Motor M500 Askoll

Česky

Translated with Deepl.

7.10.2018

For those of you struggling with the same or similar problem, read to the end so you don't pick it apart unnecessarily (like I did)!

A friend of mine asked me to fix his washing machine. He bought it new, but very cheaply with a scratch from shipping. OK for a guest house. After two years, the little fuckupper kicked in and the washing machine started acting strange. First it would momentarily fill with water and then try to drain it. It never turned the drum while doing so. There was a sort of double-clicking of the relay from the motor about every 10-15 seconds. "That would be the engine", I said to myself. And it was. When I disassembled it and found that the motor was a three phase motor with its own frequency converter, I suspected there would be trouble. And there were.

The washing machine is a Blomberg WNF 7447 AE40.

Behaviour of a working engine without communication

If you connect the motor to 230VAC only, on connector pins 4, 5 and switch it on, the following should happen:

1) The relay clicks (turns on). The motor twitches. If you measure with an oscilloscope at the center of the coils (against GND) you should see something like the picture. It looks like some kind of test.

2) After approximately 10 to 15 seconds, the relay will quickly turn off and on (within about half a second). This repeats indefinitely. This looks like some kind of watchdog. If the motor does not establish communication within a certain amount of time, it will perform a reset. If communication works, the relay does not click.

WARNING! So if the motor behaves as I describe here, it does not mean it is defective at all! The mainboard may still be faulty!

Disassembling the engine

I realize somewhat late that I didn't take a picture of the whole engine :-( So here's one from the site.

https://www.huoltopalvelu.com/Beko-Grundig-washing-machine-motor-WMB/GWN

Loosen the three long screws on the white plastic cover. It is very likely that the cover will be glued on the side with grey silicone. Cut it off with a thin knife. There's no need to glue it back on. Silicone is mainly used to fix large capacitors.

Remove the retaining ring on the motor face. Remove the rotor with a puller. It's easy.

The PCB (printed circuit board) is now between the stator and the heat sink and there is no way to hook the stator with a puller. How to take it off?

DON'T HIT IT WITH A HAMMER! Use kinetics!

Put a solid aluminum pad on concrete, or a heavy cast iron table (my case). I tried hardwood first and then plastic, but everything chipped. Grab the motor (see picture) and slam it against the pad. I recommend you still line the edges of the motor with rags (not pictured). When the stator comes loose, don't let it get damaged. For me, it only took 3 hits and it was down.

The stator is stuck on the spline. You can see the glue in the photo, but it doesn't seem to be a serious obstacle.

When you put it back on (the same way) you have to put the bolt holes in the straight first! Also, it probably wouldn't be a bad idea to measure how deep it is seated with a depth gauge before disassembling it. The bump back has to be so that the transistors are touching the heat sink via the silicone pad.

The rotor must be put back just enough to put the retaining ring in. If you put it over the top (like I did) it will scrub against the stator.

The PCB holds one screw between the relay and the large capacitors. Also the 6 connections for the coils have to be disconnected. The brown thing underneath is a reversible temperature fuse.

The Fault

We have the engine disassembled and now it would need some measurements.

The schematic is not original, I drew the diagram, see below. So respect it!

Connect 230 V.

There are several voltages to measure.

320 volts on the big capacitors. It will float according to the mains voltage.

20 volts for the relay. It could float anywhere from 18 to 22 volts. That's OK.

3.5 volts for the MCU. It should probably be 3.3 V, but again, if it's stable, it's OK. I measured 3.5 volts.

2.5 V reference. That should be accurate.

Also, there's UVP - about 2.33V and the voltage for Halls 15V etc...

Study the schematic :-)

What could go wrong?

Power transistors? I don't know what they are. Whether they're FETs or IGBTs. Taking them out is a superhuman task. They're soldered three at a time on an aluminum PCB (primary heat sink) and then soldered to the PCB and sealed with lacquer. Parametrically, they will not be overloaded in any way. With a resistance of one coil of 600R, connecting them in a star and switching two coils to one potential and one coil to the other potential, we get a load of 900R, which at 320V is about 0.3A. Not much for destruction. Unless they're undersized for voltage...

Coils? Unlikely. Load see above, unless there's a winding short, or they fried after the transistors went out. In the case of the transistors and coils, I'd dare guess that the capacitor charge protection will kick in and drop the relay (see schematic).

Source IC VIPer? I read on the net that someone changed it.

The relay? Unlikely. The relay's job is to short the charging NTC resistor, so it doesn't switch any currents.

Resistor in the signal path? NEVER!!!

...never say "Never".

Usually components that are mechanically, power or voltage stressed will fail. However, this is a rare case in my practice and cost me about 3 weeks of research.

Why the R15 150R went away is a mystery (marked in the picture). It was simply interrupted. I figured it out when I wanted to measure if there was any communication creeping in. That resistor is on a 3.5V signal, roughly 1.1V eats up the LED in the optocoupler, so there will be a maximum of 2.4V on the resistor. So the power dissipation is about 0.038W. This will not be.

Either it's some "Chinese" resistor or it's a fuckupper, but I don't believe in such a sophisticated invention.

I have another theory, but it's pure speculation.

I found someone on the net who claimed to have had an intermittent zero resistance to the optocoupler, but unlike me on the side to the Mainboard. It's actually a jumper between the optocoupler and the connector. The resistor is not in the schematic and is hidden somewhere under the silicon from the large capacitors. I didn't believe it, but if it happens more often that the resistors around the optocouplers go off then the theory is worth thinking about.

Static electricity.

The drum is plastic and moves sometimes very fast (when spinning) in a warm and sometimes dry environment (when heating), making it an ideal generator of static electricity. In the same way we can generate a discharge when putting dry laundry into a dry washing machine. Components at the interface of two potentials are the most susceptible to surges. In this case, they are the galvanically isolated source for the motor and the source for the mainboard. These two potentials are separated by optocouplers, but there are no protective elements (class Y capacitors, discharge resistors, spark arrestors). If an ESD discharge occurs (and it will), it goes either through 230V or through the signal path (optocouplers). The MCU behind the optocoupler has a good chance of surviving because the basic ESD protection is on each of its pins. A resistor exposed to a surge (short high voltage pulse) can either change value (increase) or be interrupted. I know this from experience in security where it is the resistors in the sensors that go out after a storm.

But as I say this is just speculation.

As for the schematic, it's not perfect, but the circuit diagram will (hopefully) be fine. There are no unplaced components and zero resistors drawn! The designators (Q1, R22, C17 etc) are fictional. I did not measure the values of the ceramic capacitors, so they are not described. For some components I did not trace the type (e.g. Hall, or TVS D9 etc.). I tried to read the MCU, but it is locked, which is to be expected :-).

On the forums people complain that you can see cracks in the ferrite on the stator. I don't know where the comrades from DDR made a mistake and if they made it at all, but in my engine there were cracks too and it works without problems. Ferrite is, after all, glued iron powder, so whether the particles are separated by glue or cracks doesn't matter.

I wish you the best of luck with the repair!

cztomeco@gmail.com

Edit 29.11.2020

It's been three days since I got a call from a friend that the washing machine is not working again and he said it's doing the same thing it did two years ago. The next day I decided to look into it and went to the scene. I went to find out. I removed the motor, took the plastic cover off and started measuring the resistances around the optocouplers. And sure enough, one was broken. This time it was the R11 according to the schematic, but it was also 150R. I stand by the static electricity theory, but there may still be the theory of the very defective 150R resistors. In that case, the defect shouldn't be there anymore, because from what I looked at the schematic, there aren't any more. We'll see in two years :-)