This amp came in all power fets blown, the turn out was I found bad railcaps (80v 5600uf) they were shaking and top bulgy I didn’t test them electronically because they already failed the physical parameters.
I changed the fets (3205) and fet driver (4424) all gate resistors were good 10ohms, the caps I got looked Chinese a bit slimmer and a little taller but only one had the value of 5698uf on my tester I gotten 10.76mF ???** I put them in thinking my meter was malfunctioning. It played on my bench the voltage across the caps was 63ish volts, played well and all fets both PS & OP was cool.
The customer brought it back with one cap blown, all PS fets blown, could poor quality railcaps pull down the power supply, or is there something I’m missing?
I changed the fets (3205) and fet driver (4424) all gate resistors were good 10ohms, the caps I got looked Chinese a bit slimmer and a little taller but only one had the value of 5698uf on my tester I gotten 10.76mF ???** I put them in thinking my meter was malfunctioning. It played on my bench the voltage across the caps was 63ish volts, played well and all fets both PS & OP was cool.
The customer brought it back with one cap blown, all PS fets blown, could poor quality railcaps pull down the power supply, or is there something I’m missing?
Attachments
The cap could have failed and shorted while blowing.
As a very crude way of determine if caps are OK, you can do a couple of things. Touching your meter probes across the cap and seeing how long it takes to charge to a certain reading, then reversing the probes or discharging the caps and repeating.
You can also connect a signal generator across the cap to see what frequency drops the output to about 1/2 of the no load voltage. Then you can compare to a known good cap. You can even roughly calculate the capacitance with the formula C=1/2PiFR (R being the output impedance of the generator).
As a very crude way of determine if caps are OK, you can do a couple of things. Touching your meter probes across the cap and seeing how long it takes to charge to a certain reading, then reversing the probes or discharging the caps and repeating.
You can also connect a signal generator across the cap to see what frequency drops the output to about 1/2 of the no load voltage. Then you can compare to a known good cap. You can even roughly calculate the capacitance with the formula C=1/2PiFR (R being the output impedance of the generator).
None of them tested short, I don’t know if it’s possible to be shorted while blowing then opened after the failure (like exploded transistors).
So if physically the caps are bulging and shaking within the package, then it may still be good?
I do not have a frequency or signal generator (well at least not yet), what determines the reading against time and wouldn’t that determination be varied from different values cap?
So if physically the caps are bulging and shaking within the package, then it may still be good?
I do not have a frequency or signal generator (well at least not yet), what determines the reading against time and wouldn’t that determination be varied from different values cap?
Capacitors are generally limited in capacitance, for a given voltage, by the size of the can. If the capacitor is not tight in the can, it's likely defective. New or not.
You have an audio source that you can connect in series with a resistor to make a generator to test capacitors. The test isn't ideal and best used to compare known good caps to questionable caps.
You have an audio source that you can connect in series with a resistor to make a generator to test capacitors. The test isn't ideal and best used to compare known good caps to questionable caps.
Is there any reason why the rail voltage would be 85.5v when I measure across the rail caps? Everything is apparently normal otherwise. I didn’t put any load (speaker) on it but there isn’t any DC voltage when measured across the terminal.
The caps are original (80v 5600uf) from a used board I’ve also changed various PWM(SG3525) and various fet driver (MC4424) and the volts remain the same.
If it is ok, does capacitors have a tolerable overshoot percentage or should I upgrade the caps to 100v?
The owner says he has the amp over two years working, just the power supply fets and fet driver was destroyed and the capacitors was bulging.
The caps are original (80v 5600uf) from a used board I’ve also changed various PWM(SG3525) and various fet driver (MC4424) and the volts remain the same.
If it is ok, does capacitors have a tolerable overshoot percentage or should I upgrade the caps to 100v?
The owner says he has the amp over two years working, just the power supply fets and fet driver was destroyed and the capacitors was bulging.
The owner has another one of this amp and when he went to check the rail caps was bulging also, his problem is that other amp would play and at higher volumes it would cut out until the volume is lowered, he never had that problem when amp was new.
**the one that’s on this forum is periodically having one bank of PS mosfets been getting hot, I’ve checked all the follow up components and they are good and all resistors are ok.
**the one that’s on this forum is periodically having one bank of PS mosfets been getting hot, I’ve checked all the follow up components and they are good and all resistors are ok.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.