PDA

View Full Version : spot welder



ptjw7uk
03-07-2006, 07:58 AM
Some years ago I found a circuit for a capacitive discharge welder for thin sheet. Now when looking on the web I cannot find any circuits but lots for sale $500 -$600. I do not want to spend this on a seldom used tool but would like to build one to see if it works!! Any one know where to look!!

uute
03-07-2006, 01:23 PM
Haven't seen Cap Discharge, but sounds like a neat idea. seen a few old Pop Mechanics designs around. Might try the Yahoo groups if you haven't already.

Wirecutter
03-07-2006, 02:14 PM
I'm pretty familiar with cap discharge welding. It's a really nice and easy way to fasten metal parts - even using dissimilar metals and unskilled labor. I've made my own battery packs this way - you get a 1/8" ribbon of stainless or nickel steel and cut it with wirecutters to get the strips that go between cells. The nickel steel welds nicely to the top of most batteries - the ones I was dealing with IIRC were tinned copper, or something similar.


Materials that don't conduct so well are easiest to weld using this method - I used to think of CD-weldability as kind of the opposite of soldering or brazing.

Copper, brass, and lead are a breeze to solder, but it's tough to CD-weld them reliably. It can be done, but it's a hassle. Even at that, it's much cheaper (less labor intensive) than soldering.

Probably the easiest thing I found to weld using cap discharge was anything involving stainless steel. It has to do with how well the metal conducts electricity, and while SS is a fine conductor, Cu, Pb, and brass are much better, hence harder to weld this way.

I believe the cap discharge welders available commercially use a capacitor bank that discharges into the primary of a pulse transformer. The secondary of the pulse transformer goes right to the weld head, and the distances are kept short and the cables fat. The reason is that it's easier to do it that way? It's tough to get a bank of low-voltage caps to have such a low impedance (high discharge amps) and stay that way. (You're talking about dumping hundreds of amps in a fraction of a second) It's also easier to regulate the amount of energy discharged into the weld junction if you can charge the caps to some voltage, then trigger the weld and dump the (known amount of) energy.

Commercial "stored energy resistance welders" have been made by Hughes in the past. I picked up a 500 Volt-Amp Hughes box for my old company at a surplus joint years ago. I paid $100 for it, and really regret that I didn't keep it for myself. At the same company, they also had Unitek 125VA and 250VA units.

From what I've been able to tell, the Unitek stuff I've seen out there is holding its value pretty well, so I can understand why you might want to build one instead. I used the Uniteks a lot, and I was able to weld an amazing variety of materials and physical configurations. We even had a tweezer like thing for a weld head, and I assembled a PCB one time that way - no solder at all. (The inspiration for this came from an old coworker that made oil well-head instruments. The instruments work in such a hot environment that if you use solder to put them together, the solder winds up melted in a pool in the bottom of the instrument)

Obviously I'm a big fan of cap discharge welding. One of the projects I have on my list is a welder of this type. Recently, I scored a pulse transformer and a weld head off of FleaBay for the project. They're both Unitek units, and they'll make a fine welder. The controller is up to me.

Right now, my big project is a fancy schmancy Swiss gear cutting machine - I owe my web page an update of the progress. When I come to breaks while awaiting parts and other stuff, I'll be working on the welder. Email the address in my profile if you want to compare notes.

-M


[This message has been edited by Wirecutter (edited 03-07-2006).]

rotate
03-07-2006, 02:39 PM
Spot welders require very high current with source that has very low internal resistance. It is possible to produce very large current using capacitor, but most of the energy will be lost in the internal resistance of the capacitor so it's not a very efficient method of welding.

If you do try, you'll need a relative low voltage (2 -5 V range) but lots and lots of capacitance. BTW, those "gold" capacitors which are rate at 1 farad can't be used because they have relatively high internal resistance. My guess is that if you have somewhere close 0.5 farad at 5V, you can probably weld light gauge metal. Monoster capacitors for car amplifiers are probably suitable.

Switching this circuit poses another problem. You'll get a spark and an unintended weld at switch. I've seen very high current DC switch made using mercury, however this needs to be in an enclosed chamber.

If you want to make a spot welder for cheap, make it out of a transformer found in old microwave oven. You'll have to add a secondary winding yourself.

P.S. After having read WireCutter's note, I realized that the capacitors don't drive the work directly but instead go through a pulse transformer. In light of this, ignore my post since I was thinking of something entirely different.


[This message has been edited by rotate (edited 03-07-2006).]

Mcgyver
03-07-2006, 04:30 PM
what voltage would this be done at? i thought spot welding was low voltage high amps, and basically takes advantage of the fact the resistance in steel is around 10x that of copper? I'm wonder if the cap discharge idea is really just a way to provide a timer, well not really a timer, but a way to stop the flow of electricity?

here's the delux one i'm going to build when i get around to collecting the parts

http://www.5bears.com/welder.htm

[This message has been edited by Mcgyver (edited 03-07-2006).]

Wirecutter
03-07-2006, 08:13 PM
<font face="Verdana, Arial" size="2">Originally posted by Mcgyver:
what voltage would this be done at? i thought spot welding was low voltage high amps, and basically takes advantage of the fact the resistance in steel is around 10x that of copper? I'm wonder if the cap discharge idea is really just a way to provide a timer, well not really a timer, but a way to stop the flow of electricity?
</font>

Well, let's see - going from memory I want to say the caps ran at around 200-300 volts. These would be those big "beer can" jobs with big metal straps connected to them rather than cables.

One thing to remember, at least with the cap discharge welders I worked with, is that they're not at all like the spot welders normally used for sheet metal work. Where the big "pincer" type used for auto body goes "MMMMMMMMMM" to do the weld, the cap discharge is more of a "thump" - It's all over in a fraction of a second, and the only heat is right at the joint. I held the work with my bare hands (well, fingertips, actually) and seldom even felt any temperature rise.

The size of the work is typically much smaller, too. The 250VA unit was good for batteries, small electrodes, and circuit board assembly. They do little work, and I doubt I could have joined car body panels or anything like that. The transformer I have for this has a primary that takes less than 200 volts and kicks out about 2-3 volts. Very low duty cycle, too.

But if you want to join small stuff easily and cleanly, it's a great way to go. I can imagine some interesting applications in small assembly or building model railroad stuff.

ptjw7uk
03-09-2006, 06:16 AM
Does anyone know what type the capacitors should be to withstand the high discharge - I assume they must be something like fash gun capacitors to enable them to withstand the high discharge rate!!

railfancwb
03-09-2006, 07:16 AM
FWIW, there was a recent review in one of the Village Press mags (not Live Steam) of the spot welder Harbor Freight sells -- often for $150. Positive review, actually, with negative comments only about the softness of the replacable copper tips.

http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=45690 You can download the manual in PDF, but it probably is 90% boilerplate safety warnings "Don't plug this in" type.

Charles

Wirecutter
03-09-2006, 10:29 AM
<font face="Verdana, Arial" size="2">Originally posted by ptjw7uk:
Does anyone know what type the capacitors should be to withstand the high discharge - I assume they must be something like fash gun capacitors to enable them to withstand the high discharge rate!!</font>

Photoflash caps (for driving Xenon flash tubes) would probably do it, but you'd need a lot of them. For the one I'll be putting together, I plan on using some big "beer can" electrolytic caps that I have. They're from the same family as those that used to go in big linear power supplies, but rated for higher voltage.

This may not be a feasible option for a big production welder with a high duty cycle, but for something "around the shop", I'd think it'll do. Getting big caps, at the rated voltage, that are designed for instant discharge may be pretty expensive. I'd go with the big old stuff. You may have to replace them from time to time, but it's probably cheaper than getting some "special" caps.

Wirecutter
03-09-2006, 10:34 AM
Another option to consider is just use a huge transformer, like a modified microwave oven transformer (see Mcgyver's link above), and put a controller in front of it. The controller could turn the primary of the transformer on for a set time, or for a certain number of AC cycles. I'd think a PIC or a Basic Stamp could do the job pretty easily. That might give you some tight control like that of the cap discharge welders, but without the hassle of doing the capacitor thing. If you synchronize the circuit to the incoming AC line, you could even turn it on for fractions of a cycle with SCRs or Triacs.

[This message has been edited by Wirecutter (edited 03-09-2006).]