Exploding caps

[minimutly]

Have been toying with different vsd's to run my Town Woodhouse, maybe will run two motors from this:

It's a rather old Siemens, pulled it and the control panel out of a skip years ago, tested it, it worked with an extract fan I wanted to use so I put it to one side.
I'd had a look inside it, my feeling was that it might pop some of its supply smoothing capacitors, but today I thought I'd test it.

And pretty much, every one of those white blighters went bang. I let the smoke subside, pressed go and sure enough the drive ran. Then tripped. I'll have to check but I guess it was undercurrent - the motor is only 1kW, the drive is 2.2...
I have a manual, so will buy some caps and see if it will reset and run.

 

[MetalMonkey]

A lot of drives do have specific capacitor conditioning procedures if the drive has been stored for some time.

 

[Kram]

Unusual for PP caps to go, its usually the big electrolytic ones, which in a vfd would be on the DC bus. They smell terrible and make quite a mess - but my experince is of blown small ones.

 

[Robotstar5]

Unusual for PP caps to go, its usually the big electrolytic ones, which in a vfd would be on the DC bus. They smell terrible and make quite a mess - but my experince is of blown small ones.

Used to have a bank of big electrolytics for smoothing in the first electric robots (early 70's) which had a tendency to blow a hole in the end and act like a smoke grenade the other issue was drive power transistors blowing and spraying metal onto the adjacent PCB card so if one drive board blew you had to replace two of them

 

[Hopefuldave]

A lot of drives do have specific capacitor conditioning procedures if the drive has been stored for some time.

If it's not clear (the procedure, not the bulb), a 60w bulb will limit the inrush current - best to feed through that from a variac though, start off at low voltage and build up to full mains in small steps over a few hours to reform the dielectric layer on the foils.

Those popped ones though... replace with fresh ones, they're the filter (not reservoir) caps, and the brown on the end of the foils is what's giving the leakage and can trip an RCD. Then you can try it properly - does it have any kind of display for programming / fault finding?

Dave H. (the other one)

 

[minimutly]

Replace them? Why yes of course. But tonight I just switched it on again. Ran a treat, so I connected the table drive to it. Less than 2.5A total load, not cutting. So need to order some caps, fit them and put the lid on. Then sort out a control system.

 

[Martin-Sy]

If it's not clear (the procedure, not the bulb), a 60w bulb will limit the inrush current - best to feed through that from a variac though, start off at low voltage and build up to full mains in small steps over a few hours to reform the dielectric layer on the foils.

Those popped ones though... replace with fresh ones, they're the filter (not reservoir) caps, and the brown on the end of the foils is what's giving the leakage and can trip an RCD. Then you can try it properly - does it have any kind of display for programming / fault finding?

Dave H. (the other one)

Yes, I agree. They are interference limiting caps (suppressors if you like). You can see one end goes to chassis ground.

 

[minimutly]

Of the four old vsd drives I've rescued recently, all of them have had the same issue with these mains suppression caps blowing. So I kind of expected it.
On the subject of caps blowing, I have a 240 to 415 ups which I've tried to get working, fitted new batteries. switched on, flame throwing electrolytics. First time a bit too close to my mush for comfort. I'm thinking I'll give it one last go, maybe borrow a variac as you suggest? If it does it again it's going in a skip.

 

[Seadog]

If it's not clear (the procedure, not the bulb), a 60w bulb will limit the inrush current - best to feed through that from a variac though, start off at low voltage and build up to full mains in small steps over a few hours to reform the dielectric layer on the foils.

I looked into this some years ago with the idea of making a capacitor reformer. My understanding is that full working voltage should be applied with current limitin. Not, as used to be thought, applying a gradually increasing voltage. I can't remember specifics but think that it was steps of tens of milliamps for periods of 30 minutes.

 

[Pete.]

I recall @awemawson having to reform the caps in his large spindle drive he described his procedure in detail.

 

[awemawson]

And it worked

My memory of what I did is hazy, but here is a quote from a post I made on Madmodder back in 2017 :

"When I did the ones on my Chipturn CNC lathe a few years back (as it had stood idle for 3 years before I got it) I just rigged up a lab supply, set to the rail voltage of the capacitors, but with the current limit wound down to a few milliamps. That way you could observe the voltage slowly increasing, and if it plateaued before it got to the rail value you knew there was an issue with breakdown."