RCDs on PLC inputs

Due to a couple of electric shocks, my company began fitting RCDs to field inputs to PLCs.
Generally, it was one RCD per PLC card with either 8 or 16 inputs, these were either push buttons, limit switches, PE sensors or inductive proximity sensors.
Voltage was 240V AC.

The problem - The RCDs frequently trip out.

Has anyone here fitted RCDs to PLC inputs before, and if so, what was your experience? Are there any common problems that you have to be wary of?

Back ground.
We have an old site with plant dating from the 40's to newly installed systems. All new systems are specified with a control voltage of 24V DC, however all our old plant (about 80%) is wired with our old standard of 240V AC. We have recently had a few incedence of electric shock (Luckyly only minor boots) from operators touching damaged sensors or worn insulation. To be fair to the operators, the damage isn't always obvious, and the contact is not always intentional.

It was deemed too expensive to convert all systems to 24V DC, so as an alternative, it was decided to fit RCDs to the 240V sensors. This job has just begun, and the first RCDs are frequently tripping.

I was brought on board with this proble this morning, but the guys who are handling the installation are on shift and won't be back until Monday.

So far, I have come up with the following possibilities.
1, PLC inputs out of the RCD loop. Enough inputs come on and the current inbalance exceeds 35mA, tripping the circuit.
2, Leakage from the PLC input card.
3, Leakage due to harmonics.

If there is any thing else that I should check, please let me know.

I have to wait until Monday before starting, but it's good to get some advanced ideas.

Thanks,

Doug

Very smart thing to do!...

First off, for those that aren't familiar with the term RCD, it stands for "Residual Current Device". In the US, it's known as a GFCI (Ground Fault Circuit Interrupter) or more commonly, a GFI.

Current out on one leg is SUPPOSED to equal the current on the other leg, so if the RCDs are tripping, current is going somewhere! But, you already knew that...

That "somewhere" is usually ground. Old equipment tends to have ground faults all over the place and it might be difficult to find all of them. You're probably starting out with >30ma of baseline leakage, then when the machine runs, things move around and give you those extra few ma that cause the trip. You need to get that baseline down so that additional minor leakage is tolerated. I think you can rule out idea #1. It should not matter how many inputs are on, the current on both legs should increase the same amount. Although, this might be the cause if idea #2 is true. If each input has a slight leakage to ground, then more inputs = more leakage until you reach the 35ma limit.

I don't know if harmonics can cause current leakage, so someone else will have to help you out on idea #3

Be happy that these RCDs have proven that the equipment IS dangerous. They're just doing their job of protecting!...

1. How have wired Input cards Neutral voltage, is it earted inside the card (as Eric explaind)

2. If the sum of control cables lenght is very long and there are GND (PE potential earth in Europe) wires, that means that big part of current returning by GND wire back to PLC Cubicle or through field devices GND-terminal to graund.

3. Every neutral return wires must wired back through Residual Current Device as Phase (240V) (as Eric explaind)

leakage current

Sorry, must be 3 km and 1.65 km.

Many RCDs are not happy with harmonics. Had an occurance recently on a pool filtration job where there were 2 x 37kW soft starters complete with line and load side filters. RCD tripped out every time a motor started. Used a different typs that was much more immune to harmonics.
Also, watch out that the filters do not filter to ground. You will never get the motor started.
I agree with the premise that old machinery starts to move around and earth leaks can occur from leaky cables. This is probably the major cause of the problem.

Leakage doesn't have to be in old wiring. It is common knowledge in pump station controls and other wastewater and water applications that GFIs will trip frequently. In the high humidity atmosphere there is enough leakage through terminal strips, field connections, and even new wiring insulation to trip the GFI. I have personally observed and verified this in the field.

sometimes depend on your's wiring

I follow the answer above.
and I have some difference idia is..May be induce voltage
Check your's input wiring cable...seperate them line by line

platootod

Thanks for all your responses.
I have now checked the first circuit and found that the neutrals were not segregated.
Basicly, the field devices use the nearest convenient neutral. With this setup the RCD/GFI cannot possibly work. We will be re-wiring them in a few days, and hopefully this will fix the problem.
If not, I now have a good list of things to look for.

Kind Regards,

Doug