Hi,

I have a dead coffee machine that I would like to fix. After some probing, I have found the power supply to be faulty. Problem is, I have no schematics or information I can reference.
It looks like a SMPS built around a DM0265 controller. It has a transformer without any markings that makes it quite hard.

Anyway, the transformer has 4 windings around a common core. The primary had 120 windings, the next on top of that had 15, next there was 9 and finally 17. The 15 wind coil was connected to VDD on the DM0265 and nothing else.
The other two where connected to the rest of the logic. The start (or end) of the 9 turn was VSS for the board as I traced a few tracks. The other side of the winding was joint to the start of the 17 turn winding. This junction went to a L7805 and the end of the winding was connected to the rest of the circuit. (See attached drawing)

I need to know what the expected output voltages would of been. I guess the output going to the L7805 would not matter too much as long as it is less then the max of the vreg as it does not go anywhere else. The other tap went to stepper motors so it could be anything..

What I do know is:

• The controller worked at a frequency of 67KHz.
• The transformer had a surface area of 49mm squared and made form the black ceramic like material.
• The voltage on the stepper motor controllers is between 10 and 46v.
• Input voltage to the transformer was ~300 DC.

I had a look at a few calculators on the web but can't really make any sense out of them.

As this is more of a exercise it trying to learn, I would like to workout what the output voltages would of been. I will then see if I can find a replacement transformer or wind mine own.

So, can anyone help me work this out?

The other thing I don't understand is why this failed. When I pulled the transformer out and checked the windings I did not find any broken or shorted winding but when I applied power it got hot (too hot to touch) and I did see quite a bit of blackening on the board around it.



Thanks

 

when I applied power it got hot

Do you mean applied power to the transformer alone, or applied power to the complete unit?
If the latter then it might be a shorted turn in one of the windings which wouldn't necessarily be obvious or it may be a fault elsewhere which results excessive current through the windings.

Edit: As you have the transformer core, and you know the number of turns and wire diameter, you could rebuild the original transformer.

 

Hi,

How old is the appliance? Check all the electrolytic caps.
How many watts did it say it has for input power?

To know how it worked you would have to trace out the whole circuit ideally. That would mean we could look around for a reference design and answer a whole lot more questions about how it works and what voltages to expect here and there.
If it is anything like a regulated wall wart (and it sounds so) then the circuit is not too complicated.

 

AlbertHall - I applied power to the transformer with no load. Thinking about it more, I don't think that was so smart as I only applied 50hz and it was designed for 67KHz. Would that make it self heat? In fact, I don't think I have ever asked this, but what happens to any transformer with no load?

MrAl - I had a look at them all but they all looked and checked out good. The machine is far too complex to reverse engineer, that's why I was hoping someone could help me out with some math to work it out. Wattage on the machine is for the whole thing. There is other power supplies for the heater and motors. This supply is for logic only so I have no idea. I do know that the DM02560 does not have a current limit resistor so its max is 1.5A.

This was more of a exercise in learning a little more about transformers then trying to fix the coffee machine (it is 10 years old and I have already fixed it 4 times in the last 2 years).

Does anyone know if there is a calculator out the that I can give frequency , core size and windings and it will give me voltage? All the calculators I can find workout turns form voltage, I need the other way...

Thanks

 

I applied power to the transformer with no load. Thinking about it more, I don't think that was so smart as I only applied 50hz and it was designed for 67KHz. Would that make it self heat? In fact, I don't think I have ever asked this, but what happens to any transformer with no load?

That transformer will not work at 50Hz. It's impedance will be much lower and the current much greater (depending on applied voltage of course) and so that may well lead to it overheating.
In my many years experience of repairing this type of supply, transformer faults are very rare.

For a 'normal' 50/60Hz transformer it is easy to relate voltages to turns. In a switch mode supply things are a lot more complicated.

 

Photo of the powers supply?

 

For a 'normal' 50/60Hz transformer it is easy to relate voltages to turns. In a switch mode supply things are a lot more complicated.

Indeed. Normally I could work it out in 2 secs but the frequency bit is getting me. Still would like to know how it is done. More reading needed me thinks. Anyone know any good texts on high frequency transformers?
I was only guessing that the transformer was ar fault as it and the area around it was black. No other parts looked damaged.

Analog kid - I can post a photo if you like, sans the transformer, but I understand how the rest of the circuit works. It is more of a question of the transformer used.

Thanks

 

If you feed an HF transformer with a sine wave within its frequency range then the same turns/voltage calculations apply. This circuit supplies the transformer with pulses and that's where it gets more complicated.
You could have a play here: http://www.poweresim.com/
Set up the input and output voltages and current then click the 'recommended design' button. On the circuit it shows click on the transformer and it will show you the number of turns.

 

Analog kid - I can post a photo if you like, sans the transformer, but I understand how the rest of the circuit works.

... because you've seen it.