1970 Ampeg SVT Troubleshooting Help

[Trey]

I have a 1970 SVT with 6146's. Was non op when I got. I now have sound. I'm able to bias it correctly .072V using a 4 ohm dummy load. For test purposes I plugged it in my 8 ohm cab. Played my bass through it up to about 1/3 on the volume knob. Everything sounded great. As soon as I go past 1/3 volume one of the plate resistors fries before I really even notice red plating. It happened 3 times with different plate resistors. Only one time did it actually take out a tube. I haven't attempted yet to adjust the balance control. They measurements seem to be high, but are they to high? Could the problem be in the pre amp section? How can I bypass the pre amp and test the SVT power amp section with another amps preamp? Below are the voltages and current that I measured.
What direction should I take next?

New tubes (Chinese, supposedly matched power tubes)
At Idle, With tubes installed.
Bias =.072VDC
Plate Voltage (B+)=686VDC
Screen=229VDC
Plate Current= .020 Amps

[ThermionicScott]

Not to be a jerk, but didn't you already post a thread about this?

Can you turn it up all the way into the dummy load without popping the plate resistors? Maybe you could try halving their value (or just bypassing them) until you're off the ground.

The early 6146-based SVT's were problematic because the screens fried with large signals, but that doesn't sound like your problem.

- Scott

[Trey]

Yes I did post an earlier thread, but I thought I would start a new thread and post it in the vintage section instead. Yes, I can turn it all the way up with the dummy load connected, but once I turn it up introducing a signal via my bass is when the trouble occurs at about 1/3 volume. Is there any chance the pre amp section could cause this?

Thanks

[ThermionicScott]

I was suspicious of the mismatched speaker load, but first things first -- have you played your bass into the amp when it's cranked up into the dummy load?

- Scott

[Trey]

I haven't plugged my bass in with the 4 ohm dummy load only with the 8 ohm cab.

[Alex R]

Which plate resistors are we talking about?

[Trey]

R30 & R32, 10 ohm 5w. They go to V7 & V8 Plates. Upon further inspection I have a new theory. After the plate resistor fried the 1st time I noticed that one of the bias resistors R35 was cracked. I'm thinking that the tubes lost there bias once the volume/signal was increased due to the damaged bias resistor. I replaced the bias resistor with another 1 ohm but I didn't have a 1 watt at the time so for a temporary test I installed a 1 ohm 1/2 watt I had. After powering up with a dummy load I checked and readjusted the bias successfully. I then plugged in my 8 ohm cabinet and plugged my bass in, at low volume everything was fine, then again once going above 1/3 volume another plate resistor failed. I noticed again the that the bias resistor was cracked and damaged. So my theory is that the 1st time the the plate resistor fried, the original 1 watt 1 ohm bias resistor was already damaged causing loss of bias. I then installed to low of a wattage resistor, which again caused the bias resistor to fail and the plate resistor to fry. Could this be a likely possibility? Could a 1 ohm 1/2 watt bias resistor fail so quickly for being a 1/2watt underated?
Thanks for the help and patience! It's greatly appreciated

[Tom Phillips]

Trey,
The one Ohm resistor is just a bias current sense resistor. If it blows the tubes do not “loose bias” they just turns off. (Actually all three tubes on that side turn off) As long as the tubes still have about -50V on pin 5 they should not be damaged because of bias failure.

IMO what is happening is one of the power tubes is shorting out when you play the amp hard. The current in the sorted tube spikes very high and the plate resistor and the bias sense resistor are stressed and they fail before the main fuse can blow.

They way I troubleshoot this type of problem is to mark the tube with the blown plate resistor and set is aside as a suspect bad tube. Then I take out one tube from the other side of the push/pull output section to maintain balance. DO NOT try to re-adjust the bias per the Ampeg instructions to the 72mV spec. (With only two tubes installed on each side of the power amp circuit to bias set voltage should be only ~48mV. With only one tube on each side it should be ~24mV.

Note: Before you power back up make sure that all the tubes are receiving correct bias voltage, all the tube socket pins are clean and tight and that there are no other obvious circuit problems.

The amp will run fine for test purposes with a partial set of power tubes. If no more plate resistors blow then the tube that was in the socket when the plate resistor did blow is probably bad. I find that this is a common scenario with SVTs but it usually only happens when the tube set reaches end of life.

This is one guess at what's happening. Could be something else. Hopefully, if it's the power tube then it is only one infant failure and not a bad batch.


Regards,
Tom

[Trey]

Thanks,
I'll go through the procedures that you suggested tonight. I'll report back with my findings.

[Trey]

One more question. With the push/pull configuration. Looking at the schematic there is 3 power tubes on one side and 3 power tubes on the other side. So going by the schematic the pull or push tube for each circuit oppose each (are across from each other) on the schematic? V7-V4,V8-V5, V9-V6 correct? I'm still learning about tube circuits and yes safety is priority # 1. Schematic link below:
Thanks

http://www.drtube.com/schematics/ampeg/svt6146-jp.gif

[Tom Phillips]

... Looking at the schematic there is 3 power tubes on one side and 3 power tubes on the other side. So going by the schematic the pull or push tube for each circuit oppose each (are across from each other) on the schematic? V7-V4,V8-V5, V9-V6 correct? ...

That's basically correct. However, the way a Push Pull Parallel output stage is wired each "side" just looks like one stronger tube to the resulting circuit. That is, V7 is not just paired with V4. Think of it as V4, V5 & V6 together working opposite V7, V8 & V9.. As long as there are the same number of tubes installed in each group, the output stage will be in balance. If the tubes are reasonably close that is.

[Trey]

I tested with only one supposedly known good tube on each side of the push/pull. I connected the 4 ohm dummy load. I connected a DMM to monitor bias voltage and it was approx. 24 mv at idle. I plugged my bass in to provide a signal for testing. What I observed was that the bias was stable @ 24 mv up to 1/3 volume increase and once increased upward past that point the bias voltage became unstable and increases depending on how hard I dig into the strings. I observed above 100mv and it probably would have went higher but I backed off so that I wouldn't fry another plate resistor or short the tubes. I tried different tubes, moved the tubes to all positions and the same results, stable bias voltage until past 1/3 volume. I was going to measure plate current but ran out of time, but I'm sure it's going to spike up as well porportionally to the bias voltage. One other observation, the bias voltage was a little more stable on channel 2 but did increase as volume/signal increased just not as drastic.

How stable or high should the bias voltage/current be with increase of volume/signal?

What are the possible reasons for the symptoms (coupling cap, etc.)?

Thanks Guys all your help is greatly appreciated.

[Alex R]

You will read higher current when signal is being amplified, that's normal. Does the current increase when you turn up the volume without playing?

[ThermionicScott]

My iphone ate the response i was working on, but does the bias return to its previous value when you stop playing? I wonder if the new chinese tubes just aren't stable. You may need to reduce the value of the bias-feed resistors to keep the amp from misbehaving.

[Tom Phillips]

Trey,

As Alex R pointed out everything you observed is normal. It would be good for you to read up on some basic circuit theory to help you understand what the circuit is designed to do.

The voltage across the one Ohm sense resistor is proportional to the current flowing through each half of the class AB Push/Pull section of the power amp. At idle, with no signal applied, that current is the "bias". When you play the amp the current through the tube is the algebraic sum of the bias and the signal current. There is a DC component and an AC component. Without a complete set of test equipment it will be difficult to see exactly what's going on.

Also keep an open mind about "known good tubes" A part is only good up until the time it fails. That could be the moment it leaves the factory or a lifetime later. The power tubes in an SVT are under a lot of stress and many are not up to the task.

Regards,
Tom