Mac classic blows MOSFET

[RodCastler]

Hi all,

Thanks to the valuable material found on these forums I have been able to *almost* liberate my mac classic.

Looking for the source of the voltage drop at both 12V and 5V rails, I began with the routinary full re-cap of the leaky ones.

Finally, after replacing the mosfet, the mosfet driver, the opto-isolator and *every* electrolytic cap, I was able to get rock-solid voltages from the power supply, and the machine booted from the HDD with a sharp and great-looking image on screen. Happy Mac!  

The problem now is that after almost a minute into the boot sequence, boom!: fuse blown and I could measure right away that the MOSFET got dead shorted.

After the event, 

- I've looked for shorts in the transformer and it all looks good.

- I Measured for shorts coming from any potential solder bridge or shorted diode, but all looks good.

- DP7 and DP5 measure OK although they are both identical (which I'm not sure if it's correct) and also differ from the models stated on some other forum threads. They may have been replaced previously although I didn't see any signs of previous intervention on the AB. I trust them even though one shows signs of overheating on the exterior.

- I've tested every cap including ceramic and boxed ones: No shorts.

- Rectifier, transformers they all measure okay.

- Mosfet measures fine when compared to some spares I have although I'll replace it along the mosfet anyways.

What should I look for? What might be frying the MOSFET?  I can't seem to find signs of overload on the circuit.

I have been unable to find a replacement for CP23, DP7 and DP5. They are my current suspects but with no real reason: They all test functional.

* RP1 surgistor launched a small high-voltage arc on one of its sides that actually burned the surface, but still works. Haven't been able to find a replacement for it either. This arc might be a sign of something wrong. Has happened only twice, right on power on.

Thank you so much in advance for any help you could provide,

My board is a 220v, but almost identical to this one:

 

[RodCastler]

I am re-testing some components. CP23 reads 560pF on the multimeter. It's a 271J, should read 270pF.  This is a new suspect.

I removed it and the oscilloscope shows a strange behavior when under a square wave, like it's open circuit. It's clearly NOT shorted. Assuming it's defective (maybe open or heavily out of spec), could this cap be the responsible of the MOSFET being fried?  I'll try to replace it with something else and will let you know.

 

[apm]

C23 is a reasonable possibility, if it is open circuit.

The MOSFET will blow for one of two reasons: excessive current leading to overheating, or excessive voltage from drain to source. I'd guess the latter is more likely in this case, though if you have an IR thermometer that lets you monitor the MOSFET from a distance, you could check the temperature. (As you probably know, the primary side of the power supply is the single most dangerous part of the Mac, including the CRT anode voltage.)

Assuming it's overvoltage, then there are a couple possibilities:

1. The rectified mains voltage could be too high. You say you have a 230V board. Dumb question: are you sure? Has JP1 been removed? I foolishly powered up a 120V classic on 230V and got the same result as you: chime, followed by fuse blown and shorted MOSFET, though it only took a couple seconds. The feedback loop in the power supply circuit will try to correct for all manner of bad input conditions, but in the process it may exceed the ratings of the MOSFET. Too-high rectified mains would explain the arcing on RP1.

The way to check this would be to measure the voltage across CP22, which should be about 300V. But again: literally the single most dangerous point in the machine.

2. There might be a fault with the components around the MOSFET and the primary side of the transformer. I doubt the problem would be on the secondary side, though it's possible. An open-circuit CP23 might do it, as this part looks to have something to do with catching voltage spikes when QP2 switches off. In the schematic I have, it's rated at 2kV which says something. But also give a close look to DP2, CP18, RP15 and RP30. They also look to catch the voltage spikes that will result from QP2 switching off. An open circuit in that part might in some ways be worse than a short. Finally, double-check DP12 and RP40, again especially for open circuits.

 

[RodCastler]

Thank you for such a comprehensive reply, apm.

Yes, I had check the voltage and it is close to 300V and everything seems correctly set-up for a 220v machine. Actually the previous state of this mac was that it could remain powered for ever but would show garbage on screen and the HDD would not spin (low voltages).

I double checked the components you suggested and CP23 would be the only one out of spec. I have not found a local store that carries a replacement. As soon as I get my hands on a new 270pF capacitor will let  you know. (I'll have to replace the mosfet and the driver as well, but I have found those).

Thanks!

 

[RodCastler]

Another Mac has been liberated!

Symptom was: Low voltages at the 5v and 12v rail. When under load, the rails went all the way down to 4v and 9.9v respectively.

Cure: Re-capped every electrolytic, replaced TDA4605, CP23, IFBC40 and CY17G (4N36 used as replacement).

Thank you so much for your help!!

 

[apm]

Great job!

 

[Ferrix97]

Was it a board with the newer yoke connector? it's the fifth one that I see with low voltages and bad TDA4605 switching IC.

 

[RodCastler]

Not sure which type of connector you're referring to but I can tell you that mine looks like this: 

http://www.nightfallcrew.com/wp-content/gallery/swapping-crt-tube-apple-macintosh-classic-se30/20150724_083127.jpg

 

[Ferrix97]

Yup, that's the new one. If you also own a Plus or SE, you'll see that they are different.

It seems that your mac did exactly the things I mentioned in this topic

https://68kmla.org/forums/index.php?/topic/28497-fixing-a-macintosh-classic-rev-2-analog-board-240v-that-shows-no-signs-of-life/

 

[RodCastler]

Similar issue, and yes: same board: 630-0560-240V.

 

[bengi3]

I resurrect this post because i had a weird situation.

I fixed an analog board for the usual video artifacts by replacing the usual group of capacitors followed by the TDA4605. The Mac has worked flawlessly for 1 hr. Then when I switched it on a day later it started blowing the fuse. So far I have replaced also the diode bridge, the CNY 17 and the IRFBC40 but again as soon as I switch it on it blows the fuse.

Any suggestion why it has worked and the day after it started blowing fuses?

 

[Ferrix97]

If you replace the main switching mosfet you might also need to replace the blue capacitor next to it, forget the value or it's name, also check for shorts on the secondary, mainly the big rectifiers

 

[bengi3]

Sad to admit, but I had it fixed by a professional... 30 bucks and another Mac liberated!

 

[techknight]

Any idea what they replaced? 

 

[bengi3]

tried to get as much info but was really reserved...

replaced the mosfet and some caps. did not say which. As soon as i have time I observe carefully the board and let you know

 

[techknight]

Ugh.... When I was a bench tech I gave the bad parts back. and gave the customer a rough idea the problem. 

 

[bengi3]

Yes he gave me back all replaced stuff.

Threfore he replaced:

- mosfet

- the “usual suspects”, i.e. TDA and CNY

- plus a cap located between the two ics (TDA and CNY) (in the picture I put the old cap, a 47 uF 25 V, on top of the new one).

This is all he did, including a “cleaning”.

Today I wll try the same on another analog that has the same symptoms. Will keep you posted.

He was honest (30 euro) but I really feel like that fisherman that after a bad fishing day, resources to the fish market...

 

[bengi3]

Another one liberated! I just replaced:

- mosfet

- TDA 4605

- CNY 17

- capacitors at CP4 and CP5

It blew fuse and mosfet, now, after those parts replaced, boots fine.

https://youtu.be/q-rUze0fQmo

 

[techknight]

the 47uf 25v capacitor is a critical one. Without it, the IC will fail.