Looking for advice on homemade spot welder design

(this is a copy of a post made on sci.electronics.design)

--
Hi. Lately i've become interested in building a small spot welder for
DIY use. The idea is to end up with a (relatively) portable unit, with
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Miller welding machine manuals usually include diagrams of their electrical circuits. See how they make their spot welders.
i
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Jul 26, 8:36 pm, Ignoramus16379 <ignoramus16...@NOSPAM. 16379.invalid> wrote:

I have a lot of tech documents for spot welders, both from Miller (transformer based) and Image Industries (capacitor discharge based). The electronics on these devices are rather simple - what i'm looking for is for wisdom from people who built similar devices for themselves, or have experience with both types to help me decide on the welding system.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Jul 27, 3:40 am, snipped-for-privacy@gmail.com wrote:

I have played with building a couple of spot welders. The first one I built with four big variac cores. The cores were from about a kilowatt variac. I used the existing windings which were for 120 volts and wired two cores in series in parallel with the other two cores in series. And then used 240 volts.
The secondary was made aluminum. It filled the hole in the variac cores. So I had four cores with a laminated bar going thru them. I never got to finishing it, but did use it to fuse two 3/8 diameter bolts together. Just crossed the two bolts and clamped an electrode on each side with a C clamp. As the bolts melted into each other I tightened the C clamp until the bolts were pretty much in the same plane. Took a minute or so.
Dan
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Nice! Did you recall getting too much overheat on either the output probes or the transformer secondary?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Jul 27, 6:34 am, snipped-for-privacy@gmail.com wrote:

The secondary was the laminated aluminum bar about 1.5 inches in diameter. I don't recall it changing temperature. The primarys were the original winding on the variacs. They must have warmed up a little, but not enough for me to remember.
Dan
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Most of the industrial spot welders are simply transformers with around .3 to .5 volt output to the tongs under full load. The simple hand held units http://www.harborfreight.com/cpi/ctaf/displayitem.taf?ItemnumberE689 http://www.harborfreight.com/cpi/ctaf/displayitem.taf?ItemnumberE690 http://www.millerwelds.com/products/spotwelders/msw_and_lmsw__air-cooled_/ have nothing more than a toggle switch for control. Clamp down, pull trigger for whatever you think works, release trigger, release clamp.
Basic info on spot weldeing here: http://www.millerwelds.com/pdf/Resistance.pdf
On 240 volt lines you need approximately 500:1 turns ratio. Core size and input need to be suitable for the kva rating you are trying for. Output cables and tongs need to be suitable for the amperage. Note that at 6000 amps at .5 volt, connection resistances need to be on the order of 1/100,000 ohm for things to work well.
For industrial applications in the 15kva to 125kva range, the weld current and duration is controlled for each application. It's common to have a specific series of pulses with varying timing and current. http://www.on-b.com/weld/wd-lc-nu.htm
For a home brew I think I'd look for a suitable base transformer core and primary windings in the 1.5 to 3 kva range. Pull out the secondary windings, install new secondary windings made from copper braid. Probably only need a couple of wraps. Hook them to some sort of clamped tong assembly. A simple timer/trigger circuit with user adjustable timer from perhaps .5 seconds to 3 seconds would be nice. I think you will find that this is much more of a mechanical project with fabrication issues than an electrical one.
snipped-for-privacy@gmail.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Great post - thanks for the detailed answers.

Interesting, this is a figure i didn't find elsewhere. Guess it excludes any metal but copper for the clamp.

Yes, that was pretty much the idea all along. Perhaps even isolated, bent solid copper bars to reduce resistance in the winding.

I know - i want to make a fancy uC controlled welder, and even then is a rather simple project (electrically :). I have some experience building mechanical assemblies, and i have a pretty good idea on how i'd build the clamp - basically two thick L shaped copper bars with fittings for replaceable tips and a hand operated spring assembly to close it with some force. The whole assembly kept as close electrically to the secondary as possible.
Do you know if this kind of welder is well suited for small works, where localized heat might be a problem? I'm asking because all the literature i found suggested a cap discharge assembly for "finer" work. I can control the welding time with precisions in order of 25mS (realistically, with 10ms for both latch and release),and i stil don't know if that would be enough. The nice thing about capacitor discharge gives an exponential decrease, which means that the part is stressed (a lot) for only a brief moment, with a current slope following, which i think is gentler on the parts to weld.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I've used both a capacitive discharge and a small transformer unit (the one sold at www.sisweb.com) to spotweld small filaments to posts, stuff like 0.001" thick by 0.030" wide rhenium ribbon to 0.03" thick stainless steel, and the capacitive discharge units were much easier to control. Both got the job done, though, with some practice. At work we have a 15kva spotwelder that looks like it was made in the 60's, with an electronic control box from the 90's added on. It does zero crossing switching so the time is in increments of 16.7 msec - you program by telling it how many cycles to apply. That would probably make your current switch last longer if you go the transformer route.
-- Regards, Carl Ijames carl dott ijames aat verizon dott net (remove nospm or make the obvious changes before replying)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
It pretty much depends on what (drum roll please!) WORK you want to do with it.
I'm into auto body work, small electronic boxes, robot base parts, etc. Combine a 3 in 1 (press/shear/roll) with a 2.5 kva spot welder and a whole lot of projects become quite easy. I haven't gotten those items for my home shop but I did get the prototype lab at work to pop for a shear, brake, and a borrowed 120 volt/1.5 kva spot welder.
If you are doing anything with smaller wire (thermocouples, etc) then you want the capacitive discharge units. I looked at a surplus unit a few months back: 1 to 100 joules, 30 pounds, $30.
snipped-for-privacy@gmail.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Nah, the smaller i would go would be welding small part for boxes or tabs to batteries, all the way up to welding sheet metal to build enclosures. I'm pretty sure i'll go the transformer way - my only concern was that i wouldn't be able to control output energy as finely as with a capacitor device.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
For sheet metal work is is just not that critical. The portable units do not have a timer at all. You just pull the trigger, wait for the spot to get a suitable color of red, release. After about 3 welds on a particular combination of metal and you will have a good sense of what is "not enough", "just right', and "too much" I did some pull tests of student welds done with a portable unit, the timing was not particularly critical.
In the sheet metal range it takes something like 50,000 amps per inch of total metal in the weld area. It doesn't take long to find that more power is better. Also, it doesn't work on aluminum, carbon and stainless only. It works on galvanized stock but you need extra power and the tips crud up quickly.
Projects are fun but is there any way you can just buy one of the import knockoffs? Sale prices around here are US$150. Not an issue to get some sort of suitable base transformer but a set of copper tongs will set you back US$40
snipped-for-privacy@gmail.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I don't claim detailed knowledge of spot welding technology, but I did build a capacitor-discharge spot welder with transformer-coupled output for welding weldable strain gauges to railroad rails several decades ago. About 50 Watt-seconds energy level, as I recall. All analog control circuitry using a stud-mounted SCR and a PUT (Programmable Unijunction Transistor) for triggering via a microswitch on an aluminum channel handle actuated by an aluminum electrode- holding beam mounted at the rear of the channel handle.
As I recall, there is a "resistance welding" industry group that offers technical literature and sets standards of welding tip alloys and shapes. I don't remember the exact name of the group.
Worked great for its purpose. Weldable strain gauges are mounted on thin sheet SS (I think) - or maybe nickle. They are mounted by making a string of spot welds to the test object all around the active strain gauge element.
The whole motive in building my welder was to get portability for welding remote from line power (and to have fun), so it was operated off two 67-volt "B" batteries in series. They charged the caps to an adjustable voltage (we always used the maximum available voltage, as I recall). This was before the days of readily available, inexpensive, DC-DC converters, as I would use today with a 12-volt gel cell power source.
The transformer was a steel-core inductor that had (as many inductors do) extra space available in the E-I core window for a copper-braid secondary of very few turns. I think a microwave oven transformer with secondary removed would work for a capacitor discharge output transformer. It was chosen somewhat arbitrarily on the basis of having reasonable looking core and magnet wire size - perhaps by looking at literature on commercial units. From memory, I'd guess the core was about 3.5" x 4.5" x 2" stack thickness. I used about 3 feet of relatively small welding cable between the transformer secondary and the ground clamp and welding probe.
At the time, very high capacitance, low voltage caps were rare and expensive, which was the motive for using the output coupling transformer. I would probably investigate using capacitors directly today, but worry about sparking and erosion from the relatively uncontrolled discharge current.
Which reminds me, why are you thinking of using a triac instead of an SCR in a capacitive discharge welder? I haven't priced them, but would expect to pay much more for a high current triac than an equivalent SCR, and very high current SCRs are readily available and cheap on the surplus market. Check out candhsales.com in Pasadena (but do it soon, as they are going out of business).
As one raised on tube electronics, I also wonder what the advantage is of using microprocessor control for a capacitor discharge welder. Control of both charge voltage and triggering is pretty straightforward with analog electronics.
Many years later (but still many years ago) I tried using the same homemade welder for welding nickle tabs on AA, C and D cell Nicads. It worked, but only marginally. I think about 100 Watt-seconds would be desirable for welding battery tabs reliably. However, my problem may have been contact resistance, not welder capacity. I used a two- pronged copper fork to make ground contact with the battery terminal to avoid having the welding current pass through the active part of the battery. I don't know if the welding current would damage the battery, but I didn't want to risk it. Seems like a bad idea. The tab to be welded and the welder probe fit inside the fork tines with just enough clearance to avoid direct contact.
I think it might be a mistake trying to create a welder capable of both sheet metal work and battery tab attachment, as that covers a wide range of capacities and the handpiece/clamp and control requirements are radically different.
Have fun!
awright
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Nice work! I was consdiering looking for a scrap microwave transformer with an intact primary - those have rather big cores and can handle powers in the order of kVAs.

I'm kinda leaning away from the idea of caps or caps/transformer combination - i like the way you can control the output power of such configurations with ease, but caps are still expensive and, worst, even low-ESD caps rated for such discharge work get worn out very quickly, i'm told.

I just happen to have a few triacs laying arround. SCRs would do fine though - you can get ones rated for 1,2kA surge for very cheap.

Ahh another purist :) I usualy prefer analog electronics myself, but this process is very well suited to be controled with, say, a 16F628 microcontroller. You get very accurate timing/charge monitoring and can program a chain of operation easily, which helps making it safe. The PIC is cheap and would only need a minimum of parts to interface with the rest - i was planning to add a nice LED display though.

Actually, as long as you have both contacts on the same terminal there's little risk of current flowing into the battery. What *is* as problem would be localized heat - batteries can be damaged rather quicky in this way. Pulses must be strong but very brief (i.e., perfect job for caps) in order to avoid excessive heating.

Sorry. I meant about the device itself being able to handle such jobs, i.e, powerful enough to solder thick metal yet precise enough to be useful for smaller work. Of course, the clamp assembly for soldering sheet metal won't be of much use in finer work, but that could be replaceable. Batteries are not my main concern right now - but the ability to control solder timing to work in different sheet metals would be very nice to have.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

How would you use an SCR if you are leaning away from capacitors?
I thought that the use of an SCR is to trigger discharge of the caps.
Without caps, you can use [two] SCRs to switch primary side, kind of like those solid state relays, but you do not need a big one?
Am I missing a anything?
i
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

SCRs were mentioned as a trigger if i used a capacitor-based design :) I'll use a normal SPST relay in the primary for the transformer - been doing some tests, and i can switch it on and off with a precision of arround 20mS. I expect this to be a *tad* more under load (relay contacts might arc a bit) so i'm settling on 25mS intervals just to be safe.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Jul 27, 6:33 am, snipped-for-privacy@gmail.com wrote:
i have a pretty good idea on how

You can buy the replacement tips at WW Grainger. And I would recommend doing that. The commercial tips are a copper alloy , not pure copper.
I think the major difference between the transfomer and cap discharge spot welders is that the cap discharge can adjust the energy in smaller steps. With a transfomer unit, you are limited in how fast you can turn on and off. Pretty hard to control to less than 16 ms. The current slope probably does nothing. You have a big copper heat sink clamped on the weld area.
Microwave oven transformers are a cheap source for the cores. The secondaries are would separate from the primary so can be removed fairly easily. Need bigger core, use two or more transformers.
Dan
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I just had an epiphany... what if i placed a (swictheable) half-wave rectifier in the primary winding? I'd have to test if the 50/60hz pulses stresses the core too much, but it would cut output power in half. I don't recall ever seeing this in a resistive welder, would be interesting how it performs in real usage.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Not sure that I understand exactly what you are referring to here, but putting a rectifier on the INPUT of a transformer is a no-no. The transformer core will saturate from the DC component of the rectified waveform and when that happens primary current is essentially unlimited except for DC resistance of the primary and input wiring.
That's why phase-control thyristors are used to vary the primary power in a symmetrical manner.
awright
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I was thinking of a simple half-wave rectifier - namely, a diode in series with the primary winding. This way you get a half-sine pulse in every cycle, then zero volts. Like i said, i never seen this before and would have to check on how it stresses the transformer.
Another option would be using a reactance in series (most probably a capacitor, to compensate a bit for the inductive load) to limit the primary current. The capacitor needed would probably need to be quite large (and mains rated) though.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.