Panasonic SA-MAX770PU SMPSU board fault....

[Rictech]

Hello everyone. 😀

As per thread title, this unit has a serious SMPSU fault. It blows F1001, which is a 10A M205 in the phase from the mains input.
Very first thing I tried, was simply replacing the fuse, but the replacement ceramic fuse blew immediately.

Looking at the schematic for this PSU, I then removed what I thought was the next likely candidate: D1001 - the HV main bridge rectifier.
This isolated the fault(no longer blows the main fuse), but does not point me to the actual fault.
There are no devices that have exploded or otherwise show signs of having their magic smoke gotten out - all devices look visually intact.

IC1402 is intact also.

I have printed out the schematic from the service manual, and am just following through it now, but I wonder if anyone else has seen this issue, and can perhaps help me home-in on the likely cause of the problem?

It would APPEAR to be a serious short-circuit somewhere on the hot DC side of things, cos with the main bridge in place, it is instantly blowing 10A fuses @ 230v!!!!
That's around 2300W or more(due to overload surge) - I would have expected a more violent and visual destruction of one or more of the devices on this board, but nope - they all look intact.

Measuring voltages etc is basically impossible, cos the main input fuse blows the instant you connect the juice, so there simply is no time to probe anything! 😁

Does anyone have any pointers?

EDIT: Oh, BTW - this fault remains even with the PSU board out of the chassis, and just on the workbench, so it is something in the PSU board, and not deeper into the unit, such as the amplifier board etc. Thought I should mention that.

[CapLeaker]

You gotta have a short somewhere. Measure the bridge rectifier again. Also measure between the two outer pins (inner pins are AC, outer pins are one for positive and negative DC. Do another measurement with the bridge rectifier out and go were the positive and negative (outer pins) DC is. If you still got a short there, follow line 4 (positive) there is a transistor and other PWMs on that line. You may have to split the lines here and there to rule out stuff or take components out. If in doubt, take it out! Good luck!

[Rictech]

Hi. 😀

Measuring across the bridge with it out of the board across + and - pins shows no short - it measures open circuit.
Measuring across the PCB where the bridge normally sits, across + and - tracks shows dead-short(zero ohms!)

That's not good..... 😁

I'll proceed along the lines you suggest, which was pretty much what I was going to do anyway, but I posted the thread, in case someone else had seen this fault before, and might have been able to tell me what they found as the cause. I just continue tracing. I will let the thread know, if I find the problem at what it was, for anyone else reading in the future. At least it's a single-sided PCB, so component removal is really easy. (compared to double-sided with PTH etc)

[R_J]

Likely Q1103 or Q1204, (35W60C3 ?) the smps page 2 has two lines marked 4, The RED 4 is low voltage vcc line the #4 B+ line near the top of the page connects to the two mosfets via there transformers
If the mosfets are ok it may be one of 3 disc caps, C1115, C1221, C1113

[Rictech]

Excellent, thank you. I was sniffing around the area of Q1103 and Q1204 when you posted, as those are obviously high-power MOSFETS, so.... I will pull both of those out later, and see if the short has gone, if not, will do the caps next.
I love a good mystery! 😁

EDIT: I will pull one MOSFET at a time, and test to see if the short has gone. If either MOSFET is dead, I think it would be wise to replace BOTH of them at the same time, with new ones.

[Rictech]

Q1204 short-circuit across ALL THREE PINS. With Q1103 removed, short still present on PCB, with Q1204 removed, short is GONE from the PCB.
Testing the legs of Q1204 reveals the dead-short across the device. Progress! I will replace BOTH of these MOSFETS, unless someone thinks it's OK to reuse Q1103, but IMHO, that device was probably stressed and has had a service life also, so I am of the mind to replace them both, so both sides of the SMPSU driver are then new devices. Members please let me know your thoughts.

[CapLeaker]

Not so fast! Since you figured out a mosfet is shorted on all three pins, you need to inspect the whole gate drive circuit, since it got full power dumped on it! If it looks o.k, and nothing blew up or changed value on the gate drive, you should be all right.

[R_J]

Also check R1112, R1116 and R1227 hopefully they are not open

[Rictech]

@ CapLeaker - Thanks, will do. Nothing is burnt out or showing signs of damage, but then, neither was the MOSFET! Is it acceptable to power the board with both MOSFETS out, so I can measure voltages on the gate drivers etc, or would that be a big no-no and cause more damage? 🤔

@ R_J - Thanks, I will also test those resistors. 😎

According to my research, these MOSFET's are obsolete anyway, so I can't get them through any of the legimate parts houses - everyone but Mouser does not even know of the part number, and Mouser says they have no stock. The Shindengen webstore know of them(it's their product!), but they also list as unavailable and obsolete, which is not so encouraging. 😐

AliExpress and ebay list tons of them, but I am VERY aware of the whole fake-parts or bad-clones you can get there, so I would really rather not buy from them, as they could be utter rubbish and not the genuine article. Anyway, while I search for/decide on MOSFET replacements, I have plenty of time to check the other things you both mentioned - cheers. 😍

[CapLeaker]

@ CapLeaker - Thanks, will do. Nothing is burnt out or showing signs of damage, but then, neither was the MOSFET! Is it acceptable to power the board with both MOSFETS out, so I can measure voltages on the gate drivers etc, or would that be a big no-no and cause more damage? 🤔

I would just look at the components and measure them in diode mode or resistance... If something doesn't measure up, change the blown up parts and the PWM with it. Can you turn the thing on to check the gate drive without a mosfet? Yes, but I don't like doing that for various reasons.

See if you can match a mosfet up? Look for 600V 35A mosfets and compare the datasheets. Pay attention to the capacitance as well. For example 35n60. Either get the same replacement or, match something up as close as you get it to the original values. These values do not have to be exactly the same, but they need to be close-ish.

[Rictech]

Can you turn the thing on to check the gate drive without a mosfet? Yes, but I don't like doing that for various reasons.

Yes, that was my thinking also! 😛
I was a little dubious about that idea, so I thought I would ask.
You've just convinced me NOT to do that. 😁

I'll just continue to look around at the components as listed, and check that they are all within what they should be.

One thing about the schematic from the Panasonic service manual that baffles me a little, is that caps and resistors have their values stated, but the semi's don't.
The codes listed on the schematic, are nothing to do with the part number at all. I wonder why Panasonic did the schematic like that?

Both PFC IC's IC1101 and IC1202(MH2501SC and MH2511SC) master and slave IC's seem perfectly intact - no sign of any burning, cracking or other.
None of the resistors, diodes or caps associated with these chips appear to be compromised, burnt out, cracked or otherwise showing signs of damage.
Both Q1101 and Q1203 gate-drive transistors fed from the PFC chips don't show any signs of damage, cracking, burning etc.

[R_J]

Here are the data sheets, looks like the F47W60C3 may work, it handles more current, I'm sure there are others available that might work

[CapLeaker]

I would have thought LCSC would have these... but nope... I just checked.

[Rictech]

Mouser do list the F47 transistor. Not stocked, but they obviously can get it for me, so this is the way I am planning to go unless someone else links me to somewhere else.

https://nz.mouser.com/ProductDetail/...HAKdLFNQ%3D%3D

[CapLeaker]

Get a couple… just in case. Lol

[Rictech]

😁