[sci.electronics] homemade welder

cas@mtcchi.uucp (2668-Charles A Sherwood(Z550900)0000) (10/11/90)

Hello people

Is there anyone out there that has build a welder using a surplus aircraft
generator. I have acquired a 400amp 30 volt aircraft generator and I am
considering building a welder out of it. 
Burdens Surplus also sells surplus aircraft generators for this purpose
and they provide plans, which I have a copy of and it goes like this:


	|------------|
     ||-|generator   |A-------------arc stablizer--------
	|------------|	   |				 work
	   |	 F	   |			   gnd---
	  Gnd	 |	rheostat
		 |	   |
		 -----------


The rheostat is used to control the welding current.
The arc stablizer prevents the arc from burning the generator's brushes
and provides a smooth weld.

This has lead to a few questions. 
1) Is it worth all the effort? Will it work as well as a regular welder?

2) An arc stablizer is a large coil of wire. Burdens sells them, but can
   I make my own? What is the correct wire size? How much wire is needed?
   What is it wound on? Iron or air core?

3) How much horse power is required to drive the generator? If I use a 
   5HP electric to drive it(5hp motors are cheap and available). 

   	5HP x 746 watts/HP = 3730 watts

   Assume 70% generator efficiency, 
	3730 x .7 = 2611 watts.
   	2611 watts /25 volt = 104 amps of welding current. 

	IS this enough?? or is a larger drive motor required?
	A larger electric drive motor would not be cost effective.
	A 7.5HP single phase electric motor retails for about 400, 
	which would by a nice welder, with accessories.
	A 8HP gas engine would be an alternative, but is a little
	noisy.
		8HPx 746 x .7 / 25 = 167amps (more than enough)

chuck sherwood

jgd@rsiatl.UUCP (John G. DeArmond) (10/11/90)

cas@mtcchi.uucp (2668-Charles A Sherwood(Z550900)0000) writes:


>Is there anyone out there that has build a welder using a surplus aircraft
>generator. I have acquired a 400amp 30 volt aircraft generator and I am
>considering building a welder out of it. 
>Burdens Surplus also sells surplus aircraft generators for this purpose
>and they provide plans, which I have a copy of and it goes like this:

>The rheostat is used to control the welding current.
>The arc stablizer prevents the arc from burning the generator's brushes
>and provides a smooth weld.

>This has lead to a few questions. 
>1) Is it worth all the effort? Will it work as well as a regular welder?

Jeez, you're bringing back old and somewhat horrible memories.  The short
answer is No and No.  Sell the generator and spend the money on a nice
DC buzz box.  It'll weld better and you'll have time to actually weld 
instead of futzing with this thing.

Before I explain why these things don't work well, let me quickly outline
what happens when you stick weld and what characteristics a power supply
must have.  The process of stick welding is actually quite complex and
very dynamic.  To pick an arbitrary point in the cycle, assume the arc is
struck, a weld is in progress and we want to take a series of snapshots.
An arc is flowing in the plasma formed by the ionized air and the
vaporized flux.  A drop of molten metal forms on the end of the rod and
as it grows to about the size of the flux coating, a combination of
forces propels it toward the work.  As this drop travels across the gap,
it momentarily shorts the electrode to the work and extinguishes the arc.
 Then the trailing edge of the drop breaks with the rod, the  arc
reestablishes and the process repeats. 

Now let's look at the electrical requirements.  When the arc is flowing,
the power source must supply whatever voltage it takes to sustain the arc
at the designated amp level, normally 15 to 50 volts or so.  As the drop
of metal starts flowing across the gap, the gap is narrowed and thus the
impedance of the arc  decreases until at the moment of short circuit, the
impedance is zero. At that point, only parameters outside the arc
environment limit the short circuit current.  This job falls on the power
source.  If the power source is not current limited, for all practical
purposes, an unlimited amount of current will flow during the short.  
This has several very negative consequences.  The strong magnetic forces
involved will tend to distort the weld puddle and blow it and the plasma
out.  And as the short breaks, the extremely high current cause a violent
arc to establish which forcibly blows much of the filler metal out of the
joint. 


To address this problem, stick welding power supplies are designed as
constant current sources.  In other words, the power supply will supply
its setpoint current into a wide variety of loads ranging from a short
circuit to an established arc.  In reality, most power supplies are not
perfect constant current sources but a range of variation as much as  2:1
is usually not noticed by the weldor.  For inexpensive line operated
welders, this constant current characteristic is obtained by designing
the  transformer for high flux leakage and quick saturation.  For more
expensive sources, the regulation is done with a combination of magnetic
amps and electronics.  

For your basic generator set, the constant  current characteristic is
achieved by designing the generator as a compound differential generator.
In this scheme, the field consists of both a series and a parallel winding
and they are connected to buck each other.  Therefore as the load
increases, the field actually become weaker as the series winding
bucks the parallel winding and as a result, there is less voltage to
force current through the circuit.  This extremely poor voltage 
regulation is just what the doctor ordered for welding.

The second major consideration is arc stabilization.  Because the weldor's
hand is not perfectly steady and because local conditions in the arc
can vary, the voltage necessary to restrike the arc after the metal
drop breaks the short is highly variable.  It is desirable not only
from a restrike standpoint but also from an initial strike standpoint
to have a fairly high voltage available instanteoutly.  This is normally
achieved by including some excess inductance in the output path.
This inductance induces a "kick" voltage on top of the transformer/generator
voltage as a function of the dI/dt through it at the moment of break.

For basic line operated welders, this inductance is typically built onto
the transformer core but magnetically isolated from the main winding. For
generator type welders, the inductance is typically in the form of a
stabilization coil of the type you mentioned in your post.  Generally the
more inductance the better and I've actually added more inductance to
marginal welders by winding a toroidal inductor from copper tubing placed
around the stator of a fractional hp motor and insulated with  fiberglass
tape!  Too little inductance makes the arc hard to start and very hard to
maintain. 

Now let's analyze the generator you have.  Your generator is most likely
a shunt field generator which means the field consists of many turns of
fine wire, draws relatively little current and operates connected across 
the armature output.  This configuration is designed for good voltage
regulation - exactly the opposite of what you want in a welder.  The
aircraft system required a regulated voltage not only for consistent
operation of its loads but also for controled charging of its batteries
- just like a car.  

What you'd see if you build a welding power source from this unit is
an arc that is impossible to regulate and which splatters and spits and
blows metal all over the work.  The stabilization inductor will mitigate
the severity of the problem by virtue of an inductor's tendency to resist
changes in current.  The fact that the field is driven from the armature
lead will also help a tiny bit.  But because of the electrical and magnetic
time constants of the field assembly, any correction from this mechanism 
is too slow to regulate the arc and tends to make the voltage oscillate.

Now it IS possible to melt two pieces of metal together using one of
the "scratchie-stickie" rods like a 6011 but you won't be proud of the
results, it won't be strong and you'll develop a chronic "I'll never
be a weldor" complex.  Trust me, I know.

Your best bet is to sell the generator, perhaps to someone else with
the welder gleam in his eye (If your concience would allow you to - mine
won't) and perhaps add a bit of money to your proceeds and buy a DC
buzz box.  I've got a very nice one like I used to sell from my shop for
under $300.  You can't even start to touch a gasoline engine for
that amount of money.  And a brand name like Miller or Lincoln won't cost
much more.  Then if you want to heliarc, you can add an outboard high
freq arc stabilizer.  Or you might want to look around for a nice
used gasoline powered unit.  I consider a 230 amp gas welder just about
the optimum unit if you have any remote work to do at all.  I rarely
paid more than $150 nor got much more than $300 for these units even
mounted on small trailers.  Welding duty is pretty easy on engines
so they should last a long time and even after the engine gets weak,
the govenor will compensate.

>3) How much horse power is required to drive the generator? If I use a 
>   5HP electric to drive it(5hp motors are cheap and available). 

Most commercial welders use gasoline engines in the range of 12 to
~20 HP.  I'd consider an electric motor to be economically impractical
for this application.

Hope this helps,
John

-- 
John De Armond, WD4OQC  | We can no more blame our loss of freedom on congress
Radiation Systems, Inc. | than we can prostitution on pimps.  Both simply
Atlanta, Ga             | provide broker services for their customers.
{emory,uunet}!rsiatl!jgd|  - Dr. W Williams |                **I am the NRA**  

rcte2q@jetson.uh.edu (10/16/90)

It's been a while since I read this aarticle, but in the late 80's, AUDIO
magazine reviewed a (then) new Mark Levinson monophonic power amp, model 31
or 16.    The review team waxed poetic about the near invulnerability of the
design.   As an extreme attempt to make the amp malfunction, they fed a 1 khz
test tone in, hooked welding rods to the speaker output, and cranked up the
volume.   They showed a picture of the rig as it cut metal.   As the cut was
being made, the test tone could be clearly heard emanating from the plasma.
Talk about esoteric....

--Jazz Ninja