Enzo, MarkusBass, and other SVT 4-Pro Masters

[hodgy]

I've got an SVT 4 Pro here that I'm spinning my wheels on. The short story is- blown A channel, replaced the fets and all the surrounding goodies, all that is good. It ALMOST works. But! My A_Hi side is fine, but A_Low is not. I know it's gotta be something ridiculously easy that I'm overlooking, but my brain is frying, and my theory is terrible.
TP10 is around -0.96V, should be -3.5.
Base of Q109 is about -.71, I think it should be around -.3
I've got about 4V on pin 1 of IC1, definitely way too much. Replaced IC1, and IC101, checked out just about every component in the surrounding circuit(s). Any insight will be appreciated!

[Enzo]

Q109 is a limiter, enough on its base, and it turns on and clamps the signal to the gates of the outputs. Lifting D113 eliminates it. If it has 0.7v on its base, then it is being turned on. The base drive comes from all the source ballast resistors in the row of outputs. See if one of those outputs is drawing excessively - which of those 0.47 ohm 5w resistors has excess voltage across it?

Or, since this amp had been blown, did you replace Q109? were any of those 1.5k resistors open?

[hodgy]

The 1.5k's are all 1.48k or so, and I did replace Q109. Should I measure the .47 ohmer across the resistor or to ground? We're talking millivolts here, right? Either way they're actually drawing less than the corresponding B channel, which has more discrepancies, voltage-wise.

[MarkusBass]

Enzo is right; Q109 is just a limiter that works with high AC signal. You have to find out the reason for incorrect voltage on the base of Q109. There may be several reasons for this. This is a DC coupled amp - a failure of any component on the input can cause incorrect voltages in any place in the amp. So replacing components close to Mosfets is not enough. I suggest looking at the schematic. The amp is DC coupled but e.g. IC101 is separated with C103 so replacing IC101 was not needed. Also before replacing IC1 you should make some measurements. It is connected to the output of the power amp (R112) and you could at least measure voltages on IC inputs and R111 and R112.
I would measure voltages on ballast resitors (accross), voltages on A_HI, A_LO, voltages on Q106 and Q107. Actually, you could measure voltages starting from Q102 and compare them with B amp, which is working correctly. It's a pitty that you replaced the Mosfets because it is easier to desolder Mosfets first, fix the amp without them, and only then solder them back (of course a new ones).
I usually do not replace any component if I don't have a clear indication what is wrong and I don't understand why it is wrong. Sometimes you may be lucky but often you replace components that didn't need replacement.

Mark

[Enzo]

In my schenario, I am looking for the reason there is too high a voltage at the base of Q109. It has to come from somewhere. I am suggesting one of the MOSFETs is responsible - one or more actually. You would measure voltage ACROSS each 0.47 ohm resistor. VOltage to ground is meaningless here. Ohm's law tells you that if the voltage across one is higher, so is the current through it.

[hodgy]

Here's where I'm at today- looks like my A_HI side is less healthy than I thought. the 0.47 resistors on that side run from around 2.3 to 6.3 mv across them, where on the LO side they all read 0, with the exception of R159 which is a mighty .2mv. The B side also looks a bit choppy, the hi side not so bad with 18.2 to 33.5 on one resistor, but the low side reads from 14 all the way up to 41 mv. I tweeked AP101 and 201 just a bit to see what would happen, but didn't want to get too crazy with it.
You guys have mentioned fixing the amp without the MOSFETS- can I just do the A or B side or do I need to remove all 20?

Don't worry- this is a personal amp and I more or less have unlimited time and unlimited resources at my disposal. Good training exercise, yes?

[MarkusBass]

You guys have mentioned fixing the amp without the MOSFETS- can I just do the A or B side or do I need to remove all 20?

I mentioned this but this option is for advanced users with deep understanding how the amp works. If you are asking whether you should remove all 20 MOSFETs, it seems that you don't have a basic idea about the amp. I don't know what are your soldering skills but if you are not good at it then soldering MOSFETs and desoldering them several times may result in printed circuit damage. In case of your amp it was already suggested that you have to look for the cause of incorrect voltages on Q109. Did you follow this suggestion?
I would say even more; you are looking at results of the amp failure but you should be looking at the cause of the problem. Incorrect voltages on ballast resistors and on Q109 are most probably caused by some failure in the preceeding stage.
So if you are sure that you soldered correct transistors and you soldered them correctly, I wouldn't desolder them. Incorrect voltages on ballast resistors are caused either by lack of power supply rail (less possible), or by incorrect voltages on MOSFETs' gates (more possible). So this is the first thing I would check. Please compare A_HI - A_LO with the same voltages in B amp. Check whether bias transistor is not shorted. And look at the schematic and try to understand it. Without this it will take ages until you fix the amp.

Mark

[Mark Black]

Please take a hard look with serious magnification & light at the condition of the solder joints at all heatsink-mounted components as well as (and maybe especially) at the smaller transistors (& other components) between and/or adjacent to the heatsinks. I have found problems in the past on these models at those locations and it made for a long day...