Spot Welder Update

[Mcgyver]

as i rotate through the projects on the go, recently I made it back to the spot welder. Most of the esteemed advice of the forum is incorporated. I did away with the plastic block, made it steel, and made a toggle as the closing method, to be hooked up to a foot pedal, yet to be built. Used, as per AK's suggestion a plastic sleeve for isolation. lower arm is held in place by a split steel sleeve over top of the plastic that is then compressed by three set screws.

Upper arm bracket is keyed to the shaft, and then secured with taper pins to the arms. linkage is mostly but not completely finished, ie temp 3/8 bolts for temporary pins etc

I found a massive transformer and ran 1.5 turns of #4 wire x 4 (parallel turns) and am going to run the primaries on 220. This is a lot of wire, still doesn't come up to the copper arm cross section. hopefully any shortcoming will manifest itself in less of a duty cycle, which i can live with. I enlisted my 9 year old's help in removing the secondary, actually did a great job. I made a frame of 1" sq tube, the roll bar piece is so i can attach a handle after the case is in place - this puppy is heavy and awkward, with a handle it will be easy to get from bench to underneath

Controls. I read in the archives where someone (can't remember who) here had made a nifty micro-controller timer. you dial in, from an encoder, a number and when you hit start the micro-controller opens up solid state relay (opening the circuit on the primary's) for that many cycles, ie dial in 120 for 1 second. i may do away with the 2nd stage of the foot switch and just have button on the unit to start the cycle? if you hold the work with two hands, clamp with the foot switch, should be able to use one hand to hit the button - any insight on best commercial units?

made a steel case and will have a nice brushed font with all controls on it.

i think i recall reading somewhere that spot welder tips are made from beryllium copper? any know? mine need a clean up, but I'm chicken to machine beryllium copper.


since this is basically a compilation of the forums ideas, keep em coming!


here's some pics

[jungle_geo]

Single Shot Relay

I used single shot relay mentioned below (t2k-00010-461) to trigger a higher amp relay for my HF spot welder. I will eventually mount it on a stand and enclose the electronics in a panel, but it's a low priority project for now.

http://preview.ametek.com/content-ma.../T2-series.pdf

I believe I got it from McMaster-Carr (part number 7630K41?)

This single shot relay coupled with a foot switch has worked well to control the length of the pulse and get perfect welds.

I hope this helps.

george

[ptjw7uk]

If memory serves it should be Chromium copper alloy for spot welding rods and wheels. Beryllium copper alloy is for high temperature spring contacts

Anyway very nice job so far!!

Peter

[Forrest Addy]

Lessee if I can find any weakness. Nope - except for the #4 copper. Spot welders need a couple thousand Amps at low voltage to weld even thin stuff. A short high intensity zap that fuses the "nugget" and then off. #4 is good for maybe 100 Amps per. So you have about 400 Amps. A little weak maybe.

What did your tranfsormer start out as? I KVA or there abouts? Again on the bottom end but workable if the secondary is up to snuff. I predict this spot welder will heat the work but make a weak easily broken weld in light stiff like 20 gage. Try it and see.

If my dire prognostications are true, fix it by winding the secondary with as wide a copper buss material as will fit in the eye of the transformer. Make the stuff about 1/8" thick 1 1/2 turns doubled or any combination so its 1/4" effective thickness. Insulate with fishpaper bought at the motor repair shop. Lead the buss material out and bolt it directly to the electrode arms. Put a flex wrinkle in the buss leading to the movable jaw.

Otherwise, nice work.

[Alguy]

I tend to agree with forrest,,, it takes lots of amperage to make a spot welder
work ,,, I worked in a truck cab fab shop had many spot welders we did steel, and aluminium,, the alum cab weld cycle was about 1/10 second but required well over 3000 amps,, steel was lot less, voltage and amps but longer cycle time all were water cooled systems ,, my memory has faded on the details. I hope you have lots elec. service to handle the load you are asking of it.. also how much clamping force are you expecting? The final proof will be the "nugget" as always i am impressed with your work ,,

[CCWKen]

Peter is right about the copper. For best longevity and strength, use chromium copper. As Forest says, the secondary winding needs to withstand high amperes. Both my spots have a one-loop flat copper winding for the secondary. If I were thinking of making one, I'd get a piece of copper tubing/pipe and slit it to open it up and flatten. That's what I did to make some heat sinks. The copper tubing was 2" and about 1/8" wall thickness. I picked up about 2' at the salvage yard. I don't know if the tubing was schedule 40 or 80 though.

[bob ward]

Old stick welder into spot welder?

Like many people, I have a single phase stick welder sitting unloved in a corner ever since I bought a MIG.

I had thought that 'one day' I would be able to use it as the basis for a spot welder. After reading this thread, and now knowing that a spot welder needs maybe 1000 amps plus to start to be succesful, as opposed to the 150 amps that the stick welder was capable of on a good day, there goes another good idea.

[Mcgyver]

thanks for the encouragement...... i did listen to you guys first time round and agree the circ mills of current set up (#4 x 4 in parallel) will be the weak point. I got the wire cheap and the transformer cheap so thought it worth a try.

Forrest if your prediction comes true, I will have to try a heavier secondary, either that or use it as a crappy shop heater.....hopefully won't have to. short of cutting the transformer in two, not sure how'd it would get wrapped using a 1/8" flat section....seems like an easy(ish) job with the transform apart but next to impossible with as is (welded together). might be easier to add a second transformer and 4 more #4 wires? hmmmm.

I'll report back when trials have commenced

[J Tiers]

Even if you get the secondary wired with heavy copper, the primary can limit current due to its resistance. You can figure out whether you may have a problem with that (assuming the wall can provide the power), as well as what the real limiting factor is likely to be.

Many transformers of that size have around 1 volt per turn. You can very easily determine your volts per turn by connecting the primary to power and measuring the secondary open circuit voltage. Divide the volts by 1.5 (for your turns count), and there you have it. For more accuracy, wind several turns of light wire and measure that. Use the number of turns in your "exploring coil" to calculate the volts per turn.

Assuming 1 volt per turn, for example, your open circuit voltage would be 1.5, and going back to the 220V primary, you would have 220 turns. Turns ratio is then 220/1.5, or about 146 to 1

Primary current is divided by the turns ratio, so for every 146 amps in the secondary (welding current) you have 1 amp in the primary.

If you needed 1460 amps to weld, you would have 10 amps in the primary. Probably quite do-able. Even 2000 amps would only ask for 14 amps, still do-able.

However, the secondary is another matter altogether. With 1.5 volts open circuit (per the example, your voltage may be different), you would need a total resistance of only 0.001 ohm to allow that current to flow.

Your 4 strands of 4 gauge have a resistance of 0.00025 ohm per foot. A rough visual estimate puts your wire length at 3 feet or so at least.

At that, your wire resistance is 0.00075 per strand, and 4 parallel strands are 1/4 the total, or 0.00018 ohm total.

You should NOT have any problems with the wire resistance if the estimates are close to correct. The wire cross section will carry the current just fine for spot welding at up to 2000 amps.

The concerns might be other things like getting the wire connections to have very low resistance.

And, I was under the impression that spot welding open circuit voltages were a little higher, like 2.5 V, in order to be sure to drive current through the joint including a bit of oxidation, etc.

In your case, using our assumed 1.5V, and a current of 2000 amps, you would lose about 0.38 volt in the windings, leaving just over 1.12 volts to drive current through the metal. The total resistance of the un-welded joint would need to be no more than 0.0005 ohm.

It is possible that a lower current will pre-heat the joint and allow the final high current to flow. But it needs to happen fast.

An open circuit voltage of 2.5V would allow 2.12V for welding, assuming little added resistance in the secondary (for instance by reducing primary turns and increasing core size). That would allow almost double the possible joint resistance at full current.

Incidentally, the power delivered into the 0.0005 ohm joint is 2000W. The same current will deliver 4000W into a higher resistance joint, IF you can drive the current via a higher open circuit voltage. The difference between that and the total power of 14A x 220V is losses in the wiring and transformer, amounting to about 1000 W.

There are one or two other matters that limit transformer currents, but for this discussion I am only considering resistance. (The various leakage inductances and other second order considerations may have an effect at very high currents.

And I have neglected primary side resistance, mostly because I have no idea what it is. At a guess, 220 turns would have a length in that transformer of perhaps 220 feet or so. If wound with 18 gauge wire, at 6.5 ohms per 1000 feet, there would be a resistance of 1.4 ohms (cold). At 14 amps, that is a drop of 20 volts, or about 9%.

"referred to the secondary", that would be equal to another 0.14 volt drop, leaving less voltage to drive the current. The new loaded voltage would be 0.98 volt "for the joint", requiring a resistance of no more than 0.00049 ohm to allow 2000 amps, and adding some more transformer losses. It would still be less than half the actual drop in the secondary, however, so not a huge concern aside from adding to the concern over open circuit voltage.

Mind, if you DO change things to get 2.5V open circuit, the calculations of primary current etc, have to be done over, and you may find that you have lots more primary current, maybe enough to make the primary side resistance quite significant.

[Mcgyver]

JT, thanks for taking the time to work all that out.

Finally today got around to testing. Using the 220 mains, I get 2.35 volts on the arms so I'm encouraged that the 1.5 turns worked out (couldn't have fit anymore wire anyway).

The transformer is for a sodium lighting system, the secondary area was much larger than the primary. that worked out well as obviously the rewinding with bulky #4 takes up a lot of space.

The other nice feature is that the transformer had several Primary taps so that it would work with a variety of input voltages. I'm on the 240 tap now, but could have gone down to 208 if the voltage at the arms was too low - nice to have that flexibility.

Now on to the controller. if i can get a timer surplus, that'll be the way to go, still, I like Rob's method

http://bbs.homeshopmachinist.net/sho...ht=spot+welder

then I've got to learn programming, finish the cnc so i can route the circuit for the microcontroller (i hate etching). might take a awhile.