Welding work at machines during machine power ON

[AB2005]

Hi;

Its recommended that whenever a welding work is require at any machine where inverter drives, PLCs etc are present, switch off the power of machine before welding. It may be possible that the sensitive electrical things can damage during welding work.

I want to know the technical reasons behind it. Has anyone faced such kind of electronics failure during welding work?

As now a days electronics welding machines are being used, is the risk of damage (if presented with conventional welding machines) of electronic things still there if welding work is done at machine during power is ON?

I do agree that machine power should be off during any welding work but sometime due to some reasons, machine's power can not be off.

 

[GaryS]

I don’t know where you got you information from but I can tell you that it just not true
I have multiple welding machines running using VFD’s and PLC’s with touch screen controls
All run with no problems. They all run the VFD’s while welding in fact you can’t run the welds without the VFD’s running.
One that been running about 10 years now, has 2 Lincoln 400Amp Submerged Arc welders
Multiple VFD’s for travel and head position. Controlled with a PLC. It is capable of running 2 continuous weld beads over 200 Ft, the bridge span on that machine is 25 Ft wide

 

[AB2005]

Hi;


Recently I personally faced an issue at one of flexo printing machine. A welding work was carried out at folding beams without power off. Just after welding, two motorized linear actuators malfunctioned. Their motors were OK but electronics kit failed.

 

[parky]

Unless the plant it'self is the welding plant (which will be designed in such a way that there is no possibility of ground spikes) then it should be isolated.
I have seen on a number of occasions where contractors welding on plant have caused sensors (especially PT100's) to be destroyed, this is usually because while welding the welder earth is not clamped directly on the piece being welded, even seen where a welder welding on a metal bench where there was a wet tiled floor, there were sparks tracking from the legs to the wet floor it took out two PT100's on the machine located within a metre or so.
Often engineers welding on plant fail to move the earth clamp from the last weld position to the next so the current has to travel longer distances, should the pipework not be connected to the main frame of the machine large currents & voltage spikes can be introduced. Electronics are more likely to survive if not powered up but as a general rule switch off equipment isolate as much equipment as possible where the welds are going to be done & instruct the people weldng to ensure the welding earth clamp is securley clamped as close to the parts being welded, tig or mig seems to be less prone to causing large spikes but it is better to be safe than sorry.

 

[Peter Nachtwey]

I don’t know where you got you information from but I can tell you that it just not true
I have multiple welding machines running using VFD’s and PLC’s with touch screen controls
All run with no problems. They all run the VFD’s while welding in fact you can’t run the welds without the VFD’s running.
One that been running about 10 years now, has 2 Lincoln 400Amp Submerged Arc welders
Multiple VFD’s for travel and head position. Controlled with a PLC. It is capable of running 2 continuous weld beads over 200 Ft, the bridge span on that machine is 25 Ft wide

VFDs are noise spike generators themselves! PLCs are tested in the labs of the respective PLC manufacturer. I have seen them do the "gutter' test where cables leading to and from the PLC AND our motion controller are put in a gutter and a 3000V noise spike is induced. The equipment can shut down but can't be damaged and should run after a noise spike. I have been to the Rockwell and Modicon testing locations and watched them do these tests.

I don't remember going to Johnson City, TN for doing any testing on the old TI505 boards but we had little fuses that would blow and protect the circuitry. The fuses were easy to replace.

I doubt a commercial printer goes through these tests.

Turn off the electronics! My company makes hydraulic servo controllers. Although our controllers are relatively robust, the Temposonic and Balluff rods are not. We have seen too many get fried. Any electronics with long cables are susceptible to damage.

We recommended disconnecting the wires if they can.

 

[I_Automation]

I worked in a stamping plant that for a year almost strange things blew up in control panels. Servo drives, power supplies, VFD's, etc.


After a year the maintenance dep't welder was pushing his portable MIG welder out of the area and said out loud "I wonder what will blow up this time?


I stopped him and asked him what he meant and he said for about the past year whenever he is welding at a machine (with the power off to it) a machine nearby had something blow up.


I sent his welder out for testing and it was sending weld current through the power cord, back to the switchgear and through neighboring machines.

 

[GaryS]

I just had to repost here to try and clear thing up a bit.
I see one common thing with all the post’s about the problems. I am surprised that others don’t see it
The person doing the welding tried to short cut the job and failed to locate the weld return lead in the proper place.
The return / ground lead must be connected to the work as close to the weld joint as possible
Yes return is the correct phrasing some welders are isolated and then you have the reverse polarity
I know of one case where a welder was hired to repair a casing in a 30ft pit where the pump was located
The power to the pump control panel was disconnected
The welder short cut the job and connected the return lean to a pipe coming out of the pit
Part way through the job the pump motor and control panel burned up they were just lucky the pump house didn’t burn down. It turned out there was a vibration coupler in the pipe so the weld current found a path through the motor connection / conduit and went through the panel. They all had to be replaced control panel, motor, wiring even some of the service.
And before you ask the welder was a portable engine driven unit mounted on a truck 100% isolated from the equipment. Electric current is a funny thing it will bind a path even if it’s not the one you want.
Most welders are current regulated so even with a high resistance path the welder regulater will just increase the voltage ( to the limits) until the requested current is supplied
I build and hooked up many welding machines if they are done right there is no problems.
While you are welding some interments may show bad reading due to the magnetic and electrical fields in play.
As a side note here, I have my own welder and do my own welding repairs

 

[parky]

@ Gary, yes tht is what I posted, in saying that, never in all my years have I seen (well not definitive) that welding has caused a PLC fail, but have seen PT100, Sensor converters (like PT100/4-20ma) and a number of industial switches/modems & other electronic equipment suffer.

 

[Jim3846]

My main concern with welding is the return path for the ground. When the ground is further away from the welding machine, this usually introduces problems with equipment between the ground connection and the welding machine. I think GaryS above had an example of this issue.

 

[I_Automation]

My main concern with welding is the return path for the ground. When the ground is further away from the welding machine, this usually introduces problems with equipment between the ground connection and the welding machine. I think GaryS above had an example of this issue.

Plus I think GIT just posted on another thread a month ago about a PLC with values going wild so he put in an alarm. When it went off he found the welder up off the floor with his ground cable connected to the wire trough.