Problem with induced low voltage across 120V line 3

[actnup]

Do to a combination of problems in the field I have as much as 45 volts induced across a 120VAC control circuit. The power wiring (480VAC) is run in the same conduit along with several control loops and even water has been found in part of the underground conduit, but rather than correcting the field conduit problems at this time I need to suppress the induced voltage at the actuator. The induced voltage across the control circuit picks up relays intermittently and is pulsing an actuator motor.
Could someone help me with the best way to lower the induced voltage and where would it be most effective, at the field device connection or within the control building? Thanks for any help you can offer.

 

[peebee]

I'm a little confused by the details of your existing problem -- are you reading a common-mode voltage to ground of 45 volts? Or are you reading 165 volts line-to-neutral?

If this is a common-mode voltage to ground, I'd suspect that the main problem is not induced voltage, but rather a missing neutral-ground bond at the upstream separately-derived system. I'd definitely check that out first. If this is truely an induced voltage problem, a small 120 volt transformer at your load could be used to eliminate common-mode voltage, BUT YOU SHOULD REALIZE THAT THIS WOULD ONLY BE MASKING OTHER POTENTIALLY SERIOUS PROBLEMS.

If instead you're reading 165 volts, line-to-neutral, the only way that induced voltage could cause that situation is if you have not run all circuit conductors together in a common raceway system, which would be a violation of code and for which the easiest, cheapest, and fastest correction would be to properly rewire it.

Provide more installation details if you'd like a more specific answer.

You should keep in mind that "correcting the field conduit problems at this time" could likely be the easiest, cheapest, and fastest way to correct your problems, as well as the only legal and safe way.

 

[actnup]

Thanks peebee for your reply. I am passing info I have received from sales reps in the field which are doing the best they can. The 120VAC supply is a separate source that picks up an interposing relay in the actuator for remote control. They are reading the 45VAC across the L and N without power being supplied from the source. The field techs are telling us it is induced voltage from problems in their conduits but we are being asked to help correct the problem. I agree that the true problem in the wiring should be corrected rather than looking for a way around it.
Again thank you for your help.

 

[peebee]

A small buck transformer would probably do the trick.

 

[peebee]

Actually, if you're looking for very quick and dirty, dropping a small load like a 100w incandescent bulb across the line near the relay might drop your voltage enough. Just keep hanging light bulbs on there till the voltage drops to a reasonable level. Make sure you don't overload your control power source.

I still recommend you get the thing fixed right, though.

How long a wiring run are you talking about here?

 

[busbar]

First, are the "control loops" installed complying to NEC Article 725? That could be the biggest problem.

There are three voltmeter readings necessary to troubleshoot the problem. Besides across the pair of wires, what is the voltage from each conductor to ground? How much do those readings change when paralleled with the 10KΩ resistor?

As for water in raceways—is the wiring suitable for the application at hand? There is almost no way to prevent moisture ingress in an outdoor underground raceway over the life of the installation. In most cases, NEC Chapter 3 requires wiring suitable for wet locations for jacketed or non-jacketed cables.

 

[actnup]

Thank you both for your help on this. I am passing on the questions to the field to get some answers. I have also asked them to check all the control wiring on the conduit for different voltages. I will post the information I get back so we can all know how it turns out.
Thanks again for your help.
actuator

 

[jbartos]

Suggestion: The shielding of those conductors that can have induced higher voltages and cause the nuisance trips is recommended. Another solution is to run the control or sensitive circuits in a different raceway from the power raceway.

 

[jOmega]

Gosh, used to fix this problem by using a tight, twisted pair for the low voltage control ckt. Try to get about 6 - 10 twists per inch. Made the tisted pair from stranded 12 ga.. One end of pair in drill chuck (hand drill) other end held with a pair of vise-grips..... run up the drill and twist 'em up. Original length should be about 20% longer than actual length to allow for pull-up in the twisting... Use twisted pair to replace existing wires.

Used this fix to solve a problem on Dumbo at Disneyland in Calif.... about 20 years ago. Dumbo's still running...

HTH

 

[jbartos]

Suggestion to the previous posting. There may also be a need for shielding. That is hard to implement.

 

[jOmega]

Jbartos:
as an F Y I:

In actual test run at Reliance Electric circa 1970, twisted pair was compared with shielded cable, and with twisted shielded cable.

..... and the winner was: tight twisted pair....

The shield serves as an antenna.... and reduces the effectivity of the twisted pair (do you ground one end?.....if so which end?...do you ground both ends?....What is the consequence?...)

Certainly, reducing the strength of the magnetic flux field to which the wires are subjected, improves attenuation. However, the addition of the shield can exacerbate the situation by acting as an antennae and providing a closer coupling.

The straight twisted pair, when tightly twisted, usually provides sufficient cancellation of the induced fields.

 

[actnup]

I have been passing on the information you have all given and hoefully they will let me know what in the end correctes the problem. I have been told the electrical contractor claims all the wiring is to code and the panel manufacturer claims there control panel is correct so it appears it will be a battle to get the issue corrected.I mentioned the information I passed on was a consensus of several engineers and electrical individuals so it should be weighed heavily. I appreciate all the input. Thank you.

 

[jOmega]

Actuator:

While the wiring may meet code, the code does not address the issue of noise immunity. Mixing power wiring with 120v (and below) control wiring has always been an invitation to cause the type of interaction you stated in your original post. Usually, such issues are addressed by equipment OEM's on their installation drawings. Responsible OEM's will provide wiring do's and don'ts... in the form of notes on the installation drawings or wiring diagrams, etc.

Electricians that are good at bending conduit and pulling wires, and hooking them up are, for the most part, oblivious to issues such as EMI, EMC, RFI, etc.... and noise in general. If a straight line is defined by two points, and you give them the two points, they will go in a straight line. But don't expect them to be able to handle parabolas and hyperbolas. Their education, training, and experience, generally doesn't equip them for such tasks.
But they are exceptional within their limitations.

(Conversely, most graduate engineer are ill prepared to function as electricians).

When the drawings that are given to the electricians to use for the installation contain such notes, they can plan accordingly. i.e., if the drawings say that 120v control wiring should not be run with power wiring or should be run in separate ferrous metal conduits, electricians will heed these notes and plan the installation so as to comply.

These are my opinions, which have been formulated in over 40 years of experience in both domains.

For what its worth...........

 

[jbartos]

Suggestion to Omega (Electrical) Nov 26, 2002 marked ///\\Jbartos: as an F Y I:
In actual test run at Reliance Electric circa 1970, twisted pair was compared with shielded cable, and with twisted shielded cable.
..... and the winner was: tight twisted pair....
///Very impressive.\\The shield serves as an antenna.... and reduces the effectivity of the twisted pair (do you ground one end?
///Recommended up to 1MHz.\\.....if so which end?
///Sending end, not receiving end.\\...do you ground both ends?
///Recommended above 1MHz.\\....What is the consequence?...)
///A reduction of noise in very low power signals.\\\

 

[jOmega]

jbartos:
Appreciate your comments.

We used to ground shields at the drive end. I think it had something to do with the impedance to ground of the shield. IF grounded at the source... then you had the cumulative impedance over the length of the shield that would (as the impedance became significant) support noise. Never did get to evaluate the hows and whys.... but on more than one occassion, a noise problem was solved by moving the ground end from the source to the terminating end (drive)...

Most of the electrical noise encountered over the years wasn't in the HF domain... Grounding at both ends of the shield introduced more noise than it cured. Something about the difference of ground potentials and the impedance in-between.

Have you observed that some Belden Shielded cables also have a drain wire....? Usually see that in cables with a foil shield as opposed to the braided wire shield.

Well, sir, I think we've beaten this one into the "ground"
(pun intended.... no apologies offerd)...

Cheers

 

[jbartos]

Suggestion to the previous posting: Reference:
1. IEEE Std 142-1991 IEEE Recommended Practice for Grounding of Industrial and Commercial Power Systems, (Green Book),
Page 213 Par. 5.7 Grounding of Shields. (Excerpts or Eng-Tips)
1. The overall shield is grounded whenever and wherever possible for low frequency application. (My Remark: This is an obvious intent since the shield replaces Conducive Metallic Conduits, e.g. RGS, EMT, IMC, etc., where there are no major shielding questions.) The individual shields are grounded according to the applied frequency: low frequency at one end and high frequency at multiple points. ......
Shields may be:
1. Braided copper wire
2. Metalized foil with a copper drain wire
3. Metallic conduit (if steel conduit it also serves as a magnetic shield)
4. Other shielding methods
...
For signal frequencies of up to about 1MHz, it is a good practice to ground a shield at only one end, preferably at the source of the signal end leaving the load end insulated from ground. This is to prevent the shield from acting as a conductor for voltage differences at the two ends.

The IEEE Standards are often interested in volunteers; especially, if the volunteers are experts as you are. Visit

http://www.ieee.org

for more info.

 

[jOmega]

jbartos:

You have earned a star.
I am impressed and appreciative of the information from IEEE Std 142-1991. I was not aware of this reference and will order it come Monday.

Kind regards,