PLC controlled equipment and welding

[bigbuckaroo]

Hi all. I am looking for some feedback on what your company policy is with regards to welding on equipment controlled with a PLC. My company recently had an outside contractor in our facility doing some extensive stainless pipe welding. They decided to weld a bracket on a piece of equipment that was running production. This was an ice cream freezer with an older Siemens S5 PLC and the machine stopped making a quality product as it would not power the suction valve to cool the product. We spent some time with the OEM working on loading programs and parameters back into the thing but we lost 1 full shift of production that was in process plus another 2 full days where it could not be scheduled to use. i am not a welder nor do i no much about it. Is it reasonable to assume the welding process caused the problem? This is our thought but how do we prove this? Thanks.

 

[gclshortt]

http://cr4.globalspec.com/thread/25231/Effect-of-welding-work-near-PCs-and-PLCs
Here is a thread on welding near PLCs.
In our company we have a hot work permit that is filled out at the time. This includes welding / grinding etc. The project engineer/maintenance manger must sign off on this, in order to perform the work.

Regards,

 

[Firejo]

I’m not a welder and haven’t done any electrical welding since high school (many years ago) but I would never weld on anything that has active (or even powered up) electronics connected to the same ground. Even with cars you should disconnect the battery and ground the cars chassis to earth ground before welding. Two things to consider, one, you’re sending a lot of current through the metal and you’re relying on a ground connection between the item being welded and the welding machine that may or may not be a good connection. Shift the ground plane and you run the risk that some of that (or all of that) current will find a path to ground through the electronic equipment and bye bye tiny components. Second, the welder is producing a lot of EM noise and this noise can be powerful enough to damage electronic components all by itself. We’re talking about components that are very sensitive so much so that they usually are handled with special care to protect against static discharge (which is a lot less powerful than a typical welder). Having said that I would expect a PLC to be fairly well protected against that sort of damage given the intention of where they are used.
I don’t know what all is involved in getting a welder’s certificate and/or degree but I would expect that this sort of thing is part of the curriculum and a pretty good reason to use someone who does have a certificate or degree.

 

[Ken Roach]

Sometimes the EMF fields from the welder are going to be too much for the electronics no matter what you do, but it's a good practice to shut down the control system and carefully bring it back up.

My limited experience says that it's the placement of the ground clamp that causes the most major damage. I had a welder choose the wrong conveyor frame and he ended up putting the full current of the welder through the DH+ braided shield for a few milliseconds before it vaporized.

 

[GaryS]

Keep the Ground lead / Return lead as close to the weld joint as possible.
It’s never a bad idea to turn things off and isolate them if possible but sometimes that’s not an option.
Keep in mind electricity will seek the path of least resistance. And that can happen weather it’s turned off or not. That path could be through your PLC modules if you’re not careful.
I have built many machines using a PLC and VFD’s to control the welder.
All the machines have been working without problems for well over 5 years.
To my knowledge they have never had to replace any of the components.
Just to clarify they are running 2 400 amp MIG welders per machine.
Robot welders are common in almost all industries with many hours of operation without problems.
Pay close attention to the ground connection and current path.

 

[bigbuckaroo]

Thanks for the replies so far. The fact that this was running in production is what causes me concern. Comments regarding the location of the ground I think are important for this case as well. The contractor actually did two welds on this piece of equipment, one on each end. The thought is he did not move the ground to the other side of the machine closer to where he was welding. I am no expert with welding so to me it seems to make sense that the ground clamp should have been moved. More importantly it should not have been in produciton and powered down. My job now is to convince the higher ups that this needs to be standard practice.

 

[Firejo]

GaryS has a good point. Shutting the equipment down does little to protect circuits. Removing the power connection from the electronic equipment should be what’s done. Removing the power connection makes it less likely that the stray current can find ground through the equipment.

 

[Aabeck]

I worked in a factory that occasionally had a machine in various parts of the plant do catastrophic things. Then the welder finally asked why whenever something happened he was in that area welding. Turns out his portable MIG welder was sending weld current back through the power cables, to nearby machines and then to his ground.

 

[willxfmr]

What I have always done is to power down the equipment, and disconnect all the ground connections. Usually I can just un-mount the ground bar. So far I have never had an issue.

 

[just the cowboy]

It was a no way thing

When I worked in a large plant if welding was to be done to the "FRAME" of the machine we would shut down and pull the swing arms and processor cards out. We left no other option for the welder due to memory loss or damage.

 

[NetNathan]

Sometimes the EMF fields from the welder are going to be too much for the electronics no matter what you do, but it's a good practice to shut down the control system and carefully bring it back up.

My limited experience says that it's the placement of the ground clamp that causes the most major damage. I had a welder choose the wrong conveyor frame and he ended up putting the full current of the welder through the DH+ braided shield for a few milliseconds before it vaporized.

How well the weld cable is grounded is VERY important.

I have seen a lot of ground wires melted due to a bad ground during welding. Nothing worse than having a ground wire toast in a conduit. More than once, I have had to remove all the wire from a conduit for the purpose of removing the melted ground wire (14 AWG has a hard time handling weld current) that also melted itself to the control wiring......arghhh.

I do not allow spring clamps for ground at all. I actually had my company remove all spring type ground clamps from the shop floor.
ONLY a bolted ground clamp on a cleaned metal surface is allowed. Do not clamp on a painted or dirty surface.
The ground connection is also required to be as close as possible to the welding to be done.
Absolutely no rotating components between ground and weld area.... I can't count how many bearings I have seen destroyed because of this.

You also want to make sure he weld leads are not near any electronics....due to the VERY strong EMF field created.

 

[bigbuckaroo]

Thanks all for the replies. My goal is to try to change the thinking at my facility to take situations such as this into account. All too often there is a clash between production and the need to get work/projects done. The result is short cuts are taken and things like affecting a machine while producing product happen. Now someone is tasked with trying to pin the lost production on the welding contractor when in my opinion some of the blame should be taken by my company for poor planning.

 

[RussB]

Sometimes the EMF fields from the welder are going to be too much for the electronics no matter what you do, but it's a good practice to shut down the control system and carefully bring it back up.

My limited experience says that it's the placement of the ground clamp that causes the most major damage. I had a welder choose the wrong conveyor frame and he ended up putting the full current of the welder through the DH+ braided shield for a few milliseconds before it vaporized.

I have seen a similar incident using a 1.5" conduit with about 50 or so wires in it. It lasted a few seconds before the insulation on the wires in the conduit flamed up and spewed smoke out both ends. It only damaged a few other components.

 

[Highland Controls]

Just yesterday a customer called and sent pictures. Machine was down, and a 20 gauge ground wire melted all the insulation off. Only power supplying that part of the machine was 2.5A 24VDC(fused at 2A). I told them it had to be something external. Today he calls and said someone connected a welder to the machine yesterday. Took out a motion controller and a touchscreen(at least).

 

[Firejo]

Thanks all for the replies. My goal is to try to change the thinking at my facility to take situations such as this into account. All too often there is a clash between production and the need to get work/projects done. The result is short cuts are taken and things like affecting a machine while producing product happen. Now someone is tasked with trying to pin the lost production on the welding contractor when in my opinion some of the blame should be taken by my company for poor planning.

Simply point out the time and eventually cost (of parts and time) of repairing the units compared to the cost and time to take the unit off line and prep it for proper welding. If they don’t see it they probably never will.