Plasma Cutter

[Excavatoree]

(ExEE longwinded warning)

Summary:  Does anyone know about Plasma cutters and can explain why four out of 12 wires in a cable assy failed without visible damage to the insulation?

Had a new experience working with a company formed by some former co-workers at the excavator plant.   They do metal fabrication and welding.  My former boss called me and asked if I would look at their plasma cutter lead.   At the time I knew nothing about plasma cutters, and I still find it difficult to wade through the marketing wank and get real information.

He called me when an interlock stopped working.  When a part of the torch is installed, it closes a switch to enable operation.  That wire was open.  After checking both connectors (hoping for an easy solution)  we realized the long (6-12 feet, or 1-3 meters) cable assy had a broken wire in it.  This was an otherwise ordinary multi conductor cable (about 12 22 AWG, .3 sq mm)  with an air hose inside it.   I cut into the cable and got lucky - I found the bad spot in the interlock wire.   

I've seen burned wires in cable assys and harnesses before - usually the insulation melts and or burns and it's visibly damaged.  This wire looked perfect, but for about 1-2 mm, there was no copper inside the wire.  It was a plastic tube.   I cut and stripped the end for soldering in a new wire, and did a quick check to the other end to be sure there were no other breaks.   "While I'm there..." I decided to check the other wires.   There are 7 wires spliced together at the torch that go to   7 connector terminals at the other end.  Three of them were broken in the same way.  (Fortunately, in the same area that I had already cut open.)   I replaced these as well.  The guy using it says it cuts better than it ever has, leading me to believe that the wires failed one after the other, slowly.

I did learn that plasma cutters operate at about 100-200 volts and about 30 amperes.   The seven 22 AWG wires seem to be adequate for that.   I just can't figure out what happened and why, and if it's doomed to happen again.  Can anyone offer any insight?

My thought is that if it fails again, I'll run a 12 or 10 AWG (3-5 sq mm)  wire alongside the cable, if it is not objectionable to the operator, and consider if the unused connector terminal positions can be used to add more parallel wires going into the machine.

OF course, I'm hindered by not knowing the model of the machine.  I'll have to ask.   

EDIT:  It's a Miller 875 rated for 60 Amperes at 140V, but max open circuit voltage is 400V.    Duty cycle is supposed to be 40 percent at 60 amperes, but we know how people are when there's a job that needs to be done.  I saw no  mention is made of the machine failing to operate if duty cycle is exceeded or if the unit overheats.

manual with wiring diagram, if anyone is interested.  (open wire 15 stopped it from working.   I see the diagram only shows 5 operating current wires (1-5)  two more were added, spliced together using only one connector position.)  This indicates they know they have a problem and have made at least one revision to try to stop it.
http://eagleplasma.com/cutters/manual-miller875-autoline.pdf

[Samogon]

Looks more like physical damage to me



PS Stupid iphone throws wrong words, making me look like i have dyslexia

[Excavatoree]

Looks mole like physical damage to me

I agree, but there was no visible sign of any physical damage on the cable cover or the insulation of the individual wires.  That's what I don't understand - what could have damaged the copper inside without showing any sign of damage to the insulation?

[Samogon]

You mentioned, that there is hose to supply gas for plasma. What psi rate it has? Is it possible that they exceeded psi rate for hose?
I can allow thought that at some conditions it has expanded hose and while insulating material pretty flexible, copper wire could brake.
And can you tell how wires tightened to the enclosure or hose? Or the freely laying there?

[Excavatoree]

You mentioned, that there is hose to supply gas for plasma. What psi rate it has? Is it possible that they exceeded psi rate for hose?
I can allow thought that at some conditions it has expanded hose and while insulating material pretty flexible, copper wire could brake.
And can you tell how wires tightened to the enclosure or hose? Or the freely laying there?

Gas pressure is 90-120 PSI.   In this case it's just air, but some use pure nitrogen.   The cable looks like a typical multi conductor cable, just with the plastic hose inside.  Thick outer layer, rope like cushioning, non-woven cloth like wrapping.  The wires wrap around the hose.  I suppose the hose expanding and contracting could cause a fatigue failure right in one spot - the four wires were physically together.   No chance of me finding out exactly what that hose.  Also, I'm not sure if there's a valve at the torch end.  If it's open all the time, there wouldn't be as much pressure difference and thus not as much flexing.

 When I repaired it there are splices, so even thought it's warped, it isn't as tight and my wires aren't twisted around the hose but laying beside it.  Maybe that will help. 

[cs.dk]

Usually there is a valve in the machine. Not in the torch it self.
You feed the machine with shop air 8-10 bar, and reduce it down to ~4-5 bar for cutting and cooling..

[Samogon]

I wonder how did you find broken spot? Did you use toner, or there ware visible signs. May be they twisted it couple times with air running.

[Excavatoree]

I wonder how did you find broken spot? Did you use toner, or there ware visible signs. May be they twisted it couple times with air running.

I'd really like to say I used some scientific method or a piece of equipment, but I'll be honest:  It was dumb luck.   My first thought was to cut into the middle of the 7 meter (or so) cable and find which half, then determine which quarter,  then which eighth and so on and so on until I found it.  However, there was a report on a machinist forum that the breaks are near the torch.  So I guessed about 100mm from the torch, cut the cover, and extended the cut to find the blue wire.  (interlock wire) When I found it and pulled it a bit, I felt the hollow section.

[Samogon]

Luck chooses worthy man.

[tautech]

I wonder how did you find broken spot? Did you use toner, or there ware visible signs. May be they twisted it couple times with air running.

I'd really like to say I used some scientific method or a piece of equipment, but I'll be honest:  It was dumb luck.   My first thought was to cut into the middle of the 7 meter (or so) cable and find which half, then determine which quarter,  then which eighth and so on and so on until I found it.  However, there was a report on a machinist forum that the breaks are near the torch.  So I guessed about 100mm from the torch, cut the cover, and extended the cut to find the blue wire.  (interlock wire) When I found it and pulled it a bit, I felt the hollow section.

As another member stated before......fatigue.
Without suitable management of the everyday bending that it's subjected to it will happen again in time.
Better quality cabling and some better strain relief from bending you will get the cabling to last much longer than the original.

[EPTech]

Hi There,

Is there a strain relief in the torch cable,like a kevlar strand or steel cable? If not and the design relies solely on the current carrying wires and switch wires and the gas hose inside, then that is the problem. If the problem were current capacity you should see some heat damage on the wires. If you can, you could try to pull an additional steel or kevlar strand that is shorter than any of the wires through the sleeve. Or teel the workers not to get the feet caught on the cables. 

Greetings,
Pascal.

[TerraHertz]

It's just bending fatigue. I've seen wires do this a few times. When a cable is flexed over and over, the copper strands start to work harden. Once a strand or two have fractured, the others come under more strain at that point, so that's where they break too. And because your cable has the airhose at the center and the wires spiraling around it, the copper strands are placed under tension half the time with random flexing. Hence the ends pull apart a little.

If the manufacturer wasn't cheap-arse, they'd have used cable with very many extremely fine strands, like this:
https://www.aliexpress.com/item/16AWG-Black-5M-Flexible-Silicone-Wire-Cable-Soft-High-Temperature-Tinned-copper-UL-VW-1/32431498039.html

Btw, you can find breaks in unshielded cables by putting a signal (say 10KHz since, 10Vp-p) into one end of the known broken wire, then running a scope probe tip along the outside of the cable. Especially with a bit of metal foil or something attached to the probe tip, at high sensitivity there's enough capacitive pickup to detect the signal. Where the wire break is, it will stop.

[Excavatoree]

There is a strain relief spring, but I'm not sure how effective it is.  As the poster in the machinist forum said, the cable is pretty tight, which leads me to believe it flexes in one spot.  The repaired cable isn't as tight, plus my new wires are not twisted around the hose.    The manufacturer obviously was trying to use several smaller wires to help, but these wires have the larger strands, not the very fine strands someone mentioned.   

When I was working at the bus company, customers would specify (arc) welding wire for battery cables - designed to be flexed, and designed to survive on a shop floor with oil, grease, etc.   Funny, the spec. would pass through many people and often got corrupted to "welded wire battery cables."

I haven't seen the torch put together, but should it need repair again I'll encourage them to go with a large wire of very fine stranding run alongside the hose, or at the lease, I'll run smaller wires along the area that is flexed, near the torch.

I thank everyone for their insights.   

[tautech]

A trick I have used in the past is to extend and taper the strain relief to spread the bending forces over a larger/longer radius. Quite easily done with staggered layers of shrink sleeving.