SVT-VR lost power

[boroman]

I have a pretty rare SVT-VR here, made in the USA (1st year).
I just got new set of Shuguang tubes, and NOS Mullard ECC83 which all have been tested for over 100% emission.
Amp biased, balanced and sounding strong.

It sounded awesome for few rehearsals but we did some cabinet swaps and hooked it up to a Ampeg 410HE which is 8 ohm so a little impedance mismatch here. After some time, sound was suddenly gone (with a distorted fade out) and red light came on. No sound at all. We put standby off and then on, but it will not engage to playing mode so I shut it off. Came next day, turned it on and it's OK but I feel like it's half power, deafinitely not what you expevt from 300 clean watts, and deafinitely not the clean volume what was before. Ten o-clock and it's overdriven like V4B. Not channel dependent, I tested the tubes, all get right voltages and all test OK. No hum/hiss, smokes, shocks... Any ideas what might be wrong here?

[doombass]

I would expect the power tubes to be ok if you have green lights. If you have a signal generator and a scope you can do a signal trace. The early distortion may be developed somewhere in the preamp.

[nevetslab]

Do you have heater glow on all of the tubes? You'd have to remove the preamp chassis from the cabinet to verify the five preamp tubes inside there to be sure. If you see both GRN bias LED's lit on the rear panel, that's an indicator that you have +/- 15VDC on the main PCB, AND that you have plate voltage present for the five tubes in the preamp, as well as the HT supply to run the power tubes. The BIAS LED's sense the sum of the cathode current flowing thru the power tubes, so it's also an indicator that you have Plate Voltage and reasonably healthy power tubes. The SVT-VR doesn't have the Fault Detector circuit that's in the SVT-CL, so it can't tell you when you've had screen resistor failures in the power tube section. The one IC in the preamp runs off of +12V, -15V, and it drives the power On LED (GRN if good, RED if not). 480VDC comes into that preamp, is dropped down to 315VDC, and besides feeding all the plate circuits in the tubes, also connects to the comparator circuit on IC1B to drive that indicator light. No plate supply, it will go RED. The five preamp tubes in the preamp chassis run off of +12VDC, which comes from the main power amp PCB assy.

If you have GRN bias LED's on the rear panel, see if the power amp section works....patch in signal, or even your bass into the Power Amp input jack. it will be low level, not having a preamp to drive it, but will help isolate where the problem is.

[boroman]

Do you have heater glow on all of the tubes? You'd have to remove the preamp chassis from the cabinet to verify the five preamp tubes inside there to be sure. If you see both GRN bias LED's lit on the rear panel, that's an indicator that you have +/- 15VDC on the main PCB, AND that you have plate voltage present for the five tubes in the preamp, as well as the HT supply to run the power tubes. The BIAS LED's sense the sum of the cathode current flowing thru the power tubes, so it's also an indicator that you have Plate Voltage and reasonably healthy power tubes. The SVT-VR doesn't have the Fault Detector circuit that's in the SVT-CL, so it can't tell you when you've had screen resistor failures in the power tube section. The one IC in the preamp runs off of +12V, -15V, and it drives the power On LED (GRN if good, RED if not). 480VDC comes into that preamp, is dropped down to 315VDC, and besides feeding all the plate circuits in the tubes, also connects to the comparator circuit on IC1B to drive that indicator light. No plate supply, it will go RED. The five preamp tubes in the preamp chassis run off of +12VDC, which comes from the main power amp PCB assy.

Thanks!!!

The amp turned off (to RED) couple of times to day but I saw once arc'ing in one of the tubes. Disassembled the amp and found 5W resistor open/blown on tube board. (Tube itself tests OK, so I hope it was not a shorting tube...). I also re-set J26 pin to J28 which I've read is the common mod to set the signal out of the "30s time delay relay" that causes problem with shutting off standby when voltage is just few volts below the "ideal". So I will replacing this resistor and we'll see what's going on then.

What's the purpose of BALANCE in the bias section? Tried to read schematic and opinions about it but I don't get it. You have two halves that you bias separately to get identical sum of currents, right? So then both triangles work about the same way push-pull. So why another balance pot? If I just set the bias right with a matched set of tubes, would it be OK to just set balance in the middle position?

[Jazz P Bass]

" but I saw once arc'ing in one of the tubes"
" found 5W resistor open/blown on tube board. (Tube itself tests OK, so I hope it was not a shorting tube."

Sounds like you have a bad output tube.

[boroman]

" but I saw once arc'ing in one of the tubes"
" found 5W resistor open/blown on tube board. (Tube itself tests OK, so I hope it was not a shorting tube."

Sounds like you have a bad output tube.

It tests perfectly fine on my both tube testers for emission, shorts, grid leakage etc.
It might arc because of that burnt resistor... I hope so.

[Helmholtz]

You can't be sure with an intermittent short between electrodes. Arcing inside a good tube can only produced with excessive voltage. What's the function of the burnt resistor?

Please always post a schematic.

[nevetslab]

Thanks!!!

The amp turned off (to RED) couple of times to day but I saw once arc'ing in one of the tubes. Disassembled the amp and found 5W resistor open/blown on tube board. (Tube itself tests OK, so I hope it was not a shorting tube...). I also re-set J26 pin to J28 which I've read is the common mod to set the signal out of the "30s time delay relay" that causes problem with shutting off standby when voltage is just few volts below the "ideal". So I will replacing this resistor and we'll see what's going on then.

What's the purpose of BALANCE in the bias section? Tried to read schematic and opinions about it but I don't get it. You have two halves that you bias separately to get identical sum of currents, right? So then both triangles work about the same way push-pull. So why another balance pot? If I just set the bias right with a matched set of tubes, would it be OK to just set balance in the middle position?

You didn't mention if the GRN Bias LED's were both lit. I'd expect the side having the blown plate resistor would NOT have lit. The Bias LED's have a plate current window from around 64mA to 78mA on each side. The side with that open plate resistor wouldn't have allowed the associated bias LED to light under normal conditions. The Balance LED and associated circuit is a fine-tune of the plate current balance of the upper and lower halves of the output stage. It's normally set by injecting a 40Hz Sine Wave signal into the Power Amp input (rear panel jack), and setting it to 25V RMS output into a dummy load (would be VERY LOUD if using a speaker!). The Balance pot is adjusted for YELLOW, or, if you have a scope and Distortion Analyzer, adjust for minimum distortion. Some times you have to slightly tweak the bias adjust pots to get the YELLOW LED to light up. Another less-scientific way to adjust that is to listen to the residual 'hum' on the output, and adjust that Balance for minimum hum....and hopefully it corresponds to the LED lighting.

With regards to bypassing the relay-switched AC Mains (J26 is the return AC Mains wire from the Standby Switch, input to the relay HT terminal & J28 is one of the two relay-switched output terminals that the pwr xfmr primary wires attach to). If your AC Mains is 220-240V, you could get away with this 'bypassing of the HT Power Xfmr's relay switching'. If you're at 100VAC or 120VAC, both J27 & J28 will be occupied by the BLK and BRN wires of the split primary wires of the HT power xfmr. I've never read anything about this mod, and it's potentially dangerous if you don't know what you're doing.

I've attached the Chassis wiring diagram for an SVT-CL (don't have one for the SVT-VR, but the primary wiring is the same with regards to both Heater & HT xfmr's primary wiring, along with that of the AC Mains PCB, which is common to both models. And the Power Xfmr schematics for the SVT-VR I recently drew up, since Ampeg has never provided them.)

SVT-Classic-Chas-Wiring-Schematic.pdf
AC Terminal PCB Schematics and Parts List (07P319-04_2).pdf
Ampeg SVT-VR Power & Heater Xfmrs.pdf

[Helmholtz]

Can't find the power amp schematic.
Which one is the burnt "5W" resistor?

[nevetslab]

Ampeg SVT-VR Power Tube PCB Schematic & layout drawings attached:

Power Supply PCB Schematics (792SCH_A).pdf
Tube PCB Schematics (365SCH_0).pdf
PWA Tube PCB Layout (365PWA_0).pdf
PWA 300 Watt Power Supply PCB Layout (792PWA_A).pdf

[g1]

A tube tester will not check the tubes at the kind of voltage the SVT is running at. If you saw an arcing in the tube, it is highly likely the tube is bad.
If you had another working SVT you could test the tubes in that.

[boroman]

Helmholtz Sorry I messed up... I meant 5R not 5W. These are 5 Ohm / 3W plate resistors mounted on the tube board. One of them is fried. I don't have one here, must order so will check it later this week.

nevetslab Yes it can be biased, both green light are lit, no problem with that even with 1 bad plate resistor.
The instruction for bypass the time delay relay is here https://www.talkbass.com/threads/joh...ay-mod.617347/
- I think it was modified in the later models and SVT 50th Anniversary so there's no problem playing when wall voltage drops a bit.

And back to BALANCE potentiometer, so if its the icing on a cake of a properly biased two halves, means that I can leave it in the center and not bother? I dont have 300W/4ohm resistor and true RMS meter... (I must check, maybe my Fluke is true RMS but I doubt it). I've tried to listen to the distortions/hum when its turned all the way clockwise and then all the way counterclockwise and I don't hear any hum or more distortion when I play through it.

g1 - I think I should swap those from SVT-CL... good idea, but let me test those when I solder new plate resistor in place. We'll see.

[nevetslab]

Good article in the thread you included. It actually answered a problem I hadn't considered, where one SVT-VR we rented had dropped out on Stage in a performance. I was never able to reproduce the problem in the shop, but....I hadn't dropped the AC mains down to see just where the control circuitry that feeds the Relay circuit drops out. THAT's where the adjustment needs to be made.

In the case of the SVT-CL, it has a VERY EFFECTIVE protection circuit which the SVT-VR doesn't have. If you ever have a power tube fail on you, it will shut down the HT Xfmr immediately, preventing damage, which the SVT-VR will not do. All you have there is hope that the HT supply will drop enough to get the fuse to blow. I'll have a look at the SVT-VR circuit to see what the threshold is and what's needed to keep these powered up in the range of 100VAC to 105VAC.

I would advise AGAINST making that relay bypass mod on the SVT-CL!

[boroman]

I would advise AGAINST making that relay bypass mod on the SVT-CL!

Me too. I wouldn't bypass it on CL. It has different purpose there and it's designed well.

But in VR, it;s like 20-30s time delay from powering it on, not to mention that stupid voltage protection. I've read, that when bypassed, people have set variac to 95V and still was playable (just lower headroom), so this might be the key. And as I said, I've read that threshold was brought down in later revision, so in current production VR's that problem do not exist anymore.

[nevetslab]

In looking at the power supply circuits on the SVT-VR....the delay-turn-on comes from J9 on the main power amp PCB. Specifically comes from the emitter of Q3, with the base of it getting the delay ramp voltage of C19 & R45. Those two parts set the delay. The voltage source for this comes from the zener regulator on the LV bipolar supply. The unregulated side of that has a pair of 1500uf/35V buss caps, while the negative side only has 470uf.....not much current is being pulled from the -15V circuit. The Relay circuit is being powered from a 5V Regulator IC2 (LM7805T via J1, feeding J36 on the relay board. It won't drop out at low line, but, with low line bringing down the unregulated voltages off the Heater xfmr....those 1500uF Buss caps may be an issue. It's the negative end of the triangular ripple on those caps that can cause the loss of hold-in voltage at Q3 to keep the darlington xstr on the AC Mains PCB turned on.

I'll have to wheel over an older SVT-VR, as well as the latest SVT-VR in our rental fleet to have a look at this issue. As well as with the SVT-CL. When I read the link you included in the previous post, I had egg on my face from chasing the problem found with a client having the SVT-VR drop out on stage. I never stopped to drop the mains voltage down under the battery of tests I was running. Had blinders on during that exercise, sad to say.