Kevin's Hafler p-500

There is a design error in that section of the schematic...
 
The node between C114 and IC7A pin 6 and 9 should have a series resistor there instead of a wire. It is extremely bad design practice to design a stage with positive feedback that dumps into a capacitor (C114 in this case). The positive feedback is to prevent oscillation at the comparator decision point by reinforcing the decision that the comparator has just made. Dumping into a cap causes oscillation at this critical decision point.
 
I asked a question the other night during our marathon SP-12 fix and I think it just slipped by ya--- Can PS ripple be a reinforcer of oscillation in the output section??

Ok, was the series resistor overlooked or not understood by the designer??
 
It did not slip by me Lee, just old and forgot to answer you :-)

Probably in the rare case it can be a reinforcer Lee but generally oscillation starts elsewhere in the circuit and is limited to the offending non power supply section.

In the case of an active (non-open loop) power supply such as the SP12 has in it built around the LR8 and LT1085 regulators, if those are unstable, you will have a difficult time squelching the effects of them oscillating elsewhere in the circuit. It will be everywhere.

On your other question, it looks for the most part that the guy designing this Hafler amp knew what he was doing so would put it into the category of overlooked (and it never caused a problem so he did not have to go in and fix it :-)
 
The DC I measured at the outputs was .009 VDC on the good channel and .404 VDC on the bad channel. Clearly off, but not horrible.

Would the value for that resistor be 150K as the other two legs of that RC network are?
 
I checked the signal on the PC-19C board at R33 between the pre-drivers and it was the same display as I posted earlier, a small positive-only distorted sine wave. I was pretty sure that would be the case, but I did not know if that was a valid check with the ouputs in place. I have not worked with MOSFETs much.

So does this mean the problem is before the pre-drivers or the pre-drivers themselves? I suppose if D11 through D14 were shorted, it might create this problem. Likely D11 or D12 because the N-type MOSFETs have no output at all.
 
It means that either an output device has a gate to source or drain short or that there is something wrong upstream from this (which would be my hunch.)
 
Recommend you compare the bad channel signal to the good channel signal as you probe along as an effective means to troubleshoot it Mark.
 
The layout of that Hafler is not conducive to poking like the PL amps are (with the front dropped down). I feel like pulling the main board out for that channel and doing some cold checks. I wish I had another PC-19C to drop in there, maybe I will look on ebay.

An easier check would be to pull all the output MOSFETs and check the signal at R33 again. That would eliminate the possibility of shorts somewhere in the outputs.

I'll let you know what I find out.
 
OK Mark, Like I said, I do not have any real familiarity with the layout of this amp.
 
Hi Mark
The observation I made about the DC protect has nothing to do with the current problem you are chasing.
 
Oh, I knew that. But I think it is worth an experiment once I get the amp working properly. I appreciate knowing how that circuit works and it is clear you have spent some time with similar circuits.
 
LoL... YEA... WHAT LEE SAID??????????????? Not That I would ..... or have ... understood one thing yall have said for the past two days but I sure enjoy reading all this stuff....
 
Osmosis Kevin, osmosis, that's all some of us can hope for.
 
Outputs pulled off the heatsink and next I will run a signal in to the amp and check the output of the pre-drivers. Standby to standby...
 
Handle your MOSFETs with care when they are out of the circuit Mark. They are a lot more sensitive to static discharge damage than those rugged bipolars we are used to.
 
You must log in or register to reply here.
Back
Top