How do you weld anything by melting only the surface?
Sounds like the welding equivalent of a dry soldered joint,
Welding needs penetration.
--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)
It is an experimental automatic welder, following the idea of TIG (with shielding gas - argon etc.). I don't know the details. My task was to design a controlling circuit which lets the user to set the duration of pulses, PWM parameters and some delays (the process should be repeated). Now, the controller generates all signals (this was checked separately - without the welding part, only turning on/off a small load). After connecting the ionizer, the microcontroller goes crazy after about 5 seconds of electric sparks.
Here is a simplified schematic (hand drawn):
The "ionizer" is a factory made device to ignite electric arcs in welders. It should be connected through a transformer (e.g. 24:3 windings). The secondary winding (connected in series) adds HV pulses to the circuit. The circuit for these pulses should be closed by R and C (just above the ionizer on the schematic). Precise parameters are not known. But it generates "thick" sparks about 5 mm long, at the frequency of the mains (50 Hz). It may be about 7 kV.
And, these HV pulses are getting somehow trough the "high current" wires to the IRFP transistor AND, through the optocoupler (!), to the microcontroller. And resets it. After disconnecting two wires (G and D of the transistor) the microcontroller does not reset.
Regards P.
On a sunny day (Wed, 13 Feb 2013 20:51:09 +0100) it happened Piotrne wrote in :
Here is a simplified schematic (hand drawn):
That is a very inductive load. No way will a simpel VDR protect the MOSFET.
What is that transformer to that ioniser, current transformer? Inductance? Termination resistor? Looks all like a disaster to me.
The pulse at the drain will likely also feed through to the gate, and cause gate oxide breakdown. No RF screening? Incoming 12V filtered? No current sensing?
Forget it.
Does keyboard pick up EMF?
I'll add more to this later as real work invades play.
Why don't you consider using three or six triacs in the AC input to the FW bridge? Or even as the bridge? Then you can stay away from the very high DC currents. Also, it would use pulse transformers to drive the gates, giving additional isolation from the HF junk.
Or even better, use an inverter tig welding power source? Most of them have a low voltage control input that allows doing exactly what you are after. Not only that, the pulser device is readily available OTS. Some welding sources include the pulser built in.
Regards, tm
only
Check into battery tab welders.
?-)
From a simplistic point of view, is there some way to move the ioniser to the other electrode? Placed on the FET switch side it will leak HV pulses into the FET no matter what you do.
?-)
Still needs full penetration through the tab, partial penetration into the terminal.
Not exactly easy to control.
--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)
be
Easier than you might think. Non-engineer builders have made reliable ones DIY. It is a variant of a spot welder.
?-)
Hi,
we have solved the problem of overvoltage in the welder. There was a basic mistake in the circuit: a missing diode.
Here is the current version of the circuit:
The diode turns on after turning off the transistor and directs the current to the load.
So - initial attempts to absorb these overvoltage pulses were wrong (they may have several kW!) They should be directed to the load. Now, everything works as it should, even with the ionizer.
Regards P.
Huzzaa! Who would have thought it would be that easy?
?-))
That should work just fine. There are gasless plasma welders using water now:
If our attempts to bring the device to work failed, the next option was to buy a welder with a spot welding function. For example, a Kemppi welder with the MicroTack function. This could be even more expensive than this plasma welder.
Hopefully our welder will work correctly. A few things have to be done: synchronization with a stepper motor moving the electrode, measurement of the current, shaping the I(t) function.
Regards PF
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.