Looking for advice on homemade spot welder design

[[email protected]]

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
controlable power, suitable for spot-welding anything from battery
contacts to 1.5mm sheet metal. I've ruled out buying a commercial
welder - mainly because they're rather expensive where i live, and
also because i happen to like the challenge

So, right now i'm contemplating two different designs, and having a
hard time deciding which one to go for:

* Capacitor discharge welder: The idea is to have a rather large
(100,000 to 300,000uF) capacitor bank which is discharged by a large
triac or similar electronic switch into the weld contacts. A cheap
microcontroller with an integrated ADC would control the charging
voltage, allowing for variable power, plus triggering the output
device.

* Mains transformer welder: The idea is to modify (or pay to have
built) a large (1 kVA to 3 kVA) transformer with only a few turs in
the secondary to get a low voltage, high current output. The power
controlling would also be handled by a microcontroller timing a relay
in the mains winding - i've had good sucess in switching off relays
very quickly (arround 5ms) by using a zenner in series with the back-
EMF catching diode.

I've considered also running a high voltage cap discharge through a
transformer to get a strong amperage pulse, as done in commercial
welders, but ruled it out - if anything, because it's overly complex
and overkill for a simple device as intended. Anyway, in both cases
i'm planning to build a simple clamp arm with copper contacts as weld
tips.

Right now i'm favoring the transformer option, if only because it ends
up being way cheaper than buying large electrolyitc caps + SCR. I have
concerns about it being used to weld small parts, as i feel that a
capacitor bank allows for much finer power control - yet, caps won't
perform as well in thicker parts... or will they?

So, my questions are... does anyone with experience with this type of
welders have any advice to share? If i go with the transformer option,
should i look into limiting the output surge current with an inductor
of some kind? How long can one expect electrolytic caps to last when
stressed with charge-discharge cycles like these? Any comments that
can help me make up my mind will be appreciated, thanks!

 

[neon]

yor problem begins with power @ 120v 15 amp standard circuits in the USA that is 1800 w per circuit and you might trip the breaker every time at that level to chage 100.000 ufd with this setup it will take time to charge to peak #14 = 2.35 ohms /1000' don't forget 2 wires. so you might have timing problems to beging with. triacs scr are good for ac only unless you learn how to shut them off. and for that kind of amperage they might exibit emitter crowding and blow uo. transformers are the way to go it is simpler and more relaible in the long run.

 

[Chris Jones]

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
controlable power, suitable for spot-welding anything from battery
contacts to 1.5mm sheet metal. I've ruled out buying a commercial
welder - mainly because they're rather expensive where i live, and
also because i happen to like the challenge

So, right now i'm contemplating two different designs, and having a
hard time deciding which one to go for:

* Capacitor discharge welder: The idea is to have a rather large
(100,000 to 300,000uF) capacitor bank which is discharged by a large
triac or similar electronic switch into the weld contacts. A cheap
microcontroller with an integrated ADC would control the charging
voltage, allowing for variable power, plus triggering the output
device.

* Mains transformer welder: The idea is to modify (or pay to have
built) a large (1 kVA to 3 kVA) transformer with only a few turs in
the secondary to get a low voltage, high current output. The power
controlling would also be handled by a microcontroller timing a relay
in the mains winding - i've had good sucess in switching off relays
very quickly (arround 5ms) by using a zenner in series with the back-
EMF catching diode.

I've considered also running a high voltage cap discharge through a
transformer to get a strong amperage pulse, as done in commercial
welders, but ruled it out - if anything, because it's overly complex
and overkill for a simple device as intended. Anyway, in both cases
i'm planning to build a simple clamp arm with copper contacts as weld
tips.

Right now i'm favoring the transformer option, if only because it ends
up being way cheaper than buying large electrolyitc caps + SCR. I have
concerns about it being used to weld small parts, as i feel that a
capacitor bank allows for much finer power control - yet, caps won't
perform as well in thicker parts... or will they?

So, my questions are... does anyone with experience with this type of
welders have any advice to share? If i go with the transformer option,
should i look into limiting the output surge current with an inductor
of some kind? How long can one expect electrolytic caps to last when
stressed with charge-discharge cycles like these? Any comments that
can help me make up my mind will be appreciated, thanks!

I can share my unsuccessful experiences:
I wound a roughly 1kVA (continuous rating) transformer with a few (tried 1
to 5) turns of welder cable (also tried paralleled lengths of 10mm2 wire.
I put a toggle clamp with copper contacts on this thing. What I found was:
most of the power was dissipated in the secondary winding, it was way too
thin, and, I was able to weld thin metal, but only if I pressed very
lightly on the electrodes so that they made bad contact with lots of
resistance. When I pressed down hard, there was not enough resistance in
the material to be welded and so the weld area stayed cool whilst the
secondary got hot. It was impossible to get consistent results because the
pressure and surface condition of the metal to be welded caused the contact
resistance to be either too high or too low many times.

I have also used a proper commercial spot welder which had a *much* bigger
transformer, much fatter secondary, much more pressure on the (water
cooled) electrodes, and which did much better welds.

You could try asking on sci.engr.joining.welding

Chris

 

[[email protected]]

What I found was:

most of the power was dissipated in the secondary winding, it was way too
thin, and, I was able to weld thin metal, but only if I pressed very
lightly on the electrodes so that they made bad contact with lots of
resistance. When I pressed down hard, there was not enough resistance in
the material to be welded and so the weld area stayed cool whilst the
secondary got hot.

Sounds like a 1kVA transformer doesn't offer enough output current to
weld consistently. This was also a fear of mine - that a trasformer-
based device wouldn't offer enough power or would do so only for a
brief moment...

You could try asking on sci.engr.joining.welding

I just did, thanks for the heads up!

 

[Clifford Heath]

So, my questions are... does anyone with experience with this type of
welders have any advice to share?

I have a small spot welder that uses a small Variac to charge a largish capacitor,
and a pressure-switch that activates a contacter to dump the cap through a transformer
when the electrodes clamp down hard enough. It's designed for tack-welding small s/s
wires in orthodontics before they're brazed into position. I can send internal photos
if needed... however:

It doesn't work worth a damn on anything more conductive than stainless steel, or on
anything above a couple of mm thick, and the parts must be touching between where the
electrodes touch. In short, it's great for its intended purpose, but not much use for
anything else.

If you want to make one, I'd modify a microwave oven transformer to replace the high
voltage secondary with a few turns of heavy-duty flat copper sheet, or flattened tube.
Bolt the I side of the transformer core together before grinding through the welds
that attach them to the E side, then you can load your new secondary easily before
re-attaching the I cores (clamp tight and weld again).

Clifford Heath.