Bias current too high on Carvin TS100

[guitarpresence]

I am wondering if 93 mA (per side) is an acceptable bias current for this amp, recommended is 50mA. (Schematic attached)
Here is the background:
- I am working on a customer's Carvin TS100. I put in EL34's in it and new tubes for the phase inverter about 6 months ago. Biased it but could not get it down to the 50mA that is recommended but is sounded clean and looked good on the scope.
- The amp just came back to me, the complaint was that audio stopped working in 1 channel and then he switched it to bridge more and shortly after no sound at all. He said he smelled something burnt.
- When I opened it up, nothing was actually burnt. When I powered it up there was indeed no output (I used an 8 load for each channel). I tested the power supply and both OT and found no problems. I did find that R76 value had gotten out of spec, so it was probably what the customer was smelling. It was about 530 and should be 350. Then proceeded to retest and the output was back and working fine (I hate intermittent issues). I replaced R76 and rebiased the amp. I got the same results as the first time I worked on it which was about 93 mA on each side for the lowest I could set the current. I'm sure if I switched the bias more switches to 6L6 more and put 5881s in it I could get the current down but Carvin only gived 1 value for all tube configurations.
Thanks in advance,
Kevin

[Helmholtz]

I am wondering if 93 mA (per side) is an acceptable bias current for this amp, recommended is 50mA. (Schematic attached)
Here is the background:
- I am working on a customer's Carvin TS100. I put in EL34's in it and new tubes for the phase inverter about 6 months ago. Biased it but could not get it down to the 50mA that is recommended but is sounded clean and looked good on the scope.
- The amp just came back to me, the complaint was that audio stopped working in 1 channel and then he switched it to bridge more and shortly after no sound at all. He said he smelled something burnt.
- When I opened it up, nothing was actually burnt. When I powered it up there was indeed no output (I used an 8 load for each channel). I tested the power supply and both OT and found no problems. I did find that R76 value had gotten out of spec, so it was probably what the customer was smelling. It was about 530 and should be 350. Then proceeded to retest and the output was back and working fine (I hate intermittent issues). I replaced R76 and rebiased the amp. I got the same results as the first time I worked on it which was about 93 mA on each side for the lowest I could set the current. I'm sure if I switched the bias more switches to 6L6 more and put 5881s in it I could get the current down but Carvin only gived 1 value for all tube configurations.
Thanks in advance,
Kevin

but is sounded clean and looked good on the scope

High idle current always sounds clean and looks good on scope. But it causes excessive plate dissipation and eats tubes. Never bias above 70%!

Check bias voltages at C9/C44. Bias voltage should be adjustable between roughly -35V and -50V.

What is actual B+ ? And what is heater voltage?

[g1]

Where did you find the idle current suggestion, and how are you measuring it?
'per side' and 'per tube' do not mean the same thing to me, especially when dealing with a stereo power amp.

[guitarpresence]

Here is what I am seeing:
B+ = 517Vdc

Heater EL34 pin 2 = 3.1 vac, pin 7 3.3 vac

Voltage from ground to C9 = -43.76Vdc, C44 = 43.79Vdc

I used the bias procedure in the end of the manual, basically read the current at the OT.

[guitarpresence]

Measured idle current per side, I can't post the manual because it's too big. 2 current meters, one per side, connected to the center tap of each OT.

[Helmholtz]

Heater EL34 pin 2 = 3.1 vac, pin 7 3.3 vac

Always measure heater voltage across tube heater pins or across the heater winding and NOT wrt ground.

Voltage from ground to C9 = -43.76Vdc, C44 = 43.79Vdc

Is that the max neg. bias you can adjust?

I used the bias procedure in the end of the manual, basically read the current at the OT.

Don't have the manual.

[g1]

Ok, I read elsewhere that they call for 50mA at the OT center tap, so 25mA per tube.
Have you checked that the 6L6/EL34 bias switch is making a difference?

[guitarpresence]

OK, so heater voltage from pin 2 to 7 is 6.4V.
Sorry I get a little confused when asked about max voltage when talking negative voltages. The C9 and C44 voltages are at the most positive direction, in other words I can adjust the voltage more negative only.
Here is the excerpt from the manual for the biasing procedure (I did the series current reading method):
Power Tube Biasing
(Note: Biasing should be left to a qualified technician due to the fact that lethal voltages near 500 volts are present inside the amplifier).
The TS100 can be set up to use either EL34’s or 5881’s (6L6GC) in each channel. If different tube types are used in each channel, the amp should only be used in STEREO mode.
Inside the amplifier, on the component side of the power tube printed circuit card, are switches to select the tube type for each channel. Above those switches are markings that indicate the different tube types. Bias potentiometers are also near those switches for fine-tuning. Each side of the amplifier must be biased with ALL four power tubes installed.

1. Remove from the printed circuit board the red wires connected from the output transformers to QC2 and QC16.
2. Two milliamp meters will be needed. Insert a milliamp meter in series with each of the red wires to the printed circuit board. Current can now be measured through both output transformers while no input signal is present. Make sure the leads don’t touch anything such as the chassis, each other or you- VERY HIGH VOLTAGES ARE PRESENT!
3. Make sure the bias select switch is in the correct position for the tube type and the meters are set to “mA” or milliamps.
4. Power up the amplifier – switch the stand-by switch on. Adjust the bias pots accordingly to obtain a 50mA reading on each of the milliamp meters. Leave the amp on for a few minutes making sure the readings don’t change. Turn the amp off, leaving the standby switch on and let the residual high voltage bleed down. Remove the milliamp meters from the series connection and re-attach the red wires directly to QC2 and QC16 again. The amp is now correctly biased. CAUTION: The power supply capacitors will remain “charged” for a period of time after the amp has been turned off with voltages near 500 volts.

(If only volt meters are available, an alternate method of biasing can be used by substituting 1, ½ watt resistors where the milliamp meters would be as described above. The voltage across each resistor should read 50mV corresponding to 50mA. The rest of the procedure outlined above is the same.)

[The Dude]

........ I can adjust the voltage more negative only.......

More negative bias voltage will give you lower bias current. Are you adjusting the pot in the right direction?

[g1]

Maybe I'm misunderstanding but you want to make the bias voltage more negative. That will reduce the idle current.
You have 93mA and you want to reduce that down to 50mA.
You measure approx. -44V bias, you want to move that toward -50V (more negative).

[guitarpresence]

After flipping the bias switches to the 6L6 position and verifying that the switches had an effect, I then I switched them back and the levels significantly improved.
Now I am able to get the bias current to -57mA for both sides, and the voltage at C9 is -47.7V and at C44 -49.05V. One thing that is confusing to me is that the voltage at C9, C44 I though should have been inverse from the current but that did not appear to be the case (maybe I did something wrong here but I don't know what).
Here is what I think I might have done: the trim pots were all the way in the CC direction. This circuit appears to balance the current between the two channels and is very interactive, when I turned the pots clickwise, the current increased even more so I did not want to push it. Since the voltage at C9, C44 could be measured in standby mode, I turned the pots more CW and then when I went back to test the current, I was able to bring in the balance between the channels in better. Honestly I'm not sure what happened, just a bonehead thing I guess. Anyway, seams like -57mA is barely within spec. What are your thoughts on all this? Also I looked back on my notes from the first time I biased this amp and I had set the bias to -56mA, so pretty much the same as it is now.... Thanks

[Helmholtz]

]After flipping the bias switches to the 6L6 position and verifying that the switches had an effect, I then I switched them back and the levels significantly improved.
Now I am able to get the bias current to -57mA for both sides, and the voltage at C9 is -47.7V and at C44 -49.05V.

The bias switch might have a contact problem. The 6L6 position should allow for somewhat more negative bias voltage than the EL34 position.

One thing that is confusing to me is that the voltage at C9, C44 I though should have been inverse from the current but that did not appear to be the case (maybe I did something wrong here but I don't know what

Now you have a little more neg. voltage and lower idle currents, just as expected. What's confusing?

Since the voltage at C9, C44 could be measured in standby mode, I turned the pots more CW and then when I went back to test the current, I was able to bring in the balance between the channels in better. Honestly I'm not sure what happened,

In standby mode you will see more neg. bias voltage. So don't rely on adjustments made in standby.

If your adjustments are not repeatable, the trim pots might have bad wiper contact.

I don't see how the amp could balance between the channels.

I think 57mA idle is a good value - provided it's stable.

[guitarpresence]

Now you have a little more neg. voltage and lower idle currents, just as expected. What's confusing?
I was confused because the voltage adjustment went in the opposite direction than I had expected.

In standby mode you will see more neg. bias voltage. So don't rely on adjustments made in standby.
The adjustments in standby mode was just an experiment, actual adjustments were not in standby mode.

If your adjustments are not repeatable, the trim pots might have bad wiper contact.
Yes the adjustments are repeatable. I sprayed some cleaner on the pots and switches also.

I don't see how the amp could balance between the channels.
I think there is a loading effect when adjusting the pots.

I think 57mA idle is a good value - provided it's stable.
Yes, stable within 2mA

[guitarpresence]

Thanks everyone for your help! It appears to be functioning correctly and in spec now.