Got several mtx 1501d amps in (old design) that a fellow before me tried his hand at fixing. This one in particular I got the power supply running however i cannot get to output side to oscillate and the board mounted 42n20 mosfet (its entire drain pad is soldered to the board) gets quite hot with even a short power up of about 2 to 3 seconds. What is that mosfet's purpose?
(If it is any help, I currently have one 1 42n20 in the high side output bank and 1 in the lowside output bank. Led 399 illuminates (and of course 436 does as well) but led 238 does not. The previous person also replaced diode d199 and there are char marks around that SMD component.)
Thanks in advance
Steve
(If it is any help, I currently have one 1 42n20 in the high side output bank and 1 in the lowside output bank. Led 399 illuminates (and of course 436 does as well) but led 238 does not. The previous person also replaced diode d199 and there are char marks around that SMD component.)
Thanks in advance
Steve
Don't reinstall D199 (striped end directly connected to the RCA shields). If the amp loses ground, the amp tries to ground through the RCA shields.
The FET in the center of the board drains the rail caps. It's normal for it to get hot if the power is cycled repeatedly, in quick succession.
What's the DC voltage measured directly across pins 5 and 8 of each of the two 3120 optocouplers?
The FET in the center of the board drains the rail caps. It's normal for it to get hot if the power is cycled repeatedly, in quick succession.
What's the DC voltage measured directly across pins 5 and 8 of each of the two 3120 optocouplers?
Turns out I was too chicken keep the amp on long enough for the output to begin oscillation. LOL (the amp wasn't in the heat sink so I couldn't keep it powered on for very long.)
So I clamped down in the heat sink and powered it on. Pins 5 and 8 on the optocoupler nearest the edge of the board read 19.7 volts. On the other optocoupler voltage read 18.9 volts.
There is a Drive wave on the gateleg however it looks like complete trash. I'll post back with subsequent pictures of the waveform.
So I clamped down in the heat sink and powered it on. Pins 5 and 8 on the optocoupler nearest the edge of the board read 19.7 volts. On the other optocoupler voltage read 18.9 volts.
There is a Drive wave on the gateleg however it looks like complete trash. I'll post back with subsequent pictures of the waveform.
Ok continuing to poke around at this wave form....Yellow side has that ugly saw tooth wave pattern riding along negative rail
Blue side has a gorgeous square wave.
Last edited:
Is there any DC offset at the speaker terminals?
Does it produce audio?
The waveform is dictated by what's needed to keep the output where the input is dictating (to eliminate DC offset and produce the waveform at the speaker terminals.
Does it produce audio?
The waveform is dictated by what's needed to keep the output where the input is dictating (to eliminate DC offset and produce the waveform at the speaker terminals.
Nope, no dc offset and it produces audio..... I was being too cautious. LoL
Do you know what the value of the capacitor is that spans the pins 5 and 8 on the optocouplers?
Do you know what the value of the capacitor is that spans the pins 5 and 8 on the optocouplers?
Sorry 2 more questions about mosfet replacements for this amp.
This power supply originally ran 75339 fets. I have lots of 3205s which appear to be easier to drive and have better output specifications. Is that indeed the case in your experience?
Second question is it looks like both digikey and mouser have decided that the 42n20 is obsolete. When i compare to other fets it looks like the 38n20 is pretty darn close (1amp less current capability). Would that be an adequate replacement or should i continue to look elsewhere for a better mosfet?
This power supply originally ran 75339 fets. I have lots of 3205s which appear to be easier to drive and have better output specifications. Is that indeed the case in your experience?
Second question is it looks like both digikey and mouser have decided that the 42n20 is obsolete. When i compare to other fets it looks like the 38n20 is pretty darn close (1amp less current capability). Would that be an adequate replacement or should i continue to look elsewhere for a better mosfet?
The IRF3205s appear to be harder to drive. I'd use the originals if available.
When I was seeing these, they were finicky about outputs. The originals were the only thing that would work without making mods to the amps with only one gate resistor for each output bank. The original was an SSP45N20B. The amps that had one gate resistor per output FET used the IRFB42N20D.
I think you may have to test thoroughly if no one can give a known good replacement. If your amp has one gate resistor per FET, it may be less finicky.
When I was seeing these, they were finicky about outputs. The originals were the only thing that would work without making mods to the amps with only one gate resistor for each output bank. The original was an SSP45N20B. The amps that had one gate resistor per output FET used the IRFB42N20D.
I think you may have to test thoroughly if no one can give a known good replacement. If your amp has one gate resistor per FET, it may be less finicky.
Thank you and I'll be pulling the 3205s and running the proper 75339P mosfets. The output mosfets all have their own gate resistors on this amp. 🙂
Is it the input capacitance that determines the difficulty in driving the mosfets (assuming similar RDS and gate charges?)
Is it the input capacitance that determines the difficulty in driving the mosfets (assuming similar RDS and gate charges?)
Don't quote me on this but it seems that the Ciss is more critical for the PS FETs and the Qg is more important for the output FETs. If someone knows this to be wrong, please explain.
The 3205s may work but I don't remember trying them in this amp. I think I'd find a usable output FET before pulling the 3205s. If you can't find an output FET, the PS doesn't matter.
As a side note, I've had some success increasing the value of the 100 ohm resistor that's in parallel with the diode just after the optocouplers, when trying to find a sub for the outputs. This was on the amps with only a single output gate resistor.
The 3205s may work but I don't remember trying them in this amp. I think I'd find a usable output FET before pulling the 3205s. If you can't find an output FET, the PS doesn't matter.
As a side note, I've had some success increasing the value of the 100 ohm resistor that's in parallel with the diode just after the optocouplers, when trying to find a sub for the outputs. This was on the amps with only a single output gate resistor.