Ampeg SVT-5 Pro distortion

[bnwitt]

I am about to throw an Ampeg SVT-5 Pro up on the bench with symptoms of crackling distortion in the output through either preamp channel. This is the model with the single 12AX7 dual triode in the preamp section. I'm going to inject a signal and use the tracer to see which stage the distortion is originating in. If anyone has any experience with this issue in this amp model please feel free to share your experience.
thanks,
Barry

[Enzo]

Did we already determine it was in the preamp rather than the power amp?

[bnwitt]

Enzo,
the simple answer to your question is no. The best laid plans of mice and men oft go astray. The amp is still sitting on the bench. I had an unexpected visitor today who basically shut me down on that service job. I am going to get back on it in the morning. Since its in both channels, it has to be in the section prior to their input or post preamp in the power section as you say.

[bnwitt]

Lordy

Ok I finally got this think up on the bench and this is what I found. I have good clean signal coming out of the effects sends, the tuner output, and the preamp outputs. However, the speaker outputs have a nasty distortion when checked with the signal tracer. I tried to plug in a test speaker and that didn't go well. Made a heck of a pop for the split second I had the plug in the jack. I assume I have a power section problem at this point. I guess the next step is to pull the power transistors and test them. Any suggestions?

[Enzo]

Well, don't guess. Set your voltmeter to the speaker output and see if there is DC there.

Inject a clean signal at the power amp in jack or FX return. Follow it down to the finals.

This has two power amps, right? Set for stereo operation, right? Full range, not biamp?

This affects both channels the same? Start looking for common factors. like power supply. How does the power look for the ICs on those power amp chanels? The driver stages also run with the help of the +/-15v rails, so again they are important to verify.

[bnwitt]

Ok, here is what I have found. I have abour 13 millivolts on either amp A's or amp B's speaker outputs, but that is only until the cooling fan kicks in and then I have 69 volts on channel B and 0.00 volts on channel A. This only lasts as long as the fan is running which is about 3 seconds and then both channels' speaker outs return to 13 millivolts when the fan cycles off. This is with the fan going from the off state to on.

The signal level at the preamp outs, effects sends and tuner out, drops dramatically during the fan cycle as well. This is with the combine switch off.

This unit is very hard to work on. Everything is packed in the case and you have to remove pcb's to get to test points. None of the test points are identified on the pcb's either. Lots of back and forth schematic reading because of that.

[Mars Amp Repair]

Yes, very difficult to work on. I just got finished fighting with an SVT6Pro, so all this is pretty fresh in my mind.

I've kind of gotten to a place of zen with these Ampeg 'in the can amps' and they still give me major headaches at times.

Honestly your best bet is to just get over taking the pcb out of the unit & just do it. you can perch the pcb on it's side with a board underneath to support it on the case of the amp & even run it that way. Watch out for losing all those stupid fiber washers on the 1/4" jacks ;-]

This SVT5pro has one power amp section that handles the + of the sine & one entire separate section that handles the - sine. The negative feeds the sleeve of the jack & the + feeds the tip. You gotta be careful not to ground the sleeve of the jack with say, your scope groundlead.

One nice thing is if you don't connect a load, you are basically troubleshooting the 2 sections separately.

First of all my suggestion is to use a Variac & bulb current limiter for initial fire up. Even if you bring the unit up on a variac at low ac voltage, if there is a short the Variac w/o the bulb limiter will still try its damnedest to give you 15Amps of current. the limiter prevents that.

Unfortunately, you have to get a feel for how much ac is necessary to stablize the bias network...just experience here.

You'll need to monitor the speaker output prior to the relay or you'll never see the dc offset until the relay kicks in...if it ever does. The source resistors is a good place to do this.

Looks like you've verified that it indeed has offset, but which power amp is causing the problem? You'll need to monitor the speaker output prior to the relay on each power amp to determine which one is the culprit. You'll need to put your dvm between sleeve & ground for + amp or tip & ground for - amp. Mine was the Neg half which I belive is the one that is separate from the main pcb.

NOTE: there are 2 transistors on the main pcb that are mounted to the side of the lower heatsink that get broken away from the pcb (Q-1 & Q16 if it's like the SVT6Pro). I'm thinking that may be what caused my amp to blow in the first place. Get that main pcb out of the can & check them first.

You undoubtedly have some blown Mosfet power transistors. Since all the fets are somewhat in parallel owing to the path of the source resistors & the 47ohm gate resistor networks, you really have to lift at least one of the leads of the fets to determine which one(s) are shorted.

Usually when this occurs the little 47ohm resistors that feed the gates of those fets open, so you can start by checking those first. If you have any that are open, that's the Mosfet you should unsolder & check first. Just saves time.. Also be certain to check all the fusible 1.5K resistors that feed the limiter xistors Q108, 208, 109, 209.

NOTE: these are MATCHED mosfets & you MUST order the originals from Loud. They don't work well being mismatched & will result in the ones you replace either taking the lionshare of the current thereby blowing out again or not enough causing the originals doing the same. Note the color of the DOT on the mosfets. Loud usually uses & sends the Orange ones, but I have seen blue ones, too.

After replacing the Mosfets, be certain to run a quick meter check on all the limiter xisitors as well as the bias network of xistors. Believe me, the time you spend doing this will be nothing compared to the time you'll spend re-replacing the Mosfets when they blow again...been there-done that ;-]

When finally firing it up, monitor the idle current across one of the source resistors as well as the dc offset at the speaker output (to ground as mentioned w/o a load!).

I guess that's enough for now. They really are a pain to work on logisitically as well as electronically, however that being said, most of the time it is only the Mosfets that go bad & you fire it up once you have the shorted ones replaced & bias it & off you go. (not always, tho & then they can really suck).

glen

[bnwitt]

Glen,
thanks for the very detailed reply. At the moment I can't connect a speaker load since I am getting this fan cycle 69 VDC on the speaker output I don't want to put 4 or 8 ohms across that voltage.
I am going to just pull every thing out of the case board by board today and start checking components. as you suggest. I looked around the 3 visible sides of the main heat sync and don't see any mainboard Q's attached to it. I'm going to pull that whole heat sync/ mosfet assembly and check the components today. Thanks for the heads up on the 47 ohm gate resistors. I'll start there. On the limiter transistors, I don't see any 1.5k resistors on my schematic. My schematic shows 5.6k resistors off the base of each Q after a diode. I've attached a pdf of my power section schematics and a picture of my open chassis. I'll get back with more details after further testing.

[Mars Amp Repair]

more svt5pro

HMM,
well I am looking at the SVT5pro schemo & off of the gate (base) of each mosfet is a 1.5K fusible resistor, so I'm not sure why yours would be different. I seem to recall that some of these Ampegs I worked on have like a 2.2k resistor off the gates. Whatever they are in your amp, you just need to check them.

I didn't have time on satday to check this schemo, but you at least have basically a 'stereo' amp & not the + & - sinewave driver set up that the SVT6pro has. Makes things a bit less complicated for you...as well you have an identical amp section to make comparisons with (of course as long as one of the amps is working properly).

You've got a lot less circuitry in this version which makes it a bit easier to troublesoot.

Make certain to check all the driver related transistors on the other page of the schemo. I've had issues in the bias string as well as the constant current xistors Q's 102 & 105.

You may just get lucky & only have shorted power mosfets.

glen

[bnwitt]

Glen,
are you looking at the mosfets on page 5 of the schematics? My setup is like page 8 with 10 mosfets per channel rather than the 8 mosfets per on page 5. I'm confused why there would be two different pages like that for one amp. One type of Mosfets has a blue dot and the other type has a red dot. I assume these are the matched units you mentioned. Does Loud Technologies still have these matched sets and if so, do you have any idea how much they are per unit. I assume they are beta matched right?

[Mars Amp Repair]

Yeah,
I don't have the schemo in front of me at the moment. I didn't realize there were 2 versions.

Yes, Loud still has the prematched Mosfets available. Not in sets, but just order however many you need & they send the orange or blue dotted versions depending on which genus of mosfet you order. they cost me $6.40 each at cost...probably 2wice that for list for non-authorized folks.

BTW, I never would attempt to work on one of these amps without a variac. I'm assuming you have all that good stuff.

Once I verify I have no dc offset at the sources of the mosfets, I will monitor the output on a scope with a test speaker run through a 470ohm resistor in series and listen for signal.

The 470ohm resistor protects the speaker from being burned out if the amp decides to go into some wild offset & also keeps the load light on the amp. I always have my cheap test speakers protected with a 2amp fuse anyway.

Gotta fly....have a Merry Xmas & CCCHHAPPY CHannuka...glen

[bnwitt]

Glen,
thanks again for the info. Yes indeed I have a variac, scope etc...

[MarkusBass]

Make a google search for "mosfet matching" and you will find a lot of information about the way it can be done.

Mark

[Grant_More]

Hi Barry,

Firstly, thanks again for the schematics!

Secondly, what I've found in debug explains some of the issues you've been having here. Take a look at this:

YouTube - Ampeg SVT-5 Pro Problem

I recon two fixes are possible:

1. (Easy) increase the size of the heatsink on IC4.

2. (Hard) adjust the turn-on temperature of the fan by adjusting the control circuit.

I'm going to try 1, I'll let you know how I get on.

Cheers,
Grant

[bnwitt]

Now that's a heck of a post Grant. Thanks for sharing the fix. And with a video no less. By the way, I didn't notice an accent in your emails.