Announcement

**booj** · 08-29-2012, 09:27 AM

Gee whiz guys! Zero replies? I guess I'll just replace all the source resistors, gate resistors, and the .... whatever the 1.5k resistors are called, plus the outputs and get back with ya. Hey, what are those 1.5k resistors called?

**Enzo** · 08-29-2012, 10:22 AM

Well, not sure what to tell you. You listed a bunch of resistors, so replace them. It appears ALL your resistors are burnt. Note that your burnt resistors connect to the gate and source leads of all those output transistors. All the burn up th resistor current has to come from somewhere, and at least in the case of the 47 ohms, the ONLY place that can be is through the transistor gates. The transistors are bad. I'd be looking for S to D shorts in them, and for that matter, shorts between any of their legs. You are going to need a set of them.

Visual inspection serves one purpose - to discover obviously blown parts. But looking good means nothing - after all, the corpse at a funeral usually looks good. SO reporting nothing looking burnt tells us nothing useful.

In the case of channel A, if all those 1.5k are open, then either the channel has been repaired before and someone left those out of the effort, or there is a serious problem there now.

You are not going to like it, but what do i call the 1.5k resistors? I call them 1.5k resistors. Those 0.47 ohm ones are ballast resistors for the outputs. Those allow the parallel transistors to share current more or les equally. The 1.5k resistors sample the voltage across each resistor and feeds it back to limiting transistors Q108, Q109. That is what they do. Is there a name for that? Not that I know of. That is why we have part numbers. "R138" identifies it perfectly.

You have a majorly blown channel, most of the parts are burnt. I'd rebuild it totally. All those resistors, all those transistors. Q208, 209 are probably bad - all those burnt resistors connect right to them, and for 2 cents each, I;d replace D210,D213. D214-217 are two pairs of back to back 10v zeners. Make sure those all work, or just replace them - 5 cents each I bet. Over on the right, 3.3 ohm R269 - make sure it is not open.

Whenever a transistor burns - and by burns I don't mean it has to LOOK burnt - anything connected to it is suspect. So once all the cancer is remioved from this page, look on the other page. On the left of the power amp page are two lines to the other page called B HI and B LO. The stuff on page 1 is the driver stages.

I see 13 resistors and a trimmer. Check them all, especially the trimmer. C210 could have seen way more than its 35v rating, replce it. And I see six transistors - at least east of the op amp. ALL of them faced some stress, considering they are all cheap, and the amount of damage on page 2, I would just spend the extra dollar and replace the bunch. If I recall, those diodes are fast diodes? I;d replace them too with the exact type.

SO basically just rebuild the piss out of it.

**booj** · 08-29-2012, 11:40 AM

Thanks Enzo! Once again, you have taken me out of panic mode and back into lets do this mode. Hey, I'm getting paid for it. Your input is deeply appreciated.

**booj** · 10-02-2012, 06:03 PM

Replaced all outputs and peripheral resistors. After checking all components of the output stages for opens, shorts or damage, tried amp out. Could not bias Side B. At this point amp was drawing 1 amp at idle. Replaced the four transistors and four diodes in the bias circuit. Same results. Tried amp out with guitar signal and an 8 ohm speaker load. Side A worked ok at low level test. When I tried Side A out, even with a tiny guitar signal would cause the limit LED next to the master level control to light. The limit LED circuit ( ) is connected by two things. Limiter enable (J20 #10), and Limiter LED (J20 # 11). Limiter enable is a feed which goes into both power amps. Limit LED is the summed output of the limit circuits of both power amps. (page 1 of power amp pdf).
The circuit containing the limit defeat switch seems to be labeled wrong, with the out and in labels being on the wrong sides?
At this point I'm a little frustrated and confused. I'm probably just tired of looking at this thing and am missing something stupid simple. Any help greatly appreciated!

**MarkusBass** · 10-02-2012, 06:45 PM

I'm confused; does it mean that both power amps are not working correctly (previously you mentioned amp B only)? I suggest to focus on one power amp (I would choose amp B) and don't use it with any speaker until you know that it is fixed (unless you have money for a new speaker). Use a dummy load.
The problem with amp B seems simple since the bias current depends mainly on voltage difference between B_HI and B_LO rails. So just measure the voltages and start looking why they are incorrect. Also check whether turning the trimmer changes the voltage. I'm not sure what you replaced in the bias circuit since it consists of 2 transistors, 4 resistors and one trimmer only (and no diodes as you say). Did you mean the short circuit protection instead of the bias circuit? You can be sure that during the failure the voltages on B_HI and B_LO rails were much to high which could cause failure of Q202, 203, 204, 205 and all the resistors in the neighbourhood (especially R219 and R226). Have you check all the components? You can also start by measuring voltages in T6 and T7 test points and voltages on Q202 and Q226 (compare them with the other channel). I'm sure some of them are incorrect.
If you worry about the limiter, check voltages that cause it. For example; is there a positive voltage on the anode of the D203 diode (or simmilar)? But I would solve the problem with the bias current first.

Mark

**booj** · 10-02-2012, 07:27 PM

Thanks Mark. I guess I needed a different perspective. I've answered your questions in blue.

I'm confused; does it mean that both power amps are not working correctly (previously you mentioned amp B only)? I suggest to focus on one power amp (I would choose amp B) and don't use it with any speaker until you know that it is fixed (unless you have money for a new speaker). Use a dummy load.
Actually, I'm not sure if side A is working correctly yet or not. There is no DC on either output. I can bias side A, or at least watch the power input current rise and fall as I adjust it, so I'm assuming it at least functions to bias the side, so my main attention right now is being able to bias side B. Again, I may be jumping along with the program a bit. I have a dummy load, which I tried first, but I wanted to see what it sounded like through a speaker to try to glean a clue.
The problem with amp B seems simple since the bias current depends mainly on voltage difference between B_HI and B_LO rails. So just measure the voltages (Will do) and start looking why they are incorrect. Also check whether turning the trimmer changes the voltage. (All I know is all the parts are new or check good) I'm not sure what you replaced in the bias circuit (Everything except the resistors, which checked good) since it consists of 2 transistors, 4 resistors and one trimmer only (and no diodes as you say). (I was including, and replaced, Q206, 207, 203, 204, 205, D206, 207, 208, and 209. R225 was burned when it came to me and replaced. I also replaced R 226). Did you mean the short circuit protection instead of the bias circuit? I don�t think so. Got any ideas on those lines? You can be sure that during the failure the voltages on B_HI and B_LO rails were much to high which could cause failure of Q202, 203, 204, 205 and all the resistors in the neighbourhood (especially R219 and R226). (I replaced those transistors, and R226. R 219 checks good) Have you check all the components ? (Over and over and over) (You can also start by measuring voltages in T6 and T7 test points and voltages on Q202 and Q226 (compare them with the other channel). I'm sure some of them are incorrect. (Definitely a can do / will do)
If you worry about the limiter, check voltages that cause it. For example; is there a positive voltage on the anode of the D203 diode (or simmilar)? (I�ll see) But I would solve the problem with the bias current first.

**MarkusBass** · 10-02-2012, 08:53 PM

This is not a good approach if you want to fix the amp. I'm telling you to measure some important voltages and your answer is: all the parts are new or check good. This doesn't mean anything. I'm sure that some voltages on new or good parts are not correct. And this should be your starting point. Otherwise, you may be overlooking just one failed component and you never find it working in this way. The correct approach is to find incorrect voltage and analyze why it is incorrect. It is preffered that this is "important" voltage (eg. bias voltage). This involves understanding how the amp works. And at the very end of this painful road you will find failed component. Of course, replacing components which you find burnt is a good approach when starting the job. But now you are at the moment when you cannot find anymore failed components. In this case you have to work differently.
Start with measuring voltages on B_HI and B_LO rails. You may compare the voltages with amp A.

Mark

**booj** · 10-03-2012, 12:12 AM

I know my approach is not working, certainly, or I wouldn't be asking and taking advice. I'm sorry if you think my answer to you is what I did originally before I posted, but I did answer in blue will do, which in my mind is affirmative to check the rails. I always expect a good wrist slapping every once in a while on this forum.

I was just going to print your advice to do the work but I decided to see if anyone had posted anything additional before I went to work. I'm not the best tech there is but I'm sure the only one with this amp on my bench.

**booj** · 10-03-2012, 02:49 AM

A clue! IC1, an NE5532 reads 3.5 volts on pin 7. Pin 7 goes to R211, a 1k resistor. the other side of R11 reads 1.2 volts. 5 and 6 read zero. I don't have any Ne5532's. Tomorrow is another day.

**MarkusBass** · 10-03-2012, 07:45 AM

This is possible but I suggest that you measure voltages in the inputs of NE5532. This is how opamps work; the voltage on the output is a result of voltage difference on the inputs. And in this case the input in connected to OUTPUT_B. So incorrect voltages on the output of the amp can cause incorrect voltage on the output of the opamp. Do you have any DC on the output? It is still possible that the opamp failed but I don't think that this would cause problems with the bias. I would stick to measurements of the voltages on OUTPUT_B, B_LO and B_HI. These are the basic measurements than you should perform. And I think that it will take no more than 30 seconds. At least it is worth doing before replacing the opamp.

Mark

**booj** · 10-06-2012, 02:59 AM

Ok. It's working. Q209 was shorted. Missed it before. I did change IC1 but got the same readings, so I figured I'd look at Q208 and Q209. I thought I'd looked at them before but apparently not. Changed them both.
What's the best way to bias this thing? At this point I have just used an ammeter in series with the variac and the amp. With the bias all the way down, the amp sits idle at about .8 amps. I set the bias by "feel" for about equal increased draw with a final draw of about 1 amp. No crossover distortion evident.

**Jazz P Bass** · 10-06-2012, 01:28 PM

1 amp seems (from here) to be a bit hot.
That's 110 watts while doing nothing. (idle)
I would lean more towards a cooler Class B.
50 watts maybe.
Or just enough to get rid of the crossover notch.

**booj** · 10-06-2012, 03:46 PM

I guess for now I'll just go for getting rid of any crossover notch, then give it a good burn in. I've got another one just like it here that needs similar attention that this one got. If it gets the same idle current readings I'll post it here for reference.

**Jazz P Bass** · 10-06-2012, 04:22 PM

I would like to know (out of idle curiosity) what the mains draw is when the XOver notch just disappears.
Post the value if you have a chance.

Announcement

SVTIV Pro Side B out

SVTIV Pro Side B out

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment