Electrical shock during TIG welding
December 7, 2018 1:53 PM Subscribe
Under what circumstances would contact with the work lead clamp of a TIG welder cause an electrical shock? I know a person who received a pretty significant shock this way, enough that he needed medical treatment; I've been told that it's an inevitable in TIG welding environments. Is this correct? Is there a way to eliminate the risk? Can someone explain to me why the electricity would even pass through a person who touches the work lead? That doesn't make sense to me at all.
Response by poster: Is it fair to say that if the work piece (a valve cluster) were properly grounded, the ground clamp would not be dangerous to touch?
posted by jwhite1979 at 2:31 PM on December 7, 2018
posted by jwhite1979 at 2:31 PM on December 7, 2018
While you were actually welding? Probably, but human's are big bags of salty water so I would try not to. Also, read this here: "Welding Circuit Shock Hazards
Utilizing proper grounding in the welding environment is a good practice, but it does not remove all possibility of electrical shock. The welding circuit is energized by welding voltage. A person will receive a shock if they become the electrical path across the welding circuit. Precautions must be taken to insulate the welder from the welding circuit. Use dry insulating gloves and other insulating means. Also maintain insulation on weld cables, electrode holders, guns and torches to provide protection.
Similarly, electric shock originating from the electrical supply system can be prevented. Proper maintenance of electrical equipment and extension cords will insulate the welder from electrical sources."
posted by Brockles at 4:26 PM on December 7, 2018 [1 favorite]
Utilizing proper grounding in the welding environment is a good practice, but it does not remove all possibility of electrical shock. The welding circuit is energized by welding voltage. A person will receive a shock if they become the electrical path across the welding circuit. Precautions must be taken to insulate the welder from the welding circuit. Use dry insulating gloves and other insulating means. Also maintain insulation on weld cables, electrode holders, guns and torches to provide protection.
Similarly, electric shock originating from the electrical supply system can be prevented. Proper maintenance of electrical equipment and extension cords will insulate the welder from electrical sources."
posted by Brockles at 4:26 PM on December 7, 2018 [1 favorite]
The ground clamp (as it is commonly called, but properly a "work lead clamp") should be safe to touch, but that assumes it's actually grounded. Basically you need to make sure that the welding machine has a good ground, that its internal ground connections are correct (probably, if the machine is working, this is fine), and that the welding table or workpiece is also independently grounded (this tends to be the issue). I think this last bit is probably why your friend got zapped.
If the welding table or workpiece (wherever you have your 'ground' / work clamp) isn't grounded to
earth / 0V, and the ground via the machine isn't very good (or doesn't exist at all), then the workpiece is "floating" with respect to ground. The machine only knows that the potential between the TIG torch (or "stinger" in the case of stick welding) and the ground lead is some number of volts. Thus when you touch the ground lead or welding table, it's possible to get a shock.
If everything is set up the way the manufacturer tells you to, you shouldn't get shocked by touching the work clamp. (The TIG torch or stick "stinger", on the other hand... well, there's a reason why it's called a "stinger".) But a lot of times, due to the weird ways 240V connections work in the US, with some 3-wire and some 4-wire, and lots of shoddy adapters and cables... it's easy to get a lifted ground.
posted by Kadin2048 at 5:46 PM on December 7, 2018 [1 favorite]
If the welding table or workpiece (wherever you have your 'ground' / work clamp) isn't grounded to
earth / 0V, and the ground via the machine isn't very good (or doesn't exist at all), then the workpiece is "floating" with respect to ground. The machine only knows that the potential between the TIG torch (or "stinger" in the case of stick welding) and the ground lead is some number of volts. Thus when you touch the ground lead or welding table, it's possible to get a shock.
If everything is set up the way the manufacturer tells you to, you shouldn't get shocked by touching the work clamp. (The TIG torch or stick "stinger", on the other hand... well, there's a reason why it's called a "stinger".) But a lot of times, due to the weird ways 240V connections work in the US, with some 3-wire and some 4-wire, and lots of shoddy adapters and cables... it's easy to get a lifted ground.
posted by Kadin2048 at 5:46 PM on December 7, 2018 [1 favorite]
I'd be checking the polarity/ground for sure I got zapped good by a mis-wired welding setup years ago. but also don't be grabbing stuff bare handed while operating welding equipment.
posted by patnok at 6:56 PM on December 7, 2018
posted by patnok at 6:56 PM on December 7, 2018
A long time ago a friend who was responsible for maintaining old IBM equipment (system 34s) at a big welding shop told me about crashing problems they had there which nothing could seem to fix, including wholesale replacement of all affected machines.
Finally, someone got around to testing all the grounds of the various outlets the 34s and terminals were plugged into, and found that there was as much as a 15V potential difference between the grounds of outlets in the same big room where all the computers were, and it was determined that current flowing between the grounds -- ground loops -- flowing along coax cables connecting two or more machines plugged into different outlets was causing the crashes.
And yet there was nothing obviously wrong with the building wiring.
They concluded that the grounding system of the building simply could not handle the vast amounts of current generated by the welding machines (up to 200+ amps per rig) when the shop was busy, and that was causing 'hot' grounds.
If a similar phenomenon was in play where your friend was working, that might account for his shock, because he could have been connected to a ground at a lower potential than the ground clamp on the work piece.
If that was the situation with your friend, I'd say the shop was to blame for the shock.
posted by jamjam at 11:11 PM on December 7, 2018 [1 favorite]
Finally, someone got around to testing all the grounds of the various outlets the 34s and terminals were plugged into, and found that there was as much as a 15V potential difference between the grounds of outlets in the same big room where all the computers were, and it was determined that current flowing between the grounds -- ground loops -- flowing along coax cables connecting two or more machines plugged into different outlets was causing the crashes.
And yet there was nothing obviously wrong with the building wiring.
They concluded that the grounding system of the building simply could not handle the vast amounts of current generated by the welding machines (up to 200+ amps per rig) when the shop was busy, and that was causing 'hot' grounds.
If a similar phenomenon was in play where your friend was working, that might account for his shock, because he could have been connected to a ground at a lower potential than the ground clamp on the work piece.
If that was the situation with your friend, I'd say the shop was to blame for the shock.
posted by jamjam at 11:11 PM on December 7, 2018 [1 favorite]
Another circumstance that would make it absolutely unsafe to touch the earthing clamp would be the existence of simultaneous accidental contact with the hot side, because that would put the welding supply's full output voltage across the body.
You might not even notice that such a contact existed while the weld was actually in progress, because the voltage across the arc is quite low (around 20V) but as soon as the arc broke, you'd see the full open-circuit voltage which is usually around 80V: plenty to deliver a nasty shock even without particularly sweaty hands.
It's also remarkably easy to forget, in a moment of inattentiveness, that the tricity is still lurking in the wires even when the arc isn't lit up.
posted by flabdablet at 6:38 AM on December 8, 2018
You might not even notice that such a contact existed while the weld was actually in progress, because the voltage across the arc is quite low (around 20V) but as soon as the arc broke, you'd see the full open-circuit voltage which is usually around 80V: plenty to deliver a nasty shock even without particularly sweaty hands.
It's also remarkably easy to forget, in a moment of inattentiveness, that the tricity is still lurking in the wires even when the arc isn't lit up.
posted by flabdablet at 6:38 AM on December 8, 2018
Grounding issues can be hard to debug. First, there's just plain-old lifted grounds, one where there's no ground connection at all. These are easy to discover with a multimeter, but only if you're looking for them. If you have (say) a welding machine previously wired up for 3-wire 240V split phase, where the GND and NEU are tied together somewhere inside the box, you might not ever think to look for it, and then you plug in a machine where those two wires are not tied together, and one of them is open, and you could have a bad time.
Then there are high-resistance grounds. These will look OK to a multimeter when nothing is powered up, but they aren't good low-resistance paths to ground. So if you dump a bunch of electricity into them—this is probably what happened with the IBM mainframes and welding machines—and you get a significant voltage potential between one circuit's "ground" and another.
There's even a more interesting failure mode that crops up with high-frequency TIG, which is you have a high-impedance (but low DC resistance) ground, that looks fine at DC but has increasingly high resistance as you go up in frequency. Normally you get this because of coils and other weird geometry in the wire, or wrapping the ground lead around something. I have heard, anecdotally, about getting zapped if you have your work lead wrapped around the leg of a welding table—it will act like an iron/air core choke—and when your machine goes to do its HF "arc start" (basically a burst of high frequency electricity to punch through the air and ionize the air between the TIG electrode and the workpiece, starting the arc) you get zapped. And HF shocks hurt.
You would need some reasonably decent test equipment to really evaluate whether a ground is "good" in all these situations. In most big welding shops, especially in welding schools where they know students will be doing dumb stuff, they punt on this by just having really massive ground connections. Like a big copper bus bar running along the wall, into which everything that can be grounded is tied, and then which is earthed very nearby into a bunch of rods driven into the ground. But copper is expensive, and most small shops aren't going to do something like that, even if it's probably a good idea. AFAIK, no separate grounding network is required by NEC. (In fact NEC gives some weird exceptions for welding machines, because they behave strangely compared to other types of loads.)
posted by Kadin2048 at 1:03 PM on December 8, 2018
Then there are high-resistance grounds. These will look OK to a multimeter when nothing is powered up, but they aren't good low-resistance paths to ground. So if you dump a bunch of electricity into them—this is probably what happened with the IBM mainframes and welding machines—and you get a significant voltage potential between one circuit's "ground" and another.
There's even a more interesting failure mode that crops up with high-frequency TIG, which is you have a high-impedance (but low DC resistance) ground, that looks fine at DC but has increasingly high resistance as you go up in frequency. Normally you get this because of coils and other weird geometry in the wire, or wrapping the ground lead around something. I have heard, anecdotally, about getting zapped if you have your work lead wrapped around the leg of a welding table—it will act like an iron/air core choke—and when your machine goes to do its HF "arc start" (basically a burst of high frequency electricity to punch through the air and ionize the air between the TIG electrode and the workpiece, starting the arc) you get zapped. And HF shocks hurt.
You would need some reasonably decent test equipment to really evaluate whether a ground is "good" in all these situations. In most big welding shops, especially in welding schools where they know students will be doing dumb stuff, they punt on this by just having really massive ground connections. Like a big copper bus bar running along the wall, into which everything that can be grounded is tied, and then which is earthed very nearby into a bunch of rods driven into the ground. But copper is expensive, and most small shops aren't going to do something like that, even if it's probably a good idea. AFAIK, no separate grounding network is required by NEC. (In fact NEC gives some weird exceptions for welding machines, because they behave strangely compared to other types of loads.)
posted by Kadin2048 at 1:03 PM on December 8, 2018
« Older Looking for a simple contact management app for... | What are some sounds that English speakers have... Newer »
This thread is closed to new comments.
You eliminate the risk by how you set up for the job - properly earthing the work piece and trying not to lean on the workpiece with any part of you that is unprotected.
posted by Brockles at 2:17 PM on December 7, 2018 [2 favorites]