Protecting NI-9219 DAQ from HF interference during TIG welding

[WeldMatrix]

Hello everyone,

I am using an NI-9219 module in a cDAQ chassis to acquire temperature data from K-type thermocouples during an automated TIG welding process.

The thermocouples are spot welded directly onto the workpiece and are located very close to the weld path (approximately 10–15 mm from the arc). The welding machine uses high-frequency (HF) arc starting.

The problem is that whenever the HF start is activated, the NI-9219 either:

• stops acquiring data,
• disconnects from the computer, or
• in some cases appears to fail completely.

I suspect that the HF ignition pulse is coupling into the thermocouple wires and reaching the DAQ inputs.

I would appreciate advice from anyone who has successfully measured thermocouple temperatures during HF TIG welding.

Specifically, I would like to know:

1. What is the best way to protect the NI-9219 from HF ignition?
2. Has anyone used ferrite cores successfully on thermocouple cables? If so, should both thermocouple wires be passed together through the same ferrite core, and how many turns are recommended?
3. Would a shielded thermocouple extension cable significantly reduce the interference?
4. Is an external RC filter or surge protection recommended before the NI-9219 inputs?
5. Are isolated thermocouple conditioners necessary for this application?
6. Has anyone successfully used the NI-9219 for temperature measurements during HF TIG welding without damaging the DAQ?

Any recommendations regarding grounding, cable routing, shielding, or practical protection methods would be greatly appreciated.

Thank you.

[Henrik_Volkers]

Some hints:

Twist you TC, you don't want any loop area near the welding. 

Additional shielding (not the hole length , to reduce capacitive coupling) near the workpiece can held, grounding on one side , here I would try it first near the grounded workpiece. You can try a copper tube as shield (ok, not connected near the welding erea, you don't want additional thermal mass 😉 ) just well grounded at the workpiece (resp to the ground connection welder-workpiece ). 

Add the ferrit core , windings usually as much as you can get in the core, but I would try windings  around 3/4 of the core with some distance (one TC diameter) to reduce capacitive coupling in the wires  (Nice donut core, not a split one)    😉 

Add a capacitor (10nF to 470nF)  at the 9219 input  with short wires.

The grounding of your workpiece migth need some (RF) improvement. Keep the ground connection to your welder as short as possible, no windings and the best connection to the workpiece you can make (clean surfaces, good pressure).

The RF will look for an easy way to get to ground, resp. back to its source in the welder.  Usually the welders have a RF filter in the power supply line , but take a look at all wires to the welder and look for additional RF paths provided.