24th February
written by Todd Harrison

This is Part 3.  You may want to read (part1) & (part2) first.

In this part I was going to probe the board to find where the signal was being lost for the controls. After mapping out all the connections and looking up the functionality of the ICs in the earlier postings I was confident I was going to track down the problem. I had the motor secured in my bench vise, the power supply and speed control board connected to the speed sensor feedback and mains 120v to a switch. When I turned on the 120v mains power I could smell something getting hot real fast. I had to work 30 second at a time and let the control board cool between tests. Something was not happy in the power supply! The heat was coming from a large 4watt 1.8K ohm resistor. I didn’t dare probe for more than 30 seconds because it was clearly getting too hot. But I pushed my luck too far (KA-BAM!).

READ —>:

Yes a trace turned into a make shift fuse and the bridge rectifier was also destroyed.

I cleaned up the PCB to get a good look at the carnage.

My guess is the bridge rectifier shorted across the mains and vaporized the traces to the 120v mains connection. Strange being the bridge rectifier wasn’t getting hot at all. The below is the old 10A bridge rectifier.  It tested as shorted between the middle pins which is where the 120v mains connect. You can’t have a dead short between these pins in a bridge rectifier without getting some magic smoke!

I had a salvaged bridge rectifier from an older teardown project so I checked it’s specs online. It was good enough for testing at 8A so I dropped it in.

Plus a big fat jumper wire to replace the vaporized trace.

Time for more testing. (NO I’m not going to test it using 120v mains when its in the vise! Don’t even ask, I’m just taking close-up photos.)

But now I’m in no mood to continue probing this “widow-maker” hot with live 120v mains. I had stopped by Circuit Specialists in Mesa, AZ earlier and picked up two of each IC on the board for just a few cents. I’m done probing around hot so I’m just going to replace each IC and stand back a few feet when I switch it on this time.

Darn, it still didn’t work even with all new ICs :( . It seems like all my repair projects lately have all been failures. This one really has me stumped. I know I could probe it more and find the problem being this board is so darn simple but with it overheating so fast I just don’t want to be near it  with probes. I know it’s not the bridge rectifier or any of the ICs. I made a jumpier across the fuse to make sure that wasn’t holding back on the current somehow. I have testing all the transistors in circuit and they seem fine.

So what next? I will find a known good 200V 3300uF capacitor and make sure that isn’t somehow the problem. I have found capacitors to be the problem with power supplies many times but normally they show damage and/or don’t test. If that doesn’t get it I will just start test components cold by removing the ones that are best tested out of circuit and hope I find a bad one.

I just don’t want to test this thing hot anymore. There is plenty of magic smoke left in this power supply and I’m not looking to let anymore of it out. Working with 120v mains right in front of your face is NO FUN! I don’t recommend it to anybody.

Just posted part 4.


  1. 24/02/2011

    I know well the “magic smoke”. Some time ago I shorted a broken fuse from a PC power supply with a wire. It was a bad idea, not only the magic smoke appeared, but also magic sparkles. Since then I’ve been avoiding this kind of repairs. Don’t give up!

  2. 24/02/2011

    @Jonatan. Thanks for the support. I will be trying a few more things but it is already at that point where it might be best to just salvage another working motor control on some old equipment.

  3. Raff

    Hi Todd,
    I have some suggestions for you. Buy an isolion transformer, they can be a little expensive, but it’s a life saver when it comes to working on mains powered equipment. Wire a 100 watt standard houshold light bulb in series with the active lead from the tansfoemer to your power supply. Then you can gauge the amount of current being drawn by the lamp brightness, and avoid another ‘bang’ moment. Bright bulb equals short circuit! And lastly, have you checked the power supply to the chips? Rule one of servicing was hammered into my thick head years ago, check the power rails! I’d be suspicious of that 1k8 4 watt resistor. Hope you get it going!
    Raff, from Australia.

  4. Raff

    Sorry, it’s an ‘isolation’ transformer. I need a spell checker!

  5. 18/03/2011

    Nice talking with you. I ordered the hospital grade “Tripp Lite IS1000HG 1000W” Isolation Transformer. It was a bit spending at $314 but my safety is worth it and I will be able to use my scope a lot easier being I can use the ground clip again and not two probes in (Ch1 – Ch2) mode which is a pain.

    Other models that are not hospital grade and/or low wattage that I considered were:
    Tripp Lite IS1000 1000W $262
    Tripp Lite IS500 500W $171
    Tripp Lite IS250 250W $99

    Thanks for that great tip!

    I’m going to build that light bulb AC current indicator / short circuit protection device too. How clever! My motor will not run to well with a ~1A light bulb limiter but it will protect the board from another big bang. Thanks again.

    I did check the voltages. 17.6v DC which is just fine for these chips.

  6. 19/03/2011

    Americans don’t use fuses, hm?

  7. 20/03/2011

    There is a 250v 8amp HRC fuse on the board but it’s after the bridge rectifier so it wasn’t of much help. Go figure they would design this board with the fuse after the rectifier and it is ONLY on the motor side of the circuit. The control side with the ICs tap off the 120V mains and is half-wave rectified without a fuse. This power supply was built to a price point not for quality or safety.

  8. ExtremHybrid

    Hey Mr. Harrison.

    I tried to find you on Twitter, but I guess you dont have an account yet, so I will just comment here.

    I just wanted to point out, that at the bottom of your “part 2″ you forgot to link to “part 3″. I was reading along and I failed to perceive a link that would advance me to the next part of the PSU series, like in “part 1″ at the bottom.

    Maybe you want to correct that.

    Greetings from Germany,
    – ExtremHybrid

    P.S.: Found you thanks to a shout out on Twitter by Mr. Dave L. Jones (EEVblog).

  9. 20/03/2011

    I will have to do that. I forget to add such links somtimes.

Visit Our Sponsor: MROSupply.com - An e-commerce site of Mechanical Drives & Belting