EM pickup problem in 10 kHz coil signals when touching metal conduit by hand

[emsensors]

Hi,

I have some instrumentation on very large piece of industrial machinery. The instrumentation drives a coil at 10 kHz A.C. and looks at the resulting induced emf in two secondary coils. The instrument connects to the coil via a 20m cable, a junction box then connects this cable to a shorter cable (1.5 m) to which the coil is attached. The drive side and two secondary coil signals use shielded twisted pair with drain wires. The cable has an overall braid shield. Drain wires and the overall braid are connected to the case and common at the instrument end. Drain wires and the overall braid are shorted in the junction box and connected through to the overall braid and drain wires on the shorter cable. Drain wires and overall braid are open at the coil end. The cables run through metal conduit and finally through steel pipe to the coil. AT no point do the shields or drain wires touch any metal. We are seeing a 'signal response' when the steel pipe is touched by hand. I cannot understand how this can happen since our shields are not touching the metal. Any suggestions?

Thanks,

Marcus.

 

[crutschow]

Is the steel pipe grounded? If not, it should be.

 

[emsensors]

Is the steel pipe grounded? If not, it should be.

Well, it is all part of a very large continuous caster - pretty much everything is steel, it's all connected together and ultimately all grounded. How good that ground is, I don't know for sure.

 

[crutschow]

Well, it is all part of a very large continuous caster - pretty much everything is steel, it's all connected together and ultimately all grounded. How good that ground is, I don't know for sure.

The reason I asked is because, if it were well grounded, touching it should have no effect on the signal inside.

 

[FvM]

Presuming the signal cable shield is connected according to manufacturer instaructions, there may be a problem with ground connection of the sensor shielding respectively case.

LVDT sensor amplifiers can have problems with VFD noise, particularly of they aren't designed in aware of this kind of interferences.