3 phase DC motor/driver troubleshooting

[Stonent]

Our front load washer (2008 vintage) has stopped working and I'm trying to narrow down the issue.

It is a GE wcvh6400j1ww.

The GE front load washers use a 3 phase DC motor to spin the tub.  The issue I was having is it worked fine and then the following morning, I loaded some wash and it appeared to start but when I came back an hour later, it had obviously stopped.

Normally if you try to start a wash cycle, it locks the door and is supposed to rotate clockwise several revolutions and then rotate the other direction several revolutions then it unlocks the door waits about 5 seconds then locks it again and starts.  This is to check that the washer isn't overfilled and has weight to pop the door open.

Now it only rotates clockwise but does all the other procedures, starts filling with water for about 15 seconds then stops and drains itself and stops.

There was a service pamphlet under one of the panels that showed how to run some diagnostics test.  It had a stored error code of E54 which means problem with the motor or motor inverter.  None of the motor diagnostics will spin the motor.

According to several sites I looked at a bad inverter board is the most likely culprit.

There are two different versions of the board depending on when the washer was made.  One is referred to as the metal inverter and the other as the plastic because of the housing.  Many sites list the plastic inverter as a compatible replacement for the metal inverter.

On ebay the metal one is about 1/2 the price of the plastic one.  I have the plastic one but I took a chance and bought a metal one.

Everything connected up fine but whenever the motor relay clicks on, I hear a high pitched sound from the inverter similar to what you might hear coming from a high voltage lighting driver, and the motor does nothing.

What I'm trying to figure out is a good way to test that either board is attempting to push appropriate current to the motor.

The cable going to the motor has 6 wires.

On the board the wires are silkscreened like this:

1. Yellow - Tach 1
2. Yellow - Tach 2
3. Red - W
4. White - V
5. Blue - U
6. Black - Motor Frame

The black wire connects to a spade connector on the frame of the motor.
When I connect a volt meter in AC mode to the two yellow wires and spin the tub, I can get up to about 4 volts. So to me, that seems to indicate it is working.

I'm not exactly sure how to test the other parts.  I don't know if W, V, and U are standard terms for 3 phase DC motors or not.

I have noticed if the motor is not trying to spin there's 35 volts AC appearing on each wire when I reference to earth. But 0 volts between the wires.
Also which was somewhat concerning to me is the inverter is always electrically hot.   The relay on the control board only switches neutral. So I can have the washer powered off and measure 120VAC from the inverter input to ground.  When the relay clicks on, the inverter then measures 120VAC Live to Neutral.

So what I'm guessing is the inverter derives its standby power from Live to Earth since I do have a yellow and green wire screwed to the frame of the washer and the motor gets power from Live to Neutral.

As for differences between the two inverter boards.  The Metal one (which is the older variant) has the tach wires go directly from the motor to the control board and completely bypasses the inverter board.  The Plastic one has 2 yellow tach wires go from the motor to the inverter board but then only 1 yellow wire exits out to the connector.  The second yellow wire is not pinned up on the plug on the inverter but both wires are pinned up on the control board.

Anyway back to the main issue, seeing if the motor is being powered.  So I have red, white, and blue wires going into the motor and a black to the motor frame.  How should I test for voltage?  Should I be checking for DC PWM  between the wires, DC referenced to the motor frame, or AC referenced to something?

There's also a warning on the washer that says "Shock hazard, some components are intentionally not grounded"  I'm guessing most likely the motor.

[kleblanc]

Typically UVW is for 3 phase VAC. I wouldn't be surprised if there is an inverter for single to three phase in there somewhere.  I am not to sure how to go about checking out the power feed to the motor, but here's a link for checking the motor windings. This is where I would start first.

http://www.pcbheaven.com/userpages/check_the_windings_of_a_3phase_ac_motor/

[Stonent]

This is a picture of the label that I found on a parts site for this washing machine.


It could well be AC. One of the repair sites I was looking at called it a 3 phase DC motor so I assumed it was PWM rather than sine driven.

[kleblanc]

The repair site could have a misunderstanding/miscommunication about the motor. You generally don't see a 3 phase motor and inverter on consumer appliances. But then again I haven't changed a dryer belt in 10 years. 
The induction motor is a dead giveaway it's a AC motor and having UVW wiring I would 99.9% say it's 3 phase. There should be an inverter with some large capacitors inline with the wiring somewhere.  Be careful with those.

[Stonent]

The repair site could have a misunderstanding/miscommunication about the motor. You generally don't see a 3 phase motor and inverter on consumer appliances. But then again I haven't changed a dryer belt in 10 years. 
The induction motor is a dead giveaway it's a AC motor and having UVW wiring I would 99.9% say it's 3 phase. There should be an inverter with some large capacitors inline with the wiring somewhere.  Be careful with those.

Yes the board they call the inverter is what I changed.  The board was made by Emerson Industrial Automation.  One of the boards has 2 fat Nichicon caps, the other had 2 fat Nippon Chemicon caps.  So at least I should have a certain degree of confidence those are ok.

The boards are all conformally coated.

The older metal inverter has a fuse and a gas discharge tube on it.  The newer plastic one does not strangely enough.

[kleblanc]

To check the motor voltage you would measure u/v u/w w/v in AC volts. All three should be within 10% of each other assuming the inverter/VFD is maintaining the same frequency output. The problem with these units are some crazy Volts and Hertz readings.  A Fluke 87V is always useful since it has a Low Pass filter option that smooths out the readings. If I recall the last inverter I measured Hz on was something like 15 KHz, with the low pass filter it was 45 Hz.  Similar with AC volts, without the low pass filter it will generally make xx.x digits a blur on your meter. You'll see 1xx.x to 2xx.x VAC, with the low pass filter it might fluctuate ~5 volts.

I've worked on some these motors for HVAC equipment, so I have some experience in trying to get a useful measurement but by no means an expert.  I know there are other meters that have Low Pass filtering Dave's EEV rebadge, Flir, and Brymens.  Without a low pass, the readings are almost useless.

[Anks]

I would guess the old board has a bad get going off your description of the fault. I would guess fixing the old inverter could be trivial and messing around with new one is probably a hiding to nothing.

Just my two pence

[Stonent]

I finally got back to messing with this.  The ohms between U/V U/W V/W on the motor were all ~6 Ohms. So I think the motor is ok.
Both boards use one of these for driving the motor:

http://www.irf.com/product-info/datasheets/data/irams10up60b.pdf

IRAMS10UP60B
Series
10A, 600V
with Internal Shunt Resistor
Integrated Power Hybrid IC for
Appliance Motor Drive Applications.
Description
International Rectifier's IRAMS10UP60B is an Integrated Power Module developed and optimized for electronic
motor control in appliance applications such as washing machines and refrigerators. Plug N Drive
technology offers an extremely compact, high performance AC motor-driver in a single isolated package for
a very simple design. An internal shunt is also included and offers easy current feedback and overcurrent
monitor for precise and safe operation. A built-in temperature monitor and over-current protection, along
with the short-circuit rated IGBTs and integrated under-voltage lockout function, deliver high level of
protection and fail-safe operation. The integration of the bootstrap diodes for the high-side driver section,
and the single polarity power supply required to drive the internal circuitry, simplify the utilization of the
module and deliver further cost reduction advantages.

[Jeroen3]

And U/V/W to Chassis? You'd need a high voltage ohms tester for this. A megger.