Post History

The two parts of your question

> I believe arcing requires high voltage.

> Then how is arc welding done with about 50 volts?

match these two things which must be done,

1. strike the arc, to ionise enough gas in the area that it will conduct
2. sustain the arc, by carrying enough current to keep the gas ionised, to deliver heat

The striking of the arc in modern equipment is high voltage, but it used to be done by metal to metal contact.

# Modern TIG welding
## Strike

Older welders were "scratch start" or "lift start". Some stick or MIG welding machines are still like that, and some stick welding humans prefer it.

The arc is initiated with metal-to-metal contact so it is a momentary short circuit, then the power supply has to raise the voltage rapidly as the short is broken. I have no experience of using those.

The welding tip is kept energised and will supply current as soon as it touches ground. I think in lift start it's not the full welding current until just after the short is broken, else the tip fuses into the workpiece and has to be ground off.

Video showing the practical aspects: https://www.youtube.com/watch?v=Asb-8_eK6fA

It doesn't explain, electrically, the difference between the scratch and lift. I suspect it's down to smarts in the welding power supply.

## HF start

High frequency start is much easier to use and avoids contamination of the weld by touching it with the tungsten. (don't dip it in the pool)

One disadvantage is electrical interference. Another is greater risk of electrical shock to the operator - it can hurt.

There are plenty of safety hazards when welding: electrical, chemical and thermal. Watch https://www.youtube.com/watch?v=2hAJJky4KAQ and do your research!

The TIG welder I have is documented as 13.5kV, frequency unspecified but probably ~200kHz. The tip has no voltage until activated by button or foot pedal, but then it will jump 10~15mm.

There are multiple things helping the arc to strike but the first two are the important ones here

1. high voltage, even with low current.  At DC it needs around 4kV/mm to strike but it depends greatly on gas components, pressure and electrode shape.
2. higher frequencies jump further. _Question for another day_
3. yes, keep the tungsten tip sharp, so the electric field is concentrated; but actually when welding aluminium the tungsten tip will rapidly ball up, so there's no point (!). The HF start easily overcomes the roundness.
4. modern tungsten electrodes may have various rare earth additives which are expected to increase the ease of arc starting, identified by paint colour on the blunt end _e.g._ "purple tungsten", but I am unable to tell any difference
5. when the tip is still glowing from previous use, I expect thermionic emission may be easier?

Note that the "high frequency" of HF start is not directly related to the much lower frequency (40 ~ 200Hz and asymmetric in current and time) AC used in aluminium welding, which is used to strip the Al₂O₃ layer and expose the bare metal. My TIG machine has both if requested.

## Sustain

Once the arc is struck the power supply must maintain enough voltage to pass the intended current - that is what the feedback loop tries to hold constant. If you let the arc get longer, the voltage has to go up. Eventually the power supply can't or won't, and the arc dies.

The power supply will maintain the arc as length varies, but to produce a good weld arc length should be only a few millimetres. This is about heat penetration into the workpiece and "weld pool dynamics".

For TIG welding below about 10 Amps you may need specialised kit that can re-engage the HF start after the arc dies? Mine won't do that.

Once the arc is sustained, then heat is delivered. Current may be ramped up over a second or delivered promptly, depending on machine capability and settings. Approximately: 100 Amps × 20 volts = 2kW delivered to a (5mm)³ volume = white heat and lots of UV radiation.

# "with about 50 volts?"

## MIG
The very cheap "gasless MIG" machine I started with is just a carefully designed transformer with settings "high" and "low", and wire spool that feeds at constant speed when you press the button. The wire is maybe 0.3mm diameter and contains a flux core which vaporises during welding to make a gas shield.

The open circuit voltage is around 50V but this drops once the welding wire touches the work piece. When current flows, the transformer output voltage drops because of the load.

The wire is both electrical conductor and weld filler, and better machines have settings to control the feed rate so you can balance metal input with heat input.

There may be a mixture of metal-to-metal contact and gas arc as the weld progresses, and some of that metal may be solid or liquid.

It's cheap, robust and messy.

## Stick
Stick welding has a much thicker filler rod, 1~3mm and the flux coating is on the outside. Both are consumed during the weld.

I believe older/cheaper stick welding power supplies are similar transformers, just without the wire feeder. Modern stick welders use similar electronic controls to TIG power supplies and that may be partly for control and reliability, and partly because the silicon costs for switching end up being lower than the copper costs of the large transformer.

The even cheaper version is stick welding from lead acid traction batteries. I think it has been done but I expect there are many additional hazards, and more skill will be required to produce a strong weld.