The WPC Transformer

I reassembled my Monster Bash cabinet. No playfield, but enough to test basic functionality and get the sound running so that I have everything possible ready before the big job starts.

Almost everything works first time, so I turn it off and put the DMD panel in place so I can test my new speakers. I turn it on, and up it comes - but no DMD. A few seconds later, the rear of the machine erupts in a huge cloud of smoke. The line fuse blows.

The problem with such a catastrophic failure is that you do not want to repeat it with a new part. Total troubleshooting is the only way to go.

Step 1: Is the transformer dead? There's an ugly melted hole in the tape so this is a good hypothesis.

Disconnect the three output cables. Replace the line fuse. Power on. No surprise, the fuse blows. The transformer is broken.

Step 2: Diagnose the transformer

To do this, we need to feed a low-voltage AC source into the transformer. I pulled out an old linear power supply that I built somewhere around 1978 apart, tapped into a 20VAC line, and fed that into one of the 110V input windings. The next step is to check output voltages from each output pair. At this stage, I am just looking for voltages. Pretty soon, I come to a problem winding. the 100VAC output for the DMD is a dead short. I confirm this by taking a meter to the output connector and some of the exposed wiring. It's a match. The 100VAC winding is shorted.

Step 3: Attempt to Remove the Bad Winding

Somewhere, there's a school photo of me aged about 10 years old with band-aids on two fingers on each hand. That's from taking apart a transformer with scissors and a screwdriver, so I have learnt to leave them alone. But this one is important. I figure that I can completely cut the 100V winding, my transformer might reward me with some voltage.

Here's the transformer after surgery.

If you look at the ends of the cut wire, you will see that they are light yellow. This is bare copper, after the insulation has boiled off as smoke.

Step 4: Test

A basic bench test looks good, so the transformer goes back inside the machine. Connect it up, power on, LEDS light and the machine makes a sound. No DMD, of course. But I can navigate test menus, and even get it to play sounds.

Step 5: What Else is Wrong?

All the fuses on the AV board check OK. But there's no 100V because it is disconnected, and no 60V. maybe I cut the wiring? More surgery needed.

 

Step 6: Total Troubleshooting!

BIG WARNING!!! This can be very dangerous. Pinball machine designers went to great efforts to make sure that every metal surface is well grounded. this means that, if you are working on a plugged in machine, you are probably grounded as well. Touching an exposed live wire could be fatal.

The next step is to build an alternate supply for the 100V and 60V DMD voltages. Most people don't know that you can run a transformer both ways. For example, 12VAC in, 120 VAC out. I have bought a little bag of transformers from Fry's Electronics and building them into a new DMD supply is the next step. When this step is completed successfully, I can be confident that I won't repeat the destruction of my last transformer.

So, here's my inventory of transformers and what I did with them:

Input Voltage Output Voltage Function
120 12-0-12 Convert 110 to 12V and 24V
120 6.3-0-6.3 Step 12V up to 110V (measured) as an alternative to the 100V supply
120-208-220-240 24 With 24V in, there is 88V between the 120V and 208V taps for the 80V supply

For test purposes. I made a mess with tape and jammed the input transformer leads into the female end of an IEC connector. Here's the whole sorry mess:

 

Here you see the original transformer with its defective winding flayed off so that it does not blow the line fuse. Parked on top are the three cheap transformers that substitute for the lost windings, spliced together and into the connectors with electrical tape.

 

And here's the machine with its display up and running! Note the credit dot, it misses its playfield!

So I have validated that everything in the machine is fine, and when I get a replacement transformer it will not burn up again.

Now, the voltages that I have are about 10 percent too high. This should be within design tolerances, but it will cause excessive dissipation in the regulator transistors so I don't want to run for too long like this. To get fancy, I could probably splice into the transformer input leads and find a tap that is 10V or so below line voltage, bringing everything into spec.

 

Funnily enough, there may be another solution other than a new transformer. All that's missing is power for the DMD. So one of the new LED DMDs could be used with my existing burnt-up transformer.