Ampeg SVT classic fault

[pontiacpete]

This came with the complaint that it seemed weak. I didn’t sound that way to me but… The tubes have the date they were installed on them 2013. I thought I’d try a new set of Sovteks and now the amp plays for about 10 minutes or a bit more and then shuts down, goes into fault mode. I didn’t play the amp with the old tubes that long so I don’t think it was doing this when it came here, I think I would have been told.
I will put the old tubes back and play it for 10 min and see. If it is one of the new tubes having a problem can I run the amp 2 power tubes at a time for that amount of time and play through it to see if I can find the bad one? Thanks

[nosaj]

Here is what i just went through with one. Nevetslab has some great posts on them also.
nosaj

[nevetslab]

I had written some threads in that post nosaj included, that does detail the turn-on/control of the relay board which powers the HV transformer as well as the Protect / Power Status LED that resides in the preamp. The protect circuit begins on the power amp PCB, sends control signals to the preamp board as well as to the AC Mains Relay board. If what's written in that post doesn't address where you're having issues, I can help pave the way. Eight year old power tubes, if it's been working steadily over these years, they're bound to be out of balance, as well as weak....their transconductance/plate current balance can be all over the map relative to their partner's in crime. When I replace power tubes, even with a matched sextet, I'll still preset the bias for the two triad of tubes (set the bias voltage seen on pin 8 of the two 12AU7s). start with maybe -50VDC each, and plug in one pair of the sextet into positions V3 & V4. Target nominal plate current 22mA (220mV across 10 ohm cathode resistors).



With a fresh matched Sextet of 6550's or KT-88's, I will preset the bias, then trim to that nominal target of 220mV across the cathode resistors, and make both bias sources the same...might be -46.6VDC, for example. With bias being the same, then recording the readings of the six tube, two at a time, using the same sockets, will give you an accurate reading of where the six tubes are. From there, I'll select the best match for the two triads of tubes. I'll plug in the six tubes in that selection at this point, then check the plate current readings (voltage across the cathode resistors shown for V1 thru V6 on the middle photo above). The GRN/RED LED's threshhold is 14.9mA and 29.6mA, or close to that. 22mA is a touch above middle, and will yuield GRN/GRN LED readings. When the plate current spread is wide, then you're trying for a best match that doesn't drift up into the RED after they've been at idle for a couple hours.



I've gotten to know where the parts are, using the silk screen guide of the power amp PCB, and translating it to the bottom view on the foil/solder side of the power amp PCB. Op Amp U2B is the comparator that determines if there is a fault condition on the Fault Detect summing bus that feeds R47, input resistor to Pin 6. Earlier versions lack a 0.1uF capacitor C38 between the output pin 7 and the inverting input pin 6. That cap is for suppressing false triggering of the protect circuit. C13 in that circuit is involved in setting the threshold of at junction of R48, R50 and C13 (-) terminal. C13 sometimes goes bad over the years and needs to be replaced (10uF/35V, low-leakage type).

[pontiacpete]

Thanks Nosaj and Nevetslab!
I’m testing this sextet 2 at a time. What do you think of these results? Did the tubes take a hit?
I adjusted the negative voltage on the pin 8s of the 12au7s to -50, then put 2 tubes in and dialed them in to 220mA and the measured pin 8s again and read -51v.
Then I installed another 2 tubes and recorded their current draw and got 218mA and 230mA. The next 2 came in at 244mA and 232mA.

[nevetslab]

So, you have 22.0mA, 22.0mA, 21.8mA, 23.0mA, 24.4mA and 23.2mA for the sextet. I'd group them with 22.0mA, 22.0mA and 21.8mA as one triad, and 23.0mA, 23.2mA and 24.4mA as the other, and trim the bias pot for averaging them out. 22.0, 22.0 and 21.8 averages out to 21.9, while the other three average out to 23.5. Lower the bias on that triad to match the other. You did get a tight grouping in the matched set you obtained.

[pontiacpete]

I looked at them with the bias probe and saw they weren’t all that close like i expected. Usually when I get tubes from new sensor they match them really close. So not knowing how do deal with these amps i just did the matching with LEDs and the owners manual method of biasing.

I arranged them as you suggested and will test and see if it shuts down again.

[nevetslab]

I don't know about your bias probes.....the Compu-Bias probes I have can't handle the 650V Plate potentials of the KT-88/6550's in the Ampeg SVT circuit. That's why I measure them in pairs on the power amp chassis, using that setup I have in the photos. At that bias level, I normally see 220W Power Consumption at 120VAC, with both LED's GRN.

[nosaj]

I looked at them with the bias probe and saw they weren’t all that close like i expected. Usually when I get tubes from new sensor they match them really close. So not knowing how do deal with these amps i just did the matching with LEDs and the owners manual method of biasing.

I arranged them as you suggested and will test and see if it shuts down again.

If it shuts down check the HV cable and solder joints between preamp and power amp. Chopstickin the connector is how I found mine,
nosaj

[nevetslab]

Both ends of those connector headers....on the Power Amp PCB as well as on the bottom side of the Preamp. Those header pins are famous for having solder joint fractures.

[pontiacpete]

My bias probe is the bias master, still works. I just measured a deluxe here. I was worried about that, though. How reckless of me.
Anyway, it hasn’t shut down, but I’m afraid that one of the tubes keeps drifting very slowly up. I found myself adjusting it down to 22mA after every few moments of playing(sorry in my previous post I now realize I was wrote 220mA when I should have written mV, whoops). V5 seems to be the one that’s doing this. And possibly V1 is doing this too. I probably need to spend a lot more time playing it and measuring to be 100% sure. But this could be why it was shutting down in the first place.

[nevetslab]

Are these new power tubes, freshly bought from a vendor, or have them been in use for some time? When I've run into this sort of thing, all Matched Sets of tubes have a code common to the set. That can allow the vendor who's doing the matching to check their inventory and see if there's another tube in that batch that is close enough to match the remaining five tubes that are holding steady without drift in Transconductance, and hence plate current increasing.

[pontiacpete]

Yes these have stickers on them Gm 6,600 Ip 55. I’ll give them a call tomorrow. Maybe I can get a couple.

[pontiacpete]

What do you think of these numbers? I got up and turned the amp on at 8am and ran it for 2 hours, and played only about 15 minutes quietly.
V3 20.2mA
V1 21.4
V2. 20.2
V5. 23
V4. 21.1
V6. 20.3

after 2 hours
V3. 20.6
V1. 22.3
V2. 21.1
V5. 23.9
V4. 21.9
V6. 21.1

Do I need to worry about bias drifting, should I call New Sensor and try a couple more tube, or is this not much to worry about?
I did change C13, but that’s I realize now is for the relay issue. Do you change the bias cap?
Do you think I should rearrange the tubes if I keep these or do some adjusting with bias pots?

[nevetslab]

I don't see any problem with this small of a change. This is typical. To be sure, and if you have the time, let it run at idle for several hours. I wouldn't expect much difference in what you've already seen. I thought maybe you had one tube that was changing from, say, 22mA to 27mA and still climbing. I've seen them change more than that too, where I did need to change that one tube out. I did check the cathode resistor just to be sure it wasn't changing.

[pontiacpete]

I ran the amp for 3 hours last night, playing it during that time for about 10 minutes. Turned it on this morning and let it idle for another 3 hours and checked all the current draw on tubes and everything was looking good. Then I put it back into the cabinet and played it for less than a minute and it goes into fault mode. Very disappointing to say the least. I guess there’s something else going on with this amp and it developed while it’s been here, I think. Any ideas which direction I should go? I had changed out C13, can’t remember if I mentioned that.