Proximity sensor false triggers AFTER grounding it's shielding.

[Nealieboyee]

Hi All,
But of a strange one here. I've been getting false triggers from my proximity sensors. But only when their shielding is grounded to earth. Removing the earth solves the problem immediately. Weird

Specs:
48V Toroidal PSU for stepper drivers. . Earthed. Positive and common/negative floating.
24V PSU. Earthed
5V PSU. Earthed
All cables are shielded and all shields are earthed at one point, along with every other earth.

24V and 5V power supplies share a common/negative.
48V toroidal PSU does not share anything.
The problem happens whether I turn off the 48v supply or not.
Anything glaringly obvious I might have missed?

Thanks
Neal

[Doddy]

You don't mention: Is the machine that the prox-sensor is connected to grounded?


Entertaining read: https://www.emcstandards.co.uk/cable-shield-grounded-at-one-end-only

[Muzzer]

Yes, this screen grounding business really exercises a lot of self professed ex spurts. As with most things in life, there is no absolute right or wrong method. There are some cases where grounding one end is best and other where both ends is best. You keep both as options in your toolbox when you tackle EMC issues.

Where a lot of these people get ahead of themselves is by forgetting that there is no hard "ground" from an EMC point of view. When you have to test and certify an installation, the limits apply all the way up to 30MHz, at which point the distances on a large system start to approach the wavelength of the measurement frequency. The screens may now start to form a part of the resonant network that is causing some of the problem. I speak as someone who has had to get reasonably high power (100kW) motor / dyno installations systems to meet the EMC limits.