Macintosh Classic Analog board blowing fuse and mosfet IRFBC40

[bibilit]

All is said, usually those boards are blowing fuses because the Mosfet is bad, i replaced the mosfet (shorted) and caps and the analog board blowed the fuse immediately, destroying the new Mosfet at the same time.

So the problem is elsewhere... any idea ?

 

[360alaska]

FLYBACK Likely shorted.

 

[techknight]

Ummm... no. 

Replace the PWM IC. its a TEA or TDA something. I forget which one. 

When you lose the MOSFET, you lose the driver IC with it. just about every time. 

 

[bibilit]

TDA 4605  maybe ?

 

[techknight]

Sounds right. it connects directly to the MOSFET. 

Also if there is a resistor or diode between the MOSFET and the IC, those need checked/replaced as well. 

 

[360alaska]

Isn't this mosfet connected to the flyback transformer primary? Not saying your wrong, just asking.

 

[techknight]

No it isnt. I think the flyback has a BU408 but I cant recall. it is almost always an NPN transistor. 

its in the main power supply. Which technically could be called "Flyback" but for very different reasons, and its not the flyback your thinking of

 

[BadGoldEagle]

No it isnt. I think the flyback has a BU408 but I cant recall.

IIRC, it's a BU-406. Close enough though!

 

[360alaska]

Yeah, I guess I was going off this schematic that someone made:



I have both versions of classic analog boards and neither have a "BU406" .

 

[techknight]

huh... go figure. First time I have EVER seen a mosfet used in the primary of the LOPT circuit. But, at the same time, that circuit design doesnt have a horizontal driver transformer coupling, so its operating in voltage mode instead of current mode so I guess that would be a necessity there. 

But its an IRF640 anyways, not the same as the IRFBC40. 

I have seen the IRF640 used in car audio amplifiers, and I mean a ton of them. Still made today. 

 

[360alaska]

Oh, my bad...

 

[Ferrix97]

Usually when the mosfet blows, there are a few more components that fail with it, sometimes it's the regulator IC (but quite rare), and most of the times is a fusible resistor (can't recall value or location, possibly between the IC an the gate of the mosfet) and a polyester (blue or yellow) capacitor that's also close to the mosfet.

Will try and dig up more information later if needed.

 

[bibilit]

Will try and dig up more information later if needed.

yes i appreciate.

 

[Ferrix97]

I'm away from home, so don't have a classic AB in front of me.

Looking at my notes, the capacitor is CP23 (270pf 2000V) and the resistor is RP40 (220 Ohms Fusible). Both of them are very close to the mosfet.

You should also check for shorted diodes (mainly the two big rectifiers, but doesn't hurt to check all of them)

 

[Mc128k]

Also try to trace all components from the power plug to the transformer primary coil. I'm not an expert on this, but I suppose that if the power mosfet gets toasted every time maybe it's because it's drawing too much current, and that could be because it's not oscillating. Try to backtrace the gate driver, and debug that part of the circuit.

 

[bibilit]

Thanks,

Diodes were ok, removed from the board and tested fine. will have a look at the resistor.

TDA 4605 can be tested ?

 

[Ferrix97]

usually a bad TDA won't drive the mosfet, so it would show no life.

You can compare it with a good working board my monitoring the supply voltage (pin 6) when slowly ramping up the mains voltage with a variac. but if yours blows the mosfet, that shouldn't help.

Check the board for shorts or burnt resistors, but there's a high chance that the capacitor I mentioned earlier (even if it measures ok) is bad, I've seen that kill mosfets before.

 

[techknight]

the TDA4605 will indeed blow the MOSFET if the chip is shorted internally. So it will drive full chip VCC into the MOSFET Gate putting it into full saturation which will make the FET go KERPLOW! Which it blows the fuse instead. 

I am not too familiar with the Classic design, but I know power supplies and have rebuilt enough power supplies on various things to fill a house, and I know this to be fact... 

Usually when a mosfet fails, it fails shorted through all 3 legs. Back to Gate. the microsecond it happens, it can send 160VDC B+ back into the drive pin of the chip just before it pops the fuse or causes other things to fail. if there is a Source resistor, and it opens, it will literally drive 160VDC B+ into the Drive pin all the time... But it usually wont pop the fuse. The fact that it keeps shorting the FET and popping the fuse, tells me the circuit path is complete on the Drain and Source pins. 

Usually if there are passive components between the TDA and the Gate of the MOSFET, it will take those out. Sometimes... But not always. 

Easiest way to rule the IC out is to leave the MOSFET out, and replace the fuse. Turn on the machine, and do a voltage check at the drive pin of the IC. if its pegged at VCC then yep... Otherwise, scope it and make sure the drive signal is within tolerance. 

But experience tells me rule of thumb, when you change the MOSFET that suffered in the event of failure, you change its drive electronics too! 

Given all of this, I still cannot rule out the possibility of the oscillator not running. if the oscillator doesnt run, and the state is pulled high, it can mimic the same thing assuming the small chance the drive IC is still good. 

These days when I design electronics that need switching control, I usually try and capacitively-couple the output stage to the drive electronics with the appropriate pull-ups/pull-downs so in the event of oscillator failure or the loss of a specific regulated B supply, it doesn't blow the whole machine sky high. 

 

[bibilit]

Thanks for all this information.

So, correct me if i am wrong, TDA 4605 is receiving supply voltage at pin 6 (as Ferrix advised) and delivering output to the Mosfet gate through pin 5.