"Reccomended" SCRs for CD spot welder / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

"Reccomended" SCRs for CD spot welder

Move Thread

LAN_403

ConKbot of Doom

Thu Mar 07 2013, 06:11PM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

I'm working on a capacitive discharge spot welder, and I'm currently trying to spec a thyristor to use. I have ~1.15F worth of

capacitors on the way, with 12.5mOhm for the ESR per capacitor, inductance is obviously going to be the limiting factor in the current pulse, and I was intending on adding a series inductor(catch diode too, of course) to keep the pulse from getting too high.

Is there any SCR's/thyristors that anyone would reccomend for an application like this? I can certainly browse and hope I can spec one right, but other than surge current, I'm not sure of any other gotchas.

I'm currently looking at the KP300A off of ebay, ~5kA surge capability, maybe splitting the bank in two, and feeding two of the thyristors off the same pulse transformer.

There was also some GE stud thryristors on ebay which looked appealing too. Are the pulse ratings on the chinese branded items going to be grievously over-rated? from vishay, I'm looking at $150 or so for a 2-3kA surge thyristor, and $250 or so for a 8+kA thyristor. I probably only need 2-3kA of capacity, but obviously overhead is nice for doing other things besides little nickel tabs on batteries.

I'm completely open to other suggestions for thyristors if anyone has them, or other switching devices (i know paralleling IGBTs is bad, maybe some sort of creative mounting for TO-247 mosfets? )

Thanks!

Daedronus

Thu Mar 07 2013, 07:37PM

Registered Member #2329 Joined: Tue Sept 01 2009, 08:25AM
Location:
Posts: 370

I don't think you will use all the 1.15F each pulse...
I would use a bank of mosfets so I can adjust the pulse length.

Sulaiman

Thu Mar 07 2013, 07:44PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

in my opinion, the best thyristors for pulse discharge are 'hockey-puck' style
'Stud' thyristors are amongst the worst
the most common industrially used package is the 'block',
with a conveniently electrically isolated base.
TO-247 packaged thyristors are good value for lower currents.

I have read that transistors use more silicon area per amp than thyristors
along with a couple of other manufacturing benefits,
thyristors are much cheaper than transistors per amp.
Thyristors also seem to be 'tougher' than transistors of a similar rating
the only thing that I don't like about high power thyristors is that when testing igbts they read >1000MOhm at 1kV on our meter (for typical 1200V transistors)
when testing thyristors I consider >100 MOhm good/excellent
identically rated thyristors from different manufacturers or even batches have a wide range of leakage, and manufacturers will not publish a min./typ./max.

P.S. in pulse discharge the enormous currents cause conductors to significantly attract or repel each other - mechanical shocks, can cause problems.

ConKbot of Doom

Thu Mar 07 2013, 11:58PM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

Thanks for the info about the packaging Sulaiman, I don't know why I was looking at studs over modules, I used an IXYS MCD-72 module for a disk launcher in the past and it worked well. The whole clamp issue makes using a puck SCR not so attractive due to the bulk of the whole thing.

As for as the mosfet bank, I haven't seen any setups that Ive been too keen on as far as having a sold high-current path. Ive seen one with TO-247 devices and a laminated bus which was the best I've seen, but it still seems a bit fiddly, and a bit involved with fabrication, imo.

I was thinking more along the lines of splitting the bulk capacitance into 3 (not necessarily equal) banks one for the smaller first pulse of dual-pulse welding, and the other ones for adjusting the main pulse voltage/peak current and total energy. Along with a selectable output inductor to limit the peak current. A 4th 'bank' with a capacitor or two charged up to a voltage than the other caps should be able to commute the current from the other banks, shutting off the SCRs, quickly dropping off itself, and the current in the inductor flows though the catch diodes to avoid reverse biasing the caps, unlike the traditional v-switch.

I'll be sure to reinforce my bus bars with some garolite/phenolic so they dont try and rip my caps in half.

As far as leakage goes, at 12-50V I'm sure the capacitor leakage will be much more problematic than the leakage on the thyristor.

The prices on the various IXYS MCD devices is looking good, so I may go with those.

macona

Sat Mar 09 2013, 10:03AM

Registered Member #3272 Joined: Mon Oct 04 2010, 11:40PM
Location: Beaverton, OR
Posts: 101

The capacitive discharge welders I have worked on tend to use high voltage instead of low. The stud welders that I have worked on used two large caps in parallel and dumped the power though a hockey puck. Delivered energy is controlled by a voltage regulator circuit that limited the maximum charge in the bank, I think it was about 200v max(rectified line voltage)

My hughes capacitive discharge welder also charges up two caps up to a couple hundred max and then dumps the energy though a mercury displacement contactor (pre-scr days) and that is fed into the weld transformer for a low voltage, high current pulse.

Hockey puck are the way to go, you really dont need a big heat sink, just a clamp to attach terminals. The mass of the SCR itself is enough for a pulse this short. Couple bars of aluminum on either side and some bolts with insulating washers.

You should also know that many of these new high capacity capacitors do not tolerate dead shorts very well. Unpleasant things happen to them when they are dead shorted.

ConKbot of Doom

Thu Mar 14 2013, 11:54AM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

What were the CD welders youve worked with designed for macona? The thermocouple/battery tab/other small stuff welders IVe seen online (both professional and DIY) are generally in the 12-20v range and use capacitance in the Farad range.

The u32D range caps that I'm using are designed for high ripple current, 17A rms @60hz for the particular ones I have, which makes the bank of 40 good for 680A @60hz (!) So I think they should hold up ok to the welder duty.

macona

Fri Mar 15 2013, 10:25AM

Registered Member #3272 Joined: Mon Oct 04 2010, 11:40PM
Location: Beaverton, OR
Posts: 101

The bigger one was used to weld on those little screw studs you see on electrical panels and stuff. Like these units:

The Hughes one is intended for welding on metal tabs and stuff like batteries. You can get all sorts of weld heads for doing all sorts of things.

680 amps is nothing for a spot welder. Even a cheap harbor freight spot welder will deliver 1000+ amps depending on the length of the tongs.

-Jerry

ConKbot of Doom wrote ...

What were the CD welders youve worked with designed for macona? The thermocouple/battery tab/other small stuff welders IVe seen online (both professional and DIY) are generally in the 12-20v range and use capacitance in the Farad range.

The u32D range caps that I'm using are designed for high ripple current, 17A rms @60hz for the particular ones I have, which makes the bank of 40 good for 680A @60hz (!) So I think they should hold up ok to the welder duty.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob