Weldernator project

[Ian]

The Empire here needs a mobile stuck welder for fencing and so forth, and being money poor and salvaged truck parts rich, I decided to try to make a welder out of a Delco 22SI truck alternator (we have barrels full of dead ones at work). This particular model isn't as easy to convert as some, but I may have it worked out.

Basically, the rectifier and diodes are bypassed, the field circuit separated and brought outside the case along with each leg of the three-phase windings. The field will be controlled manually with a rheostat to tune the amperage output of the generator. The three phase will be connected to a high-voltage external bridge rectifier, and the welding leads directly connected to the rectifier output. I'll probably have to add a reactor in line with the positive output and some sort of capacitor or MOV across the rectifier output terminals to stabilize the arc and squelch voltage spikes.

The innards un-modified:



After isolating the regulator and internal rectifier leads as well as all other case connections. One leg of the field has been grounded to the case and the other (blue wire) connected directly to an i sulated through post. All six diode ring terminals were clipped off and only the insulated studs used to mechanically connect the extension leads.



Here's the whole thing put back together. Plan is to mount it in place of the ac compressor on the ranch truck.

 

[Ben]

This one will be interesting to watch.

Ben

 

[Cherokee]

You did what I had no idea you could do that. Got to see how this turns out.

 

[Ian]

An automotive alternator is nothing more than a wye-wound, three phase AC generator. The output is controlled by the electical strength of the electromagnetic field windings, i.e. 0 to 14 vdc, powered by battery feed and controlled by an internal voltage regulator. By removing the regulator, the voltage should run up at over 100 open-circuit volts at full charging rpm, and about 30 to 40 under welding load, which is about what a normal DC stuck welder outputs. The three phase outputs of this alternator are capable of delivering 150 amps each at 40 volts or so, plenty to burn deep with 1/8" rods. The internal bridge rectifier is bypassed because it won't last long at such high voltage, and excess heat is also a problem within the alternator case. The external rectifier can be mounted to a heat sink and is capable of continuous duty at 200 amps input and 1600 volts AC per phase.

If the alternator stator was wound to generate single-phase AC voltage, it could be used to generate current for AC welding once the 14 volt regulator was bypassed and voltage made high enough to sustain an arc, but they are three-phase so DC welding via a bridge rectifier is both efficient and powerful, which is why alternators are three-phase to begin with. I don't know exactly what this alternator will produce, but output voltage is dependent on rpm and load, and output amperage is dependent on field voltage (normal 14-volt vehicle charging voltage reduced as needed with a rheostat), so it should be possible to also run any brush motor power tool with this as well by controlling rpm for volt level and dialing in the field power to deliver the necessary amps.....such as a grinder at 6 amps/120 volts DC. Drills motors, saws, or anything that has a brushed AC motor should work if I only wire in an ordinary wall plug. Lack of an earth ground means care must be used when operating tools, no standing in a puddle or working in the rain.

Mainly I just want to be able to weld heavy pipe together for fence duties, the power feature will only be added if the output voltage is in the range I expect. Gotta spin this thing at a minimum of 4000 rpm to get 135 amps at 14 volts regulated, 6K to get 148, and I dont know exactly what the unregulated voltage will be. Alternator redline is 12k rpm. I'll put a recording oscilloscope on it one it's running and see what the waveform looks like while welding and how bad the spikes and dips are, and also judge by the arc start and break what kind of caps and reactor will be needed to reduce spatter and extend the life of the diode bridge.

In the end it's just a generator operating at enough volts to sustain an electric arc and pushing enough amps to melt steel, no big mystery to it.

 

[popper]

Used to be a commercial rig like that, a box you hook the welder to.

 

[Brad]

Way over my head

 

[fiver]

what he saying is an alternator actually makes A/C current.
he has to take out all the D/C converting junk.
the only problem is it runs on 3 phase cycles, and not 2 phase current like your house.
so he pulled a ground field to the housing, and then to the box on the table he pulled the other wires to hook his leads to.

now he needs to modify the bracket that holds the A/C unit on the truck to hold the new alternator, and figure a way to spin the alternator fast enough to make the appropriate amperage to melt a welding rod.
and a way to cool the alternator while he is using it to keep it from melting the internal wires or insulation around the wires.

 

[Ian]

It has a big fan on the front, goes on behind the pulley. Might be enough, might just have to pause between rods for a couple minutes and shut off the field power to let it cool down. Other people have done this same basic thing for years and having a field control switch on the stinger to shut it off beteen rods helps with cooling the case substantially. Not having the diode bridge hooked up inside (thingy with all the serrations around it that the phase outputs are hooked to) removes that major source of heat from the case too.

The crank pulley is 8", so I need a 2" pulley to get 4k rpm at fast idle. The plan is to run the engine at about 1500 rpm to get into the beginning of the torque curve and bog the engine less with a fixed throttle setting. If I could get a big enough crank pulley to get 4k alternator rpm at engine idle I'd just put a field power output to the air conditioning request input at the PCM and let the computerized idle air circuit and injectors manage the surge load just like it would the AC compressor load (about the same 12 hp load as the big 22SI alternator at full output), but that ain't gonna happen. Still have to source a pulley for the alternator, nobody makes narrower than 8 groove for the big Delco units so I'm just gonna run the truck's 6 groove in the eight or lathe bore the hole out of a 6 groove to fit. I'm a little concerned about smoking the belt under load so I might have to put a 4" pulley on the weldernator to get enough surface area and run the truck at 2000 to 2500 rpm. That would equal about six rods to a gallon of gas but if thats what it takes to make the belt live, so be it.

Overall a simple concept but lots of fabricobbling to make it happen.

 

[Pistolero]

Going to mount it on a truck motor as an extra alternator, or use some sort
of a two cyl motor and package it up? OOPs, didn't see the truck references.

Hmmm. Without a governor, may be problems, but a big motor may just have
the torque to handle it. Find a source for a used Kaw 2 cyl mower motor, they
are in the 18-25 hp range, so 750W* 20 HP = 15 KW, approx. 500 amps at 30 v. should
be plenty, even discounted for efficiency at 80%. Double V belts may be
needed. They have governors which will help a lot. Normal load the motor would
be not stressed much at 5-12 hp range depending on what size rod you are using.

Looks good, probably will need a big inductor, not too hard to create, and a good
cap to stabilize, as you said. Without enough L and C it could be a ringing fool, do
some real weird stuff.

Bill

 

[freebullet]

Turn the ac compressor into an air compressor, just add the weldnator. The 79 ford comp would run a 3/8 air ratchet or more. These days I'm more inclined to charge the ac & use the ready welder 2 off bat.

Cool project for sure. I would consider gearing it off a 6.5hp small engine. Can take it where the truck won't go, & cheap to run. 1 gal per 6-8hr vs 1/2-1+gph for v8. Carroll stream 6.5 w/ e start purty cheap new, easy to attach /gear for right speed.

Following with interest.

 

[Ian]

Bill, school me on the inductor. Heavy wall iron pipe with a 2-3 dozen wtaps of 2-gauge wire ought to stabilize 150 amps dc at 40-70 volts, but Im unclear on whether the inductor wire needs to be solid or multi-strand like the welding cable. Also, the ends of the core need to be connected, correct? Like weld a D-shape of square tubing to it? Or make a square frame out of square tubing and weld the pipe core in the middle?

The voltage ripple should be a good thing and it should have a pretty good power factor if I don't over-damp it with too much capacitance. That's why I was considering the MOV instead of a big cap as a clamping shunt. Will just have to look at the waveform and see what needs correcting.

Freebullet, I already have a Tecumseh compressor that I converted for onboard air, but not on this rig. I was going to direct-drive it off one of the pto outputs on the transmission or t-case (SM465/ custom divorced NP203 so there are 3 pto locations) but built a Blazer for wheeling instead. If this was a Vortec instead of a TBI 350, a simple $15 compression gauge hose and a toggle switch on the fuel injector control wire would give you even more air, weigh mere ounces, and not hog accessory drive space. Sure it will throw misfire and injector control circuit codes but who cares, just plug in the code reader and edase them when done.

I'd use batteries but I have a lot of welding to do, not just trail repairs. It will take at least a 9 hp engine to drive this welder, 13 would be much better. If it works out ok on the truck I'll build another on a wheelbarrow chassis using a second alternator and lawnmower battery to power the field circuit. Right now I don't have a good engine that big and dont want to spend $400 on a new one from HF until I get this figured out. So far I have about $45 into it and already have most of the odds and ends and long welding leads to finish it up on the truck.

 

[fiver]

throttle should be no problem whatsoever [a hand operated unit should take care of that, even one off a lawnmower] and your only gonna draw maybe 12 HP off the engine.

a guy with a lathe can just make a new pulley to line up, and reduce it by about 25% as an under drive.
if the truck is never goin on the highway, I'd pull the belt off the power steering pump and use the extra pulley from the crankshaft.
double up the water pump and both alternator pulleys, you could measure that stuff out with a ruler.

 

[Pistolero]

Coreless inductors work, but a core increases inductance. Yes, a straight rod would work but
a loop will get higher inductance from the same number of turns. Core permeability is the key feature.
A soft iron ring would be a good core to wind it onto. Like a piece of 1" rebar, bent into a ring and
welded closed, then wrap the turns around it. Will also tend to lock the mag field into the core, not try
to move it around by attracting itself to nearby metal. Core average permeability that is high, solid iron,
for example, gets more henries than no core (air core) or a hollow steel core. I don't think you have any
need for multistrand wire, solid should be fine, like all transformers.

The inductor dampens rate of change of current, cap dampens rate of change of voltage. To keep it
from ringing, you also need a good resistor in line. RLC circuit to get a damped output. Too long since
I studied electrical circuits to sit down and select the component values, but if you do a bit of web searching,
probably some simple refresher stuff out there. Then, once you decide you need 10 millihenries or something,
you need to design a coil, at least approximately from the standard equations for inductors.

Resistor - not sure. Might want to look inside a generator welder and see what they use.

Here you go:

You need (4L/C) - R^2 to be < 0 for a damped system. If it is greater than zero, it will oscillate (ring) and
likely blow something out.

Bill

 

[Ian]

It's a serpentine system, unfortunately. 83 chevy half ton 4x4 that I put a tbi 350 in complete with entire computer system and siamesed the electrical systems together at the firewall plug. Hand built the whole engine harness and it's fully emissions legal, though after all that I never registered it again. In the process I converted the whole belt system and water pump/fan to factory serpentine because I destest v-belts and I needed the late-model alternator to work with the PCM voltage regulation.

 

[Todd M]

Watching with interest. Had not heard of this conversion before.

 

[Pistolero]

https://kansascity.craigslist.org/for/d/kansas-city-briggs-stratton-engine/6794973077.html

https://kansascity.craigslist.org/grd/d/peculiar-16-hp-briggs-stratton/6785449369.html

https://kansascity.craigslist.org/grd/d/kansas-city-14-hp-kohler-engine/6807579923.html

They are out there.

 

[Pistolero]

Heck, I have a 4 cyl Ford car engine, complete, with the harness to the firewall that I will GIVE
you. Not certain of the condition, came with my place, just taking up garage space and the
guy who left it said it ran fine, from a wreck, was going to build a dune buggy and never got
around to it. You WOULD have to come and get it.

Bill

 

[Ian]

I figured the inductance value approximately already, just a little fuzzy on the actual construction of the frame, but that helps a bunch. The MOV was strictly to limit voltage spikes and protect the diodes when the asc is broken, a value of about 120 to 150 volts should do it. I figure a 3/4 hp motor run cap will smooth things out. The onboard welders I've been around scream like a banshee, obviously the loop wasn't clamped sufficiently. I may have the same issue using only a capacitor across the leads without a resistor and an in-line reactor. Also, zi know fhat one of the welder manufacturers uses a bunch of circuitry and timers to turn the output into dc square wave. Amps is controlled via pwm, volts is held constant. They can even be used as a MiG power supply because the output is so clean and well-regulated. I'm going for quick and dirty but beefy to burn DC rods. As long as it will support a puddle and not just splatter metal everywhere it will be fine, hence the need for a reactor. Welding upside down is no fun when the arc is splashing all the filler metal out.

 

[Pistolero]

Good luck. If you blow an alternator, sounds like you have more available. The varistor should protect
the alternator winding, I think. But the inductor will help smooth it out a lot, too. I think you will need
some resistance in there, but not sure where to get a 3-5 KW rated resistor. A chunk of 1/2" rebar
4 ft long bent into a zigzag pattern?

Bill

 

[Ian]

You need (4L/C) - R^2 to be < 0 for a damped system. If it is greater than zero, it will oscillate (ring) and
likely blow something out.

The C will be wired across the DC output terminals like a flyback diode would be and the L will be in series going to the stinger. Oscillations from the cap may cycle through the leads and across the arc, requiring R off of one or both legs of the cap to dampen the cycle, but not in-line with the leads.