ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Ampeg SVT 3 Pro output has DC

11/5/2005 12:01 PM
Erik Miller	Ampeg SVT 3 Pro output has DC I've got a tough one on the bench right now, an Ampeg SVT 3 Pro, which is a hybrid tube/solid state bass amp. The output has 8 IRFP240's, all of which measured dead short when it came in. Most of the .47 Ohm resistors had blown open, but I replaced the bad ones. I've replaced all of the IRFP240's and the driver transistors test OK, but when I power it up with a load on it, as soon as the relay clicks in, it blasts dc across the output and shuts down. At one point it melted down the voice coil on my 80 watt Celestion test speaker. I understand why I am one of the few techs in this area willing to work on solid state amps. Wendy's pays a better hourly rate. Any ideas? St. Louis Music usually wants money to mail a printed copy of schematics, so I don't have one.

11/5/2005 12:01 PM

Erik Miller

Ampeg SVT 3 Pro output has DC

I've got a tough one on the bench right now, an Ampeg SVT 3 Pro, which is a hybrid tube/solid state bass amp.

The output has 8 IRFP240's, all of which measured dead short when it came in. Most of the .47 Ohm resistors had blown open, but I replaced the bad ones. I've replaced all of the IRFP240's and the driver transistors test OK, but when I power it up with a load on it, as soon as the relay clicks in, it blasts dc across the output and shuts down.

At one point it melted down the voice coil on my 80 watt Celestion test speaker.

I understand why I am one of the few techs in this area willing to work on solid state amps. Wendy's pays a better hourly rate.

Any ideas? St. Louis Music usually wants money to mail a printed copy of schematics, so I don't have one.

11/6/2005 12:51 PM
R.G.	C'mon, Erik, you've been around here long enough to know that you put a lightbulb limiter in front of power amps to prevent the "Death of the Thousand Fuses". Hook up your lightbulb limiter to the thing so it doesn't do sudden death before you can probe it. And use a dummy resistor load unless you have a rotating magazine of test speakers.

11/6/2005 11:57 PM
Erik Miller	"Hook up your lightbulb limiter to the thing so it doesn't do sudden death before you can probe it. And use a dummy resistor load unless you have a rotating magazine of test speakers. " Yeah, well, what I did was bring the thing up on my variac, carefully monitoring the AC current draw. This has worked wonderfully for every other solid state amplifier in the past. However, this piece of equipment has a dandy soft start feature that provides a time delay or something before the power amp actually comes on. So it came up just fine on the variac, sat there, and then blammo. With the lightbulb limiter and connected to the dummy load, I get the cycle up/cycle down thing. The lightbulb glows very brightly for a second, then click, it's back down. I'll try your suggestions regarding bias; I've actually pulled the driver transistor out of the board and checked it on a meter, and it's supposedly good. Working on anything with the letters "SVT" in its name is a pain in the ass, no matter what the technology.

11/6/2005 11:57 PM

Erik Miller

"Hook up your lightbulb limiter to the thing so it doesn't do sudden death before you can probe it.

And use a dummy resistor load unless you have a rotating magazine of test speakers. 8-)

"

Yeah, well, what I did was bring the thing up on my variac, carefully monitoring the AC current draw. This has worked wonderfully for every other solid state amplifier in the past.

However, this piece of equipment has a dandy soft start feature that provides a time delay or something before the power amp actually comes on. So it came up just fine on the variac, sat there, and then blammo.

With the lightbulb limiter and connected to the dummy load, I get the cycle up/cycle down thing. The lightbulb glows very brightly for a second, then click, it's back down.

I'll try your suggestions regarding bias; I've actually pulled the driver transistor out of the board and checked it on a meter, and it's supposedly good.

Working on anything with the letters "SVT" in its name is a pain in the ass, no matter what the technology.

11/7/2005 9:57 AM
R.G.	I used to use a variac too, but I got frustrated with some smart amps that refused to play if their power supply was not right after X delay. If it's a typical class AB solid state, there may well be two driver transistors, one on top and bottom. Some amps use a bootstrap capacitor from the output to a split load for a single driver transistor, but that's pretty old school and rare these days. A bad bootstrap cap could give you a similar problem. You may have a fault in the soft start circuit itself. Is it easy to defeat? What is the power supply to the amp like? Bipolar? How many volts? Can you run it at half voltage through a variac, perhaps by defeating the soft start and using the light bulb? I agree that "SVT" means "Stupid Vicious Thing" by the way.

11/7/2005 9:57 AM

R.G.

I used to use a variac too, but I got frustrated with some smart amps that refused to play if their power supply was not right after X delay.

If it's a typical class AB solid state, there may well be two driver transistors, one on top and bottom.

Some amps use a bootstrap capacitor from the output to a split load for a single driver transistor, but that's pretty old school and rare these days. A bad bootstrap cap could give you a similar problem.

You may have a fault in the soft start circuit itself. Is it easy to defeat?

What is the power supply to the amp like? Bipolar? How many volts? Can you run it at half voltage through a variac, perhaps by defeating the soft start *and* using the light bulb?

I agree that "SVT" means "Stupid Vicious Thing" by the way. 8-)

11/6/2005 1:04 PM
R.G.	Now that I got that out of my system, it's nearly always a dead driver or a bias fault that re-kills solid state amps once they've been intially killed. In fact, I usually presume the driver transistors are dead until proven otherwise. It can also be a great help if you remember that the amp will actully work, albeit in a rough and ugly sounding way, with no bias at all on the output devices. So short the bias down to 0V and measure for deadness on the driver and pre-driver if any. If this is one of those amps with elaborate volt-amp area limiting circuits, you could have problems with that as well. It might be good to just assume you're going to have to test each component in the power section with a DMM. It's also worth remembering that a class AB audio amp has an output stage with a current gain of zillions and a voltage gain of just under unity. The amp will actually work into a 10K or so load resitor with no output stage at all, just the voltage gain stage.

11/6/2005 1:04 PM

R.G.

Now that I got that out of my system, it's nearly always a dead driver or a bias fault that re-kills solid state amps once they've been intially killed.

In fact, I usually presume the driver transistors are dead until proven otherwise. It can also be a great help if you remember that the amp will actully work, albeit in a rough and ugly sounding way, with no bias at all on the output devices. So short the bias down to 0V and measure for deadness on the driver and pre-driver if any. If this is one of those amps with elaborate volt-amp area limiting circuits, you could have problems with that as well. It might be good to just assume you're going to have to test each component in the power section with a DMM.

It's also worth remembering that a class AB audio amp has an output stage with a current gain of zillions and a voltage gain of just under unity. The amp will actually work into a 10K or so load resitor with no output stage at all, just the voltage gain stage.

11/7/2005 5:40 PM
Enzo	The driver transistors test OK. Famous last words. Your tester down't put the full power rail voltage on them, nor does it ask them to conduct any current. If the amp is spitting out DC, why is there a load on it at all? RUn it naked until you get rid of the DC. That will prevent burning out your load, and it will put a hell of a lot less strain on the electronics. The schematic is like 12 pages 11x17" and includes parts layout drawings and parts lists. Isn't that worth a few bucks? There is a speaker relay, so monitor the output of the thing on the amp side of the relay, that way you can see what it is doing before the relay clicks in. The .47 ohm ballast resistors are on the output bus, monitor things at one of them. Otherwise defeat the relay. it runs off the 65v rail, so you can't just ground the low side of the coil when variac-ing, But the NC contact on the relay does not appear to go to ground - it doesn't short the speaker terminals - so you can tack a short across the contacts and then the output bus will be on the speaker terminals no matter what. If this were my repair, I would run just one output transistor on each polarity. Why blow up 8 transistors when you can only blow up two? Once the problem is solved, stuff the remaining power xstrs. For that matter, there is probably good work you can do with ALL of them out. This is not a big circuit. If the outputs are shorting, make sure the ballast resistors are all good. Make sure the gate resistors are all good. I would replace the 3440 and the 5415 automatically, Q3,4. There is a 3.3v zener associated with each, check for shorts. In fact, just grab your meter and go down the board in diode test and check each and every diode on the board. it will take five whole minutes. A bunch of them are zeners. Q2 is the little MPSA06 in a TO92 leaning on the outside of the heat sink near the bias pot. IS it OK? it is the bias xstr. The small xstrs work at low currents, so set your meter on diode test and check teh junctions of all teh TO92s. Get the schematic and check the resistors for opens.

11/7/2005 5:40 PM

Enzo

The driver transistors test OK. Famous last words. Your tester down't put the full power rail voltage on them, nor does it ask them to conduct any current.

If the amp is spitting out DC, why is there a load on it at all? RUn it naked until you get rid of the DC. That will prevent burning out your load, and it will put a hell of a lot less strain on the electronics.

The schematic is like 12 pages 11x17" and includes parts layout drawings and parts lists. Isn't that worth a few bucks?

There is a speaker relay, so monitor the output of the thing on the amp side of the relay, that way you can see what it is doing before the relay clicks in. The .47 ohm ballast resistors are on the output bus, monitor things at one of them.

Otherwise defeat the relay. it runs off the 65v rail, so you can't just ground the low side of the coil when variac-ing, But the NC contact on the relay does not appear to go to ground - it doesn't short the speaker terminals - so you can tack a short across the contacts and then the output bus will be on the speaker terminals no matter what.

If this were my repair, I would run just one output transistor on each polarity. Why blow up 8 transistors when you can only blow up two? Once the problem is solved, stuff the remaining power xstrs. For that matter, there is probably good work you can do with ALL of them out.

This is not a big circuit. If the outputs are shorting, make sure the ballast resistors are all good. Make sure the gate resistors are all good. I would replace the 3440 and the 5415 automatically, Q3,4. There is a 3.3v zener associated with each, check for shorts. In fact, just grab your meter and go down the board in diode test and check each and every diode on the board. it will take five whole minutes. A bunch of them are zeners.

Q2 is the little MPSA06 in a TO92 leaning on the outside of the heat sink near the bias pot. IS it OK? it is the bias xstr.

The small xstrs work at low currents, so set your meter on diode test and check teh junctions of all teh TO92s.

Get the schematic and check the resistors for opens.

11/8/2005 12:35 AM
Erik Miller	"If the amp is spitting out DC, why is there a load on it at all?" It took me by surprise. I screwed up, no doubt, and lost a perfectly good bench speaker. I'm thinking I screwed up by taking the piece in for repair in the first place. It's not as if I'm lacking for tube amp repair clients at the moment. The only amp I've ever had to give back to a client as beyond my ability to repair was also a high-powered solid-state bass amp. "The schematic is like 12 pages 11x17" and includes parts layout drawings and parts lists. Isn't that worth a few bucks?" I'll ask the client. I just feel kind of weird saying "the manufacturer of your expensive amp wants you to pay them to provide me with information necessary to repair it." Everyone else is happy to email me a schematic. I fix the products I manufacture for free, no matter what manner of abuse the owner has delivered upon them. Thanks for the pointers. I'm sure I'm not the only one on here who thinks "thank heavens, the cavalry is here" when they see that Enzo has replied to their message!

11/8/2005 12:35 AM

Erik Miller

"If the amp is spitting out DC, why is there a load on it at all?"

It took me by surprise. I screwed up, no doubt, and lost a perfectly good bench speaker. I'm thinking I screwed up by taking the piece in for repair in the first place. It's not as if I'm lacking for tube amp repair clients at the moment. The only amp I've ever had to give back to a client as beyond my ability to repair was also a high-powered solid-state bass amp.

"The schematic is like 12 pages 11x17" and includes parts layout drawings and parts lists. Isn't that worth a few bucks?"

I'll ask the client. I just feel kind of weird saying "the manufacturer of your expensive amp wants you to pay them to provide me with information necessary to repair it." Everyone else is happy to email me a schematic. I fix the products I manufacture for free, no matter what manner of abuse the owner has delivered upon them.

Thanks for the pointers. I'm sure I'm not the only one on here who thinks "thank heavens, the cavalry is here" when they see that Enzo has replied to their message!

Page 1 of 4

Next>

Last Page>>