1. Home
  2. ⚡ Welding Forums
  3. Durability of the electric arc

Durability of the electric arc

Hello,

I would like to add some kind of control to an experimental welder. The welder should use short pulses of current only.

I have a big 3-phase trafo with a rectifier (90 diodes, total current up to 1000A, voltage without load: 50V). And a ionizer. Everything is connected in such a way, that when I have electrodes spaced by 3 mm and turn on the ionizer for a while, the big current begins to flow (even after turning off the ionizer). The measured current is about 300A.

Now, I would like to add a circuit to control the duration of the pulse - the big current should flow for about 50 ms or less. This should not be too difficult - I have a few MOSFET transitors (IRFP4468PbF) that should be able to work as a switch. But controlling the duration may be not enough. The shape of the pulse should be also controllable.

So, the welder should act as big controllable current source. Assuming, that there is an analogue source of the desired pulse shape, what is the best way of controlling the current? I'd like to avoid a solution where the "redundant" power is dissipated in transistors (300A * 10V would give... a lot).

My question is: is it possible to use PWM in this case? If yes - with which frequency? I have such doubts:

- when the transistor goes "off", will the arc fade out immediately? If yes, PWM is not possible, because the ionizer will be off or will provide pulses every 20 ms only;

- if the frequency required frequency is high (e.g. 50kHz), will it be easy to control such a big current?

I have some experience with PWM, but for much smaller currents (up to 5A).

Regards Peter

Reply to
Piotrne
Loading thread data ...

Yes, it is possible to control the current flow via PWM. You will need to add some inductance to the output line, one switch between the rectifier output + and the inductor, and one switch between the rectifier - and the inductor. One side of the weld circuit is connected to the rectifier - also. You will also need a way to measure the current flow. When you close the switch between the + and the inductor, the current increases linearly, when the current reaches the value that you want, you open the switch between the + and close the switch to the - side of the rectifier. When the current falls below some value, open the

- switch and close the + switch. For switches, several big FETS in paralllel would work. IGBTs are another good choice. The time between opening one switch and closing the other needs to be very short, but always greater than zero!

The frequency will be dependent on the size of your inductor. It will not be "easy" to control welding sized currents, but it certainly can be done.

When the + side switch is opened, and the - side closed, the inductor is delivering the current stored in the magnetic field. The current will fall, but the inductor prevents it from going to 0 instantly.

BobH

Reply to
BobH

Thank you for your response. Do you know similar solutions from practice? What frequencies do they use? I'd like to turn on the current for a short while only (even 50 ms); the PWM frequency should be high enough to get several pulses in this period. But high frequencies cause other problems...

Did you see any non-PWM solutions?

Regards Peter

Reply to
Piotrne

I am not sure what the frequency is. It should be possible to calculate what some of the commercial units are using based on the inductance that they are using. ST sells an eval kit for some of their FET modules that is an arc welder. The documentation on that would be a good source for information. The eval kit only goes to 100 Amps or so.

The older "big iron" welding machines used saturable core reactors or MagAmps for current control. It might be possible to control current with one while you do the pulse timing by switching the AC in via SCRs. You might get bad interactions of the MagAmp with your pulse generation stuff too.

These guys built a TIG welder:

formatting link
there might be some useful ideas in there.

Regards, BobH

Reply to
BobH

PolyTech Forum 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.