induced voltage on control wires 1

[pygmy_threat]

I read a post about this from 2009. however, I am very weak in my electrical engineering...

I have a motor run indication to the DCS that is showing 65VAC when the wires (on both ends) are not connected to anything. the wires ohm out fine so no short. I definitely have induced voltage. I will use equipment names to help differentiate motors...

The DCS sends 24VDC through these wires to a switch in the motor contactor (Q1-801). When the contactor closes, the circuit is made and my DCS shows a run (Q-801). However, when a certain different motor (Q1-880B) runs, it causes induced voltage in my (q1-801) run indication wires and my DCS "sees" a closed circuit (it shows the motor to be running). We disconnected the run indication for Q1-801 from the motor contactor, we turned off the power, but Q1-801 still shows a run.

After reading the other post, it looks like 65VAC is a normal induced voltage. How do we trace where this induction is happening? Is it happening in the "home-run" bundle. I think it is a 16pair cable. Each pair is not shielded, but the entire bundle is shielded.

I think I understand from a previous thread that if you use un matched pairs, you increase the likelihood of inducing current. Is that true? why 65V? there must be some kind of calculation to figure out where this voltage is coming from....

if anyone has the old square"D" paper from the 2009 thread I would really appreciate a copy.

 

[pygmy_threat]

thread I was referring to is thread238-250435

 

[rcw retired EE]

Try loading the DCS input with a 10K or 20K ohm resistor across the input terminals. That may be enough to swamp the induced noise voltage and allow the contact to properly actuate the input. Check your DCS power supply first to verify the 1- 2.5 mA continuous draw will not cause any problems.

Calculating the induced voltage is difficult. Estimating the coupling coefficient is just one challenge. IMHO calculating the induced voltage doesn't make the solution any easier, or give an insight to a solution. Typical solutions are:
1.Replace the signal with balanced shielded pair cables or a multi-pair, twisted pair, individually shielded cable:
2. Run separate single pair shielded cable.
3.Verify your shield is only grounded at one point.
4.Put the signal pair in steel conduit.
5.Don't route the control cable so close to the motor power circuits.
6. Load the signal to swap the noise.


Try putting a 1.5K ohm resistor across your meter leads when you are measuring the noise voltage on the disconnected wire. Usually the source driving the noise has a very high impedance and the voltage drops to 0 when the wire is loaded. Since the noise appears to be 65V AC, it is probably not coming from any 24 VDC DCS signals in the control cable. DC can't induce a steady AC noise.

Are motors' power cables close to the control cable? The motor current could be inducing the noise. Is the overall shield grounded at both ends or just one end? Current flowing end to end in the shield can induce noise. Don't waste time calculating, try things until a solution is found.

Good luck!

 

[itsmoked]

rcw certainly hit all the salient points.

Forget about finding "the leak", it's not actually a leak of any kind it's a coupling as your signal leads are acting like the winds of a transformer and "capturing signal" on the switch line you're trying to look at with the DCS.

As was pointed out, don't run signals next to power - it causes this and more problems. They should be routed in separate steel conduit or alternatively as far apart as possible if not in conduit or in plastic conduit.

Load the signal wire at the DCS terminals with resistance just as rcw suggests and your false positives will go away.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[HamburgerHelper]

Using twisted pair helps and try to keep the termination of the wires tight. Bigger the area is between your twisted pair and the termination, the more interference you'll have. Don't run power cables next to control cables.

------------------------------------------------------------------------------------------
If you can't explain it to a six year old, you don't understand it yourself.