease the -Ug1 to set the quiet current to the desired value (30 - 40 mA).

[ATTACH=CONFIG]55574[/ATTACH]
https://el34world.com/Forum/index.php?PHPSESSID=m5vc0k19qig34lvcqpnm433rt3&action=dl attach;topic=17548.0;attach=45591
Marshall Voltages

Edit 191016
http://web.archive.org/web/20110621211356/http://diyguitarist.com/Images/BiasChart-KT88.jpg
KT88/6550 Bias Chart

Wow really? Just put one tube in? That's not gonna melt the OT? This is safe?

[vintagekiki]

Wow really? Just put one tube in? That's not gonna melt the OT? This is safe?

When you set the quiet current with one tube, you can normally put another tube.
It goes without saying that the tubes are relatively matched.
With only one tube in the socket you can measure and matched other tubes (one at a time)

[Greg_L]

When you set the quiet current with one tube, you can normally put another tube.
It goes without saying that the tubes are relatively matched.
With only one tube in the socket you can measure and matched other tubes (one at a time)

What do you mean by "quiet current"? The idle bias voltage?

If I put just one tube in, won't one half of the OT not be doing anything? Isn't this bad for the OT?

[Chuck H]

What do you mean by "quiet current"? The idle bias voltage?

If I put just one tube in, won't one half of the OT not be doing anything? Isn't this bad for the OT?

Yes, quiet current would be at idle. And it's ok to have just one tube (or one bank) in for testing. Keeping in mind that with the other tubes out you will have an increase in voltage because of the lower current condition through the power supply. You might even remember reading here about testing voltages without any tubes before installing them. So yes, it's ok.

[vintagekiki]

What do you mean by "quiet current"? The idle bias voltage?

If I put just one tube in, won't one half of the OT not be doing anything? Isn't this bad for the OT?

Quiet current or calm current or idle current is defined by bias voltage.
One tube don't damage OT.
When adjusting the bias, the loudspeaker must be connected to the amplifier.