In "the lab", did the exact same PC / PLC / sensors get used in the same configuration, like sensor 1 was connected to the same terminals?

What's different?  Is there welding (or induction heating) going on anywhere near the PT100s or PLC or wiring?

engco

I use indusoft to make scada with s71200 1214dcdcdc. I measure temperature using pt100 and RTD modules 8AI. I connected and tested the sensors in lab and it was working good and scada system was working perfect. When I moved to the field and connected the pt100sensors ,the values were changing on scada with random values including the real values.i don't know where the problem. Please help

I figured that the neutral wire is shorted to earth at the field. Could this cause the problem?

I used shielded cable and I connected one of the sensors shield to ground which is neutral but nothing changed. The connection was only from the plc side 

Many OEMs consider grounding shields in the main electrical enclosure as it enters "standard practice".  Many enclosure manufacturers support this with shield bonding accessories.  Many terminal manufacturers support this with multilevel terminals with a ground for the shield.

Many approaches, and all meant to reduce electrical noise.

If you wire a spare PT100 directly to the same terminals (temporarily), does the SCADA system data still jump around?

How often is SCADA updating?

Have you done a PLC trace within TIA Portal to see what the data looks like IN the PLC (to rule out a SCADA configuration problem)?

engco

The only difference is the length of pt100 cable I used longer cable  around 10 meters cable in the field. And the ground of main panel is connected to neutral. This is the only difference between the first run and the second. 

Does the pt100 resistance needs the shield to be connected to ground?