Peavey 115D Powered Speakers guidance needed

[Delta362]

I'm working on a Peavey 115D powered speakers that came in with symptoms of blowing the main T6.3AL fuse. Opening it up, it's been worked on by someone else before. Checking out the components, I found several issues:

R919 (510/1W) and R904 (100k/1W) both show signs of excessive heating and out of tolerance - I replaced
C107 and C108 (1000uF/16V) were both swollen - I replaced
C109/C110 (100nF/50V SMD) I replaced for good measure as they were in parallel to the bad C107 and C108

**C111 and C116 schematic calls for 22uF/25V, but are actually installed as 220uF/25V. - Looking for guidance if I should replace these with 22uF or 220uF.

**Q104 (calls for 2SC4793) was previously replaced with a TIP41C. Is this an acceptable substitute?

**Q108 (IRFB31N20DPbF) is bad. I intend to replace Q107 and Q108, but am having a hard time finding this MOSFET as it has become discontinued. The suggested replacement is IRFB4020D, but the maximum continuous current drain specs are just over half of what the 31N20D spec says. Is that a parameter that will cause an issue in this amp? Are any other circuit changes needed to use this MOSFET instead of the original?

Thanks!

[Jazz P Bass]

Rochester Electronics has the mosfet: Rochester Electronics (en-US) : Part IRFB31N20DPBF (rocelec.com)
As does Quest Components: https://www.questcomp.com/part/4/irf...tandardpricing

If either Q107 or Q108 are faulty, IU would replace the driver IC, U101.
IRS2092S | Quest Components

[Delta362]

Thanks, Jazz! I saw the Rochester one, but they have a minimum quantity of 282 pieces. I don't think I'll need quite that many. Quest has minimum of 4 and I need 2, so I'll just end up with 2 spares. That I can handle.

Thanks for the Quest options on that and the 2092S, though Quest sure is proud of the pricing on that, aren't they?

I see BDEnt has the 2SC4793 on hand so I'll get it from there.

Very much appreciated!

[Enzo]

If teh unit used to work with 220uf caps, then I would replace them with same. SOunds like a schematic typo.

[Delta362]

Thanks, Enzo! I was thinking that as well, but when I looked at the datasheet for the 2092S chip, it showed those caps at 10uF for their suggested typical application. The 220uF just seemed far off which is what threw me.

[Delta362]

I received the IRFB31N20DPbF earlier this week. When testing them on my multimeter as well as two of those $8 component testers, it's giving me readings that are not making sense to me.

When using my multimeter on diode test setting, I read open between D-S with it in the off state. When I apply the positive lead to the gate, then back to the drain, I read 2.5VDC across D-S instead of something closer to 0V. Touching the gate to source with my finger turns it off as expected.

In both of my component testers, it's telling me it's a diode between the S-D with 587mV drop, C of 2.65nF and Ir of 2nA.

It does appear to turn on and off when testing with my multimeter diode setting, but the turned on voltage drop is much higher than I was thinking. Do these readings sound like the MOSFETs are ok? Given that the driver chip IRS2092S was very pricey, I don't want to replace that again if these MOSFETs are bad and I install them.

[Delta362]

Here are photos of the tester with the MOSFET in question showing as a diode, plus one of a known working SMK830 MOSFET.

[Jazz P Bass]

There is an internal S to D diode in the mosfet package.

[Delta362]

I saw that, Jazz. I'm more concerned/confused about it testing 2.5V across D (pin 2) to S (pin 3) when turned on as opposed to something close to 0V drop. Is this too high of a voltage differential when it's in the On state?

The 0.6V diode drop makes sense from S (pin 3) to D (pin 2) with it in the off condition.

[Jazz P Bass]

I don't think that you are 'testing' the mosfet.
They don't work like a BJT.

[Delta362]

Thanks, Jazz!

This is how I was testing the MOSFET with my meter.
https://www.utm.edu/staff/leeb/mostest.htm

I also built this test circuit on my breadboard.



Each of the IRFB31N20DPbF MOSFETs do turn on and conduct when the gate reaches about 4V so I'm guessing they are ok. It's just odd that every video I've watched of testing MOSFETs show that when they turn on, the voltage drop across Drain to Source is in the millivolt range, not 2.5V range.

[Helmholtz]

Thanks, Jazz!

This is how I was testing the MOSFET with my meter.
https://www.utm.edu/staff/leeb/mostest.htm

I also built this test circuit on my breadboard.



Each of the IRFB31N20DPbF MOSFETs do turn on and conduct when the gate reaches about 4V so I'm guessing they are ok. It's just odd that every video I've watched of testing MOSFETs show that when they turn on, the voltage drop across Drain to Source is in the millivolt range, not 2.5V range.

You can calculate the DSon-voltage from the Rds-on found in the Mosfet's datasheet multiplied by the current in your test circuit. Should be very low.

I think your Mosfet is damaged.

[Delta362]

Thank you, Helmholtz! I did recheck them and checked the datasheet for the Rds. It seems with the high turn on gate voltage that the the voltage drop across D-S shown with the multimeter has them only partially turned on as I don't think the gate voltage at that point was sufficiently high enough.

When I tested them again with the circuit I drew above, and applied a large enough gate voltage (5V) to turn them completely on to get max mA, I used that mA and the current limiting resistor to calculate the voltage drop across my current limiting resistor in the drain path. It's dropping pretty much the entire battery voltage indicating that the voltage drop across D-S was very small. I also checked the D voltage when it was turned on and could see it drop to the S voltage once fully on.

It just shows me that sometimes multiple test methods are needed to confirm if something is good or bad.

Thanks again!