Whirlpool L1373 repair attempt

[Maxwell]

Hi,

I was lead here by google and found this thread. Unfortunately the TS abandoned the project because the main IC was apparently damaged.

I am having an issue with this board as well. I've so far repaired the damage that was present on the board, however putting it back into the machine causes R020 to disintegrate instantly, so something must be shorted in the board itself.

Initially I replaced L003 (470µh choke) and R020 (22 Ohm 3w), i also had to repair a trace that was burned out between C021 and, i'm assuming TP147... it's hard to tell because the trace is hard to follow and i'm having a hard time discerning on the schematic what the trace is feeding into other than C021.

I've since also replaced the main filter caps C023 and C022, the diodes D029 and D030, D006, the MOV RV07 and the LNK304PN, as I had those components left over from a kit I bought and I figured it couldn't hurt replacing them.

Attached pictures show the repaired area, and a picture of the R020. I apologize for the crappy soldering job, my soldering station went up in smoke and i'm making do with what i could find in the only hardware shop I could find that was open on sunday. I had NO IDEA what a challenge it is trying to desolder components with crappy generic soldering iron.

Some extra information: the LNK304PN was NOT damaged in my case, but i replaced it anyway, since i had it in a kit. The only damaged components were the R020 and the L003, and the burnt trace. Everytime i plug it in the R020 goes up in smoke and the repaired trace gets hammered and charcoaled... I've been poking my probes at this thing all day trying to find a short, and I'm out of ideas.

 

[Maxwell]

I wired a switch between mains and the L1373 so i can test and immediately turn it off to preserve my dwindling stock of 22ohm resistors.

I've noticed that I'm getting rather large arcs on the spot where I've repaired that bridge, however I'm having a hard time determining what exactly is sparking to where, as the resistor burns out rather fast.

 

[(*steve*)]

Are you sure that the is not a mosfet or something that has failed short circuit?

The resistor may be there to act as a fuse.

Please post the schematic.

 

[73's de Edd]

Sir Maxwell . . . . .

If D29 and D30 are good . . . . . . . as well as the isolated -12 and -5 rectifier diodes on the other side of the
Switch Mode Transformer.
Pull LNK 304 off the board . . . . put a 60 watt incandescent lamp in series with your
AC input line that is coming in at the top left corner of the schema I placed below.
THAT will save your 22 ohm cache by its current limiting.
Power up and confirm no R020 bad news then . If B+ is then at the two filter caps . . .all is looking good.
THEN power down and reinstall LNK 34 power up . . .still lamp protected . . . . . if lamp then goes BRIGHT you
have a bad LNK304 (Complete switch mode power supply circuit in a chip) .(Bad FET internally )

Otherwise . . . . looks like further investigation of potential circuit board carbonization needs to be explored.


[MOD Note: image removed as it was dangerously incorrect]


73's de Edd

 

[davenn]

Sir Maxwell . . . . .

If D29 and D30 are good . . . . . . . as well as the isolated -12 and -5 rectifier diodes on the other side of the
Switch Mode Transformer.
Pull LNK 34 off the board . . . . put a 60 watt incandescent lamp in series with your
AC input line that is coming in at the top left corner of the schema I placed below.
THAT will save your 22 ohm cache by its current limiting.
Power up and confirm no R020 bad news then . If B+ is then at the two filter caps . . .all is looking good.
THEN power down and reinstall LNK 34 power up . . .still lamp protected . . . . . if lamp then goes BRIGHT you
have a bad LNK304 (Complete switch mode power supply circuit in a chip) .(Bad FET internally )

Otherwise . . . . looks like further investigation of potential circuit board carbonization needs to be explored.

73's de Edd

Edd .... you really sure it should be Phase to chassis ? that would be very dangerous

 

[Maxwell]

Thanks for the lightbulb tip ED, I forgot all about that one. Will try this as soon as time permits. I must point out that LNK304PN was already replaced, and produces the same result as the LNK304 that was on the board when it failed... ie this is not likely to be the issue here.

Included is the full schematic.

Also, i'm assuming the two isolated rectifier diodes are OK, default MM test seems to clear them, although D012 has low reverse resistance, only 500kOhm, however I'm assuming this is a normal reading for a SB140 Diode, as it also has very low Vf of 180mv. I did not yet replace the latter as 1) it tested ok far as i can tell and 2) I don't have these on hand, all i have are generic 1n4007 and UF4006, and i'm assuming that neither of those are a good replacement for the SB140 due to voltage drop. I do also have BYV27-150 on hand, which might be a good replacement for D013 (BYV27-200)?

Steve: Yes I thought as well the R020 is acting as a fuse.

Another question: What is the function of the circuit parallel to switch mode transformer? It includes 470pf 1Kv MKP in parallel with 150K resistor, in series with UF4006 diode?

 

[davenn]

Included is the full schematic.

have you a larger version of that ? or better still a PDF so it can be enlarged
that posted version is very difficult to read

from what I can make out, it does look like Edd got the lower power rail incorrect as I suggested


Dave

 

[73's de Edd]

Sir Maxwell

Here is the best detailed schematic info available . . .'fo free . . . . plus I ALSO put the manufacturers power I.C. data sheet, with its typical utilization / application at its bottom.

" from what I can make out, it does look like Edd got the lower power rail incorrect as I suggested "

THAT drawing was just an old one from the net that Hellas came up with back at that time, and some Frenchman made the error on his reversed - reverse engineered drawing.

We . . . and others . . . discussed it to some length back at the time of that post.

With this old of a machine make sure that the two 5V and one 12V storage reservoir caps are good.
Also . . . . . the 47 ufd unit that is tied into the IC.
Confirm that the 18Volt sampling zener is good.
When I was referring to the two rectifier diodes . . . . . I was just thinking of dead shorts on either of them.
Your diode and RC pair constitute a " snubber " circuit, functioning to limit PEAK voltage excursions being possible across the primary of the Switch Mode Transformer and the D of the internal FET of the '304.

If you lift that IC and power up . . . .in a pitch dark room . . .any B+ rectified supply voltage arcing should be readily visible on the PCB . . . .then you pop on a flashlight and pinpoint the problem area.
If being BAAAAAD . . . . as in a carbonized path . . . it might require a Dremel tool to cut out and remove that bad area.

Thassssit . . . . .

Here 'tis :





73's de Edd

 

[Maxwell]

Hi, thanks a ton for all the info provided so far. I still have not had time to do all the tests yet, but i have a theory of my own, and just wanted to run the validity of it by you guys...

Is it possible that the repaired trace is arching over to D006 and creating a short this way because of carbonization on the board. I've attached a picture showing the carbontrails. I've cut out the 470pf cap to get a better look at the damage underneath. It seems to be arching to the kathode of the UF4006?

 

[HellasTechn]

I was lead here by google and found this thread. Unfortunately the TS abandoned the project because the main IC was apparently damaged.

Yes that is correct. in my case the microcontroller was also damaged but that is a very rare occasion.

I had to replace the LNK chip, the capacitors, the coil and the Varistor. As far as i remember it is a +12V , -5V PSU right ?

It is good practice to change all the electrilytic caps. The cost is insignificant.

you could just order the parts from e-bay, replace them on the board and give it a shot !

P.S.
It is nice to see that some old post may help other people also

 

[davenn]

Is it possible that the repaired trace is arching over to D006 and creating a short this way because of carbonization on the board.

Yes, you just need to completely scrape away all the carbonised track(s) on the board

 

[HellasTechn]

Yes, you just need to completely scrape away all the carbonised track(s) on the board

True, it may also reveal an other cut on the copper.

 

[Maxwell]

Right.

I'm going to dremel the board tomorrow with a bristle brush, see if that gets rid of it. I was also thinking of cutting the repaired trace out entirely, and then repairing it with insulated copper wire, so the affected area is perfectly isolated from arching. I'm going to try that.

I've also removed the the LNK again from the board and will perform some light bulb tests tomorrow with and without LNK.

Hellas: I'm still waiting on a delivery of some new caps, i have some from donor boards, but god only knows how old they are so... I only really keep those around for the more rare parts, and in case i need something real quick and don't have anything on hand. Meanwhile I've mixed and matched some of the caps. The main filter caps should be ok for a permanent solution, the 330µf 25v cap was replaced with a 330µf 63v cap (good quality but it's kinda large). I didn't have anything to replace the 1000µf 16v caps with, so i've left those on for now.

Any way, i'll post back when i've ran some more tests.

 

[HellasTechn]

I was also thinking of cutting the repaired trace out entirely, and then repairing it with insulated copper wire, so the affected area is perfectly isolated from arching. I'm going to try that.

I dont think that is a good idea.

In your place i would just clean the board and bridge all damages traces with a small pice of coper wire.

I'm still waiting on a delivery of some new caps, i have some from donor boards, but god only knows how old they are so... I only really keep those around for the more rare parts, and in case i need something real quick and don't have anything on hand.

Good practice.

the 330µf 25v cap was replaced with a 330µf 63v cap (good quality but it's kinda large). I didn't have anything to replace the 1000µf 16v caps with, so i've left those on for now.

Good practice also.

I've also removed the the LNK again from the board and will perform some light bulb tests tomorrow with and without LNK.

If i where you i would also order a new one from e-bay. the cost is really low and chances are that the old LNK is dead.

 

[73's de Edd]

Sir Maxwell

I'm going to dremel the board tomorrow with a bristle brush

But NOT a wire/stainless/brass brush . . . .

Your best friend that I see for improving the PRESENT abysmal condition of the boards cleanliness is denatured alcohol-MEK-Acetone or Xylene to get it really-really clean.
THEN you will be able to make an even closer inspection.
The real serious breakdown path is to the very front foli path that runs along the board.
You can see other areas that they deemed it of such importance . . . ..such that they have cut out actual slots in the PCB.
Particularly at the '304's pin 5-8 across to to that just mentioned frontal foil.
Also back at R038 you can see its shorter cut out slot to that front foil run.
IF there is no other foil interfering on backside . . . . that slot might be cleaned out and lengthened further over to the to the right using a cut-off wheel chucked in the Dremel.
BTW right connection of 470 uh choke fails QC . . . . needs some solder reflow work.
If you get this PS quasi-operating in dim bulb test mode, you can then measure the 12 and 5 supplies and shunting those 1 K caps that you didn't have, to see if the supply voltage comes up any.


73's de Edd

 

[HellasTechn]

Your best friend that I see for improving the PRESENT abysmal condition of the boards cleanliness is denatured alcohol-MEK-Acetone or Xylene to get it really-really clean.

I can confirm that. i use pure Acetone (not the nail polish remover because it contains water) all the time for that reason.

 

[Maxwell]

I dont think that is a good idea.

In your place i would just clean the board and bridge all damages traces with a small pice of coper wire.

I'm worried that the repair will just keep arching over because it's no longer insulated by soldermask. However I'll try to make the slot bigger as ED suggested, and coat the repair with a clear varnish, or i could use a wiresock...

If i where you i would also order a new one from e-bay. the cost is really low and chances are that the old LNK is dead.

I got a 5 pack, so i'm botchproof ;-)

But NOT a wire/stainless/brass brush . . . .

Not to worry, i'm not entirely daft

Your best friend that I see for improving the PRESENT abysmal condition of the boards cleanliness is denatured alcohol-MEK-Acetone or Xylene to get it really-really clean.

You mean like... paint thinner? I have gallons of acetone, but i was worried that might eat away at some of the plastic stuff on the board.

BTW right connection of 470 uh choke fails QC . . . . needs some solder reflow work.

You're very generous, the board looks like a MASSACRE and half my solder joints look like they were smeared in place with a masons trowel. However, I've worked on that some more since i got a better soldering iron, and it's looking much better now, will post a new pic after i make some more improvements.

If you get this PS quasi-operating in dim bulb test mode, you can then measure the 12 and 5 supplies and shunting those 1 K caps that you didn't have, to see if the supply voltage comes up any.

Let me make sure there's no language barrier going on here... When you say "shunt" you mean you want me to short those?

 

[73's de Edd]

When you say "shunt" you mean you want me to short those?

You put the known good test capacitor in parallel with the "suspect" one.






[Mod note please learn how to quote some one ]

 

[Maxwell]

Hi,

I was not able to power the board on using a 40w incandescent lightbulb, it seems the voltage drop is too great for small power draws, however it did confirm the short was gone, and the short was definitely in the trace repair arching to the kathode of the D006.

The board powers up now and seems to work (as far as the smps circuit goes anyways), but now the washer has started sprouting F12 error... lol! I'm going to check a few things tomorrow and see what might be up. If there is nothing wrong with the machine itself i suspect the relais on the board might be damaged, i'm going to put a probe across them tomorrow to see if they are able to pass 230v. I can confirm that the 12v circuit is working because the module itself powers up and is able to trip the main relais to power on the board.

Also question: The MKP cap C053 was nigh impossible to get a hold of. I either had to wait 5 weeks or order something like 5000 units, so I used a class 1 ceramic cap 470pf 1Kv. I'm assuming the cap is mostly used as a snubber, for which the class 1 should be fine, but i'm looking for confirmation, as i don't want the cap blowing out every other week.

Here's some post op pics of the board, as clean as i'm able to get it. I was not able to clean off the gooey muck whirlpool put on there, and so it got scorched by my soldering iron and is now sort of a blackish residue. I coated the board with a clear varnish for better insulation. I also dremeled out the slot and made it larger to prevent arching at the point of the trace repair.

Also added a pic of the cap I used to replace the MKP

 

[Maxwell]

So...

I've inspected the machine thoroughly and found nothing wrong with it, furthermore the machine prompts the error almost instantly, before actually going into cycle, so I suspected that the fault is in the PCB itself. I've poked and prodded the PCB with my probes and found a strange measurement around the circuit controlled the K1 relay.

I found a SOT23 package digital transistor that read as a dead short between pin 1 and 3, where it should have been 4.7Kohm for this type (MMUN2132), after isolating it from the circuit i got the same reading. Hopefully this will fix the fault, I'll have to wait for replacement transistors to arrive as i don't have any of these types of transistors lying around.

I'm going to do some more checking around this part of the circuit tomorrow, to see if anything else is out of wack. Enclosed is a picture of the area in question with the desoldered SOT23