Switched Mode Power Supply PCB Blowing Fuse

[s34n]

Hi There,

I need to say first that my electronic experience is very minimal!

I have a PCB from an arcade machine power supply, the whole Power Supply had around 4 PCBs in total and I traced the fault to this PCB. This PCB takes 120V input and (i think) outputs 5V (I measured the 120V, and the wire terminal has a small 5V sticker on it, so I assumed it's +5 -5 and two ground wires). Could somebody confirm?

Here is a video of the board (it came out better than the pictures did, view in HD mode)

When I got the machine, the fuse was already blown on this PCB and the fuse was soldered onto the PCB. I have replaced the largest capacitor on the PCB, then I added a fuse case and a new fuse and it immediately blew, so I suspect something else is wrong!


I removed the bridge rectifier and tested it out of circuit with the 2M setting on the multimeter. The reading began at roughly 3.00 and started to count towards an open circuit. I stopped testing when the multimeter hit 6.53. I placed the probe on the two tildes to test it; which were labelled on the chip itself.

The markings of the chip under the clip and plastic state:
MIP0227SY is at http://pdf.datasheetcatalog.com/datasheet/panasonic/MIP0227SY.pdf.

I tested the MIP0227SY while it was still in the board as I had already removed the bridge rectifier . I used the right two pins to test it. I used the 2M Setting on multimeter. The range went from 0.00 and slowed more around 3.49, I assumed it would carry on until it ready an open circuit. When I flipped the probes, it gave the reading as negative and went up to 0 then back up the positive numbers again.

I was thinking if I couldn't fix this, maybe there would be a way to use a standard PC Power Supply to handle the job of this board? I would probably need to confirm the job of this PCB first though!

If anybody can think of anything to try next, that would be great!

Thanks :-o

 

[kam1787]

no offense but if you have just a little knowledge/experience in electronics i would not recommend trying to repair a smps [it does not have a 'big' transformer, does it]

i suggest finding a suitable replacement

 

[betwixt]

I agree, SMPS running off mains AC are death traps unless you are very respectful of them.

However, I can tell you that the fault is almost certainly on the side to the left (fuse side) of the dividing line across the bottom of the board and although it could be almost any of the parts, the most likely culprits, in order are:

1. The switching transistor - the one under the shiny metal clip.
2. the bridge rectifier, that the black rectangle with four pins sticking out of one side. (possibly marked D1)
3. any one of the diodes, the small black cylinders with a silver band at one end.

All of these can be tested with a cheap multimeter on a low Ohms range (less than 1K is best) or if your meter has one, a 'diode test' range. You MUST remove them from the board to test them. From your description of the readings changing over time, I guess you measured them while still on the board, what you saw was other parts around that component influencing the measurement. On a low Ohms range, measuring one way then with the probes swapped over, you should see either no reading both ways or a reading in one direction but not the other. On a diode test range you should see a reading of around 0.4 to 0.7 in one direction and no reading the other way. Any of these parts showing a reading in both directions should be treated as suspect.

NEVER,NEVER,NEVER measure anything from one side of the dividing line to the other or from any point on the fuse side of it to any point electrically grounded. Although it is a small power supply, when powered up it will have voltages significantly higher than AC mains voltage on it.

Brian.

 

[s34n]

Thanks for the response.

This small PCB has been removed from the power supply, I am not handling the installation and test of this, but thanks for your concern. You didn't offend me, I think you either misunderstood or didnt understand the situation correctly.

I was told that once I removed the bridge rectifier, I could test the switching transistor without removing that from the board. I will test the diodes and get back with my results. Yes I do agree that the problem should be on the left side, I just want to de-solder as little a possible!

"On a low Ohms range, measuring one way then with the probes swapped over, you should see either no reading both ways or a reading in one direction but not the other" - for a bad reading right? I tested the bridge rectifier and the transistor and they seemed to test ok.


-----

On another note
It's very hard to source something similar as its from a very old japanese arcade machine

I really want to find out the job of this board and see if I can replace it's functionality with a standard PC power supply if you can tell easily what it does? I think it outputs +/-5v

 

[kam1787]

"On a low Ohms range, measuring one way then with the probes swapped over, you should see either no reading both ways or a reading in one direction but not the other" - for a bad reading right?

I would have said, if same reading both ways it is likely defective
if both are "zero" indicates a short
if both readings are "infinite" indicates an open circuit

BTW I can not see the video

- - - Updated - - -

It's very hard to source something similar as its from a very old japanese arcade machine
I really want to find out the job of this board and see if I can replace it's functionality ...?

While it can be arduous to determine the voltage outputs .usually you can get a good clue.

Are there output [or near the output] electrolyte capacitors? What is the WVDC? The output voltage should be [much] less.
Is there a common ground - with the same electrolyte capacitors? Probably the voltage is +-

What about the load? Are the ICs or transistors? ICs have a max voltage rating, and older classes often had a standard voltage.
Are there LEDs, or motors or ................ that would give a clue. If there is 12V motor/solenoid/relay/light somewhere in the load, then it is quite unlikely the supply voltage is 5V.

 

[s34n]

Thank you above! ^

I just wanted to add.....

Sorry for delay in the update.

I never managed to fix the power supply board, I have no idea what the issue was, I replaced a lot of components on that board and the fuse kept blowing!

I did confirm that the board has a 120v input and outputs 5v, so I used a standard PC power supply and jumped the on-button pins so it auto-starts, and just connected the 5v line from the molex cable to the end of this complete power supply.

So hurray it works! I'm not sure if anybody will ever have this same issue, but if they do, this was my very bad workaround.

Thanks for the help