EMI Concerns with Underground Utility Service Feed

Status
Not open for further replies.
S

saluki_engineer

Member
Location
Colorado
I am doing a plant layout that requires a new run of underground utility to a transformer outside the new building. The new utility will need to run parallel to an electrical duct bank containing plant power and communication lines. The contractor has asked me to specify the minimum distance to maintain between the two to minimize EMI.

I have directed them to maintain as much distance as possible for the parallel runs and make sure they intersect at approximately 90 degrees, but I would feel better if someone could point me to a standard that specifies this, so I could give them a minimum distance. I don't see this mentioned in the NEC or the utility's installation requirements and my google search hasn't given me anything useful. Any help would be much appreciated.
 
I haven't found a code reference for separation of underground power and communications lines. Utility may have some standards.

If there are concerns about EMI, use rigid steel conduit.
 
One simple thing you could do is specify that the exception in the NEC allowing single phases per duct in underground not be used. But that is unfortunately a decision the utility will actually make, yes?
Or is this direct burial cable?
For electric field effects (which will be attenuated by earth) a shield on the cable or use of a metallic duct would help.
For magnetic effects, only distance will be practical.

Tapatalk!
 
Thanks for the responses!

Unfortunately, the duct bank is already in the ground with PVC conduit and the utility feed will be direct bury.

I agree the utility will be the one making the final decision since they are installing it. I got an RFI from the contractor asking about this, so I was hoping to reference a standard in my response if possible. Like I said, I didn't see anything in the utility company's standards and requirements, so I may be giving them a call to see what they generally specify.
 
140221-1501 EST

Thin electrostatic shielding, such as what might be called shielded wire (for example Beldfoil), won't do much for low frequency magnetic fields, but is very useful for capacitively coupled noise.

Twisted pair wires are quite useful for the reduction of magnetic fields. Most Beldfoil shielded wires also contain twisted pairs. CAT-5 cable has a different twist pitch for each of the four pairs to reduce mutual coupling between the pairs, and the pitch is quite fine.

.
 
140221-1704 EST

An experiment I ran some years ago with unshielded CAT-5 cable.

The data being sent was 115.2 kBaud using a balanced RS-422 signal. The noise source for the test was a MIG welder welding. It was possible to have error free data transmission when the welding cable was 5 feet or more away from the CAT-5 cable. Were errors caused by magnetic or capacitive coupling I don't know.

Intuitively I would not want to run a shielded audio cable 5 feet from the welding cable.

It is very difficult to answer the original post question. What are the signal levels, and sensitivity to noise?

.
 
Thanks Gar. I ended up specifying that they keep them as far away as space would allow, no closer than 5 ft. Communication is mainly discrete and 4-20mA signals. Also some fire alarm lines. All of the ethernet is over fiber due to the distance, so no Cat-5 to worry about.
 
Status
Not open for further replies.
Top