Threshold 400A Cascode

Jump to Latest
#26
Hi Nelson,
Thx.
Yesterday before looking to your proposal I made some quick checkts. First of all I didn't said you that I had the left chanel RAIL FUSE (4Amp) burnt. And even replaced by a same one it burnt again. Secondly I unplugged all external cables and AC power and wait enough tume for having the capacitors dischargerd. Then I removed the cables from the top of each of the 2 bigs capacitors in order to just make a quick verification of the comon parts before the sperate chanle ones. The 2 bigs capacitors are looking ok and the AC/DC weston bridge too. So the DC arrival on each board too. The left chanel board if I understood is on the right hand side (looking from the top) and is using the negative where the right chanel is using the positive of the 2 capacitors.
After this power function check. I deducted that the right hand side board is rhe issue. And looking to it (visua check only) I didn't found any specific coloration.
I am now at the stage that most likely the issue is coming from one of the 17 transitors of this board. Am I right ? And if confirmed I have to unscrew (assuming that the for each transitor removing the 2 fixing screw will be enough to isolate) each transitor to check each by each without being oblidged to desloder each. My concern even I will be able to find the dead one will be the impact to replace it. Because in this case the new one will not be paired from the same batch with the others so do I will have to change them all ?
For sure I will have to check and align the DC offset too... And clean the board a bit because I saw some old residus of resine between PCB tracks where humidity could have stay!?
Many thx Nelson for your support.
Olivier
 
#27
finding burned/dead component is trivial if you do it methodically

take lot of pictures , if you desolder some wire mark where it was , if picture doesn't cover all details

output transistors are connected in parallel in groups ; when you check any of them for shorts , you're in most cases checking entire group

if there is short , you must find one which is actually shorted , and that's easier desoldering just emiter resistor , so just base and colector are still in circuit

while there , check/renew thermal grease , check torque , always check isolation between transistor case and heatsink

in fact , that's the thing I always check first - no brain involved ,and while doing that I can look and familiarize with rest of amp
for isolating which exactly output transistor is burned
 
#28
Hi and thx I agree for the method you are proposing , I thought that isolating the collector only by unscrewing was enough to check each but seams not so I have to desolder which I do not like to be honest I like to touch as less as possible.

And dontou agree on my before most likely root cause ? Or do I missed something in checking the DC upstream power ?
 
#29
I remember when I worked for Jands Electronics (some 20 years ago) and was in charge of final testing of (up to) 2kW per stereo side PA amplifiers, what I did in case of paralleled MOSFETs, was to send a sinewave of let's say 1kHz to amp input, and then move the oscilloscope probe to the gate of each MOSFET, one at the time. The faulty MOSFET would show slightly lower amplitude and/or a bit of oscillation. I also used an Auto-Transformer and increased the mains voltage "just enough".

I never had to desolder them all... to find a faulty one. And I virtually tested / repaired / brought in-to-specs thousands of AMP modules...

I miss those times tremendously, electronics was my first love, ...."forever an ever"!
 
You must log in or register to reply here.
Top Bottom