Shop built portable welder part 2

[ironnut]

I should have posted this part first, but to continue from part 1.


The alternator has to spin a bunch to do this. You need about a 3 to 1 increase in speed to do this. This means at 3600 rpm engine speed, the
alternator is doing about 11,000 rpm. Not to worry. Check out a small block Chevy. It depends on the gear ratio, but if you were blasting down
the freeway in the old days (75 mph) the Chevy motor was doing 2500-3000 rpm. The crankshaft pulley is about the size that you are looking for, which
means that the alternator is buzzing along at 9000 or so rpm. Stomp on it and passing gear will get you well above the 11,000 that your welder is running
at. So the engineers at GM have already considered this form of abuse for you.

The existing diodes are too small, so they get canned. You need at least 25 amp 200 volt PIV diodes. Get 3, stud-to-anode and 3 stud-to-cathode.
Fortunately these are fairly cheap, few bucks, bigger is better, but anything over 50 amps is overkill.

You need 2 heat sinks with plenty of fins like you see in large audio amplifiers. You mount 3 of the diodes of one polarity on one of the heat
sinks and the other 3 on the other heat sink. If you don't insulate the diodes from the heat sinks then the heat sinks become a common conductor
for the 3 diodes and one is (+) and the other (-). This assumes that the two heat sinks are electrically insulated from each other. If you follow
this suggestion then you must follow the suggestion to buy the diodes as cathode to stud and anode to stud types. You can do it with all one type
but it is more difficult and messier as a connection must now be made to the stud portions of the diodes and you need to electrically isolate the
diodes from the heat sinks.

Electrically isolate the heat sinks from the cabinet by mounting them on isolation shock mounts. These typically come either with threaded studs,
or threaded inserts. I used ones that were about an inch in diameter and had 1/4-20 threaded studs on them. Connect your welding leads to the heat
sinks. One is positive, the other negative. I bought some female connectors at a welding shop. They sometimes referred to as Miller style. They are
a tapered brass rods that mate with a tapered sockets. I mounted these sockets on the side so that my welding cables were not permanent with
the welder. The sockets come and red and black so that you can keep the polarity correct.

You can get the shock mounts from Grangers. The diodes form factor is a D0-5 shape and most electrical/electronic supply houses carry them.
A 1N1186 or 1N1188 should work fine.

For excitation I found that the self excitation route was a pain, so I obtained a small permanent magnet motor. You need 3 amps at 16 volts.
Burdens surplus catalog has bunches of them. I used a vacuum cleaner belts to drive the motor and made pulleys to give about a 2 to 1 increase
in the crankshaft speed. This ratio will vary depending on what kind of motor you end up buying, but the permanent magnet motor will have to be
driven at least the rated nameplate speed to get anywhere near the maximum nameplate current rating.

A simple power rheostat will suffice to control the field although a simple power transistor circuit out of a Radio Shack book is a nice touch
and it is what I evolved the circuit to.

I made a small box about 12 inches on a side and 6 inches deep. On one of the 12x12 sides I cut a hole to match the shape of the alternator. Make the
other 12x12 a removable door. The older GM alternators have 4 screws to hold them together. I removed the screws and made the hole in the box to fit over
the end of the alternator with appropriate bumps to match the casing. I bored holes in the bumps to match where the screws went. Then I put the screws back
in and now I had an alternator with a box on its back side. I mounted the heat sinks on rubber shock mounts which are electrical insulators and help
protect the diodes from all the engine vibrations. If you drill some holes in the cover and on the bottom under the heat sinks, the fan on the
alternator will now draw air over the heat sinks and keep the whole thing cool.

I constructed a mounting bracket that bolted to the gasoline engine. The alternator sits above the crankshaft, although it could be mounted on the
side. Alternators are not direction sensitive when it comes to generating electricity. However, the cooling fan may not be as efficient when operated
in a reverse direction but the effect would probably not be harmful. Also the brushes are mounted off to one side for the direction of rotation.
There are many approaches to mounting the alternator and it is a matter of preference and what kind of engine you mount it on.

The welder will run 3/32 rod quite nicely and if you push it, 1/8". It is a little hard to get the arc started and an arc stabilizer from Burdens would
probably help.

I still have mine and I made a second one for my brother. I went so far as to make a foot control for it and used it with a tig torch. Works pretty
good. You do a version of the Miller lift start in that you start at a low current setting make contact with the torch and add current as you lift
the torch away from the metal. That way it doesn't stick as bad. I welded a copper tube once, that was about the limit as copper really drains the
heat away very quickly.

Best thing is I haven't fried any diodes yet.

[flylo]

I've seen engine driven welder that are built from alternators & ran by truck engines under the hood. Is this the same? They sell them & plans I think on Ebay.

[micrometer50]

Here is another http://diy-welder.com/

[ironnut]

I've seen engine driven welder that are built from alternators & ran by truck engines under the hood. Is this the same? They sell them & plans I think on Ebay.

Without a specific example, I will have to reply with a qualified yes. My version which I built a good 25 years ago is pretty simple electrically. Back then those externally regulated Delco alternators were as common as dirt. You could use the later internally regulated versions, you would simply remove the diodes and the regulator. The 4 wheeling off road crowd like the after market under the hood set-ups. The Zena welder is one of the brands available.

For stick welding and running a TIG torch you want a constant current source. A modified alternator without any regulation makes a suitable constant current source. Basically it is a 3 phase delta connected AC voltage source fed into a full wave bridge rectifier. The 6 diodes are really the only electronics in the system as the field (rotor) can be controlled by a large low ohm rheostat.

The big trick for me to get it to work was rewiring the alternator from a 3 phase Y connection to the 3 phase delta connection. Those old 60 amp alternators in their stock 3 phase Y connection had an open circuit voltage around 130 volts when the field was fed 12 volts and spun up. Indeed some after market outfits advertised running your drill and saw (universal motor, no variable speed) power tools off of the alternator in your truck or car. They provided a switch and an outlet in a box that disconnected the battery from the alternators output and plumbed it to the outlet. The switch also disconnected the voltage regulator and sent the full 12 volts to the field. Then you spooled up the engine to give you 120 volts and plugged in your saw and had fun.

I have seen postings where individuals didn't modify the stator wiring to the 3 phase delta configuration. They claimed it works fine. Perhaps the 100 plus amp alternators out there these days would indeed work, but my 60 amp version did not until I rewired the stator.

The economics of building your own, are questionable, certainly not, if portability is not a requirement. I have used mine quite a bit when I was building our house to weld rebar together and the steel columns and support beams for the floor.

[rolland]

My father had one many years made out of a 200 amp dc WW2 aircraft generator with a two cylinder Wisconsin engine on it. The welder worked great if you had the patience to get the &*&#@# engine to start. The generator was off of a B-17 engine or so I was told.

[J Tiers]

I have this Eclipse Aviation generator, and an 8 horse Kohler, which is slated to become a welder. B-29 APU originally, also used other places, this one probably came from a tank engine.

problem with these is they are about 28V open circuit, which may make starting the arc difficult. Not as constant current as an alternator, either. I understand one puts an inductor in series, which could help both ways, particularly in arc starting.

Any details of what others did with these units would be helpful, so if you recall what the B17 gen setup was, i'd sure like that. (not meaning to hijack)

[ironnut]

I found this on Google books, the magazine section in an old Popular Mechanics issue, February, 1957. From the looks of it the generator is fairly simple. It is using self excitation to generate the field current given the simplicity of the control circuit. With the alternator version I had some undesirable interactions using the output of the alternator as the current source for the control field. I found using a separate small permanent magnet motor as a generator gave me a much more stable system. Perhaps the aircraft generator has some internal regulation for the field.

It is should work fairly well. The article includes instructions on how to create your own arc stabilizer, (a large inductor). You may need more than an 8 hp engine to get the maximum from this generator.

[J Tiers]

Constant output current is variable output voltage........

But constant output voltage is what keeps the field constant (not that you always want that).

What you did, providing a separate "exciter", is exactly what is needed to allow control of the field without interaction/interference with the load.

SOME variability and interaction may provide a quasi-constant current, since with lower volts out, the field is less, reducing the current.... Otherwise with a low load voltage a heavier current would flow.

The "constant current" of a welding power supply is not all that constant..... more like "drooping voltage".

[sasquatch]

Thanks for this interesting post.

There used to be a guy in my area who built a few of these welders for folks in remote areas,, and a few were on farms, a few were installed on service trucks.

[ironnut]

I had a forum member ask about additional information about the heat sink and connecting the diodes so I included some extra pictures.

Nothing magic about the rectifier system. Just about anything with fins on it will work for the heat sink. If you can find a dead old fairly powerful stereo system you can recycle the heat sinks. The ones I used are about 4" wide and about 6" long.

A 3 phase full wave bridge rectifier set-up uses 6 diodes. In this case 3 are cathode to stud and 3 are anode to stud. Most diodes in the 35 to 45 amp range have a 1/4"-28 thread stud, so you need to tap 3 holes in each heat sink and mount 3 of the cathode to stud on one heat sink and 3 of the anode to stud on the other heat sink. As stated in a previous post, the key with the old Delco alternators was to rewire it internally to a 3 phase delta connection. Then bring each phase out and connect it to a pair of diodes. Each heat sink in-turn becomes the positive and negative output respecitvely. You need some pretty heavy wire, #1 copper to connect each heat sink to the Miller welder style female plugs.

All Electronics has 1N1190A and 1N1190AR diodes on sale for $3.50 each. They should work quite nicely up to a 120 amps output of the welder. The vibration rubber mounts can be purchased from McMaster Carr (search for Vibration-Damping Sandwich Mounts). The female/female threaded inserts are the easiest to work with. The 5/16ths-18 thread 1 inch diameter one are $4.50 each. You might want to get 6 total 3 per heat sink as the 2 allow the heat sink to flex a bit. I have never had a problem with it hitting the inside of the box, but 3 instead of 2 mounts would make it more stable.

Incidentally the black electrical tap on one of the red wires coming from the inside of the alternator protects a current sensor that senses when current flows. It is used to signal the engine controller to increase the speed of the engine to full speed. Lack of current flow signals the engine controller to go to idle speed.

The diode mounting frame is ~1/8" thick aluminum bent to fit inside of the box I made.