1. Home
  2. Electronics Design
  3. Capacitive discharge welder(again?),how to preserve capacitors?

Capacitive discharge welder(again?),how to preserve capacitors?

Hello

in years i got shelves full of surplus screw aluminium capacitors,and sometimes i needed to solder batteries without warm them.

I spent some time reading the many projects of capacitive discharge welder on the web,and i got an idea. Some are microcntroller based,and this is not a problem,some uses rugged MOSFETs or SCRs,some uses capacitors,other supercaps. I found a lack of description about the need , of limiting the capacitor discharge current,that often looks limited only by the Rds ON of the device. Of course the average power dissipated inside the capacitor can't be that big,because of the relatively long interval between operations, but having to do with currents of hundred of Amperes i wonder if some other phenomena happens,like hotspots,or similar.

The other question is the choice between capacitors directly ,or through a transformer. Transformer was considered because the most of capacitor i have ,are relatively low capacity and high voltage:

10.000 uF 100V,15000 uF 63V... I heard that the transformer would need some added tricks,as a time by time drive sense invertion to prevente core saturation. Furthermore i guess that some considerations are needed to avoid that excessive inductance would limit the rise of current, nullifying the gain of current due to the transformer turns ratio. Are there some advantages ?

Thanks for your help

Diego

Milan,Italy.

Reply to
blisca
Loading thread data ...

I've done a lot of experimenting with battery tab welding. I never got reasonable repeatability using high voltage discharges. Sure, you can make a weld, but I never got a whole battery pack welded together without at least one bad one.

I tried microwave oven transformers. Problems with them are significant. Transformer saturation was a huge problem. It is absolutely essential that you start from a known point on the B-H curve.

If you switch at line zero crossings, you can improve repeatability significantly. But the voltage is so low that the weld is VERY dependent on surface preparation, contact resistance, pressure. Some kind of repeatable fixture is essential.

formatting link
formatting link

Still not repeatable enough to use as a tab welder.

This is the output waveform for a Unitek 125 CD welder.

formatting link

It switches 400VDC into a transformer. There's a lot of magic in that transformer. Way smaller than I'd expected. Has a 4T secondary. Primary is unspecified, but the voltage ratio suggests 228Turns. Note the core reset before the pulse.

This is the schematic

formatting link

Weld results are VERY much more repeatable.

I found that debugging a spot welder is not easy. When the pulse fires, the magnetic fields are so high that every scope in the room triggers and displays noise greater than the thing you're trying to measure.

Reply to
mike

I wouldn't worry about damaging a big computer grade cap. We had a lab device that vaporized a sample in a tungsten boat. They originally used a big knife switch, but the spark as the contact was made eventually wrecked the knife switch. I put in a really big SCR and a little R-C circuit to apply a trigger pulse. This unit ran hundred of pulses a day for years. The same C-G capacitor was used in both the knife-switch version and the SCR version, and just kept working.

Jon

Reply to
Jon Elson

On Feb 23, 2017, Jon Elson wrote (in article):

The key when doing capacitor discharge, especially with electrolytic capacitors, is to ensure that ringing does not cause reverse voltage, so the circuit should be a bit overdamped.

The second issue is to ensure that the capacitor can take the pulsed current. Most computer caps can, but photo flash units are usually best with flash-rated capacitors.

Joe Gwinn

Reply to
Joseph Gwinn

Although not so much because of good performance, but intentionally poor performance: they have enough ESR to limit peak discharge current to a few hundred amperes. They probably have limited life as a result, but thousands or millions of flashes is still an awful lot for a camera!

Really pretty amazing getting that much energy into such a tiny package.

Computer grade caps, in turn, will handle more surge current, but you need to be careful that it's not simply dead-short-surge current, which is intense enough to cause damage!

The point about reversal is spot on. In general -- don't simply let things "be". You are the designer! Design that system! Add resistance or inductance if needed. Research the small signal parameters (C, ESL, ESR, etc.) and evaluate the RLC circuit. If it's critically damped or better, you avoid reversal; and if the peak current is within limits, you avoid damage (which is usually also going to be overdamped, which is nice).

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com
Reply to
Tim Williams

What!! Dammit man I thought I was just a passive observer of all the carnage. You've changed my life view.

NT :)

Reply to
tabbypurr

Some people come to watch the trains explode...

formatting link

Engineers calculate how many cars are needed to make the explosion. :)

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com
Reply to
Tim Williams

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.