Cable Pulling 2

[pithpal]

Hi Everybody,

I was working on pulling 600 MCM teck90 armoured cable from basement of a building to 10 th floor through electrical rooms on each floor. The current is 2000 Amp, that means 3 cables/ phase. As I wantd some specs, I spoke to NEXANS and ALCAN.
The ALCAN gus told me that you should go for a Alcan cable and that also three conductor cable, as this will lead to less voltage drop and you can run three cables each three conductor side by side as one has to be very care ful of the phase imbalance which arises due to different lengths of cable. Here is arrangement below:
All phase in one section: Three conductor cable:
O 0 O O
OO OO OO OO

Then I spoke to NEXANS guy, ofcourse both were application engineers not sales guys, he told me as I have three conductors per phase then I should go for single conductor cable and each in a traingular configuration to have less voltage and then he said if you have even no of conductors per phase then you should go for a rectilinear configuration. All factors are attributed due to voltage drop. A B C phase in one section adn neutral in middle.
N N N
O O O
O O O O O O O O O

Now when I explained this story to my big boss, he said we will go in a different way, and that will be like below but in outward direction horizontally.
A B C N
O O O O
O O O O
O O O O

so this is All phase in parallel outward way.

Now here is my question, which arrangement is good and why?

Please advise any free software available where I can test this configuration.

Why cable pulling is an issue and as an engineer how to figure out that I should not leave it to contractor and come out my self with the exact bending configurations.

I appreciate you time always.
Regards,

 

[cuky2000]

The first step is to decide what kind of raceway to use. The NEC and other similar codes require for parallel set of cable equal length and impedance for each set of cable.

You may consider use an underground raceway from the utility to main electrical room with three conduits with three sets of cables in a duct bank arranged in horizontal configuration. Inside the building, you may have the choice selecting conduit or cable tray if there is not construction constrains or code violation.

From the pulling tension point of view, the physical cable arrangement may not be highly significant. However, the ampacity of the cable may be less for the middle set of cables since is exposed to less cooling surface for heat dissipation.
_{For this building, the total feeder length may be estimated within 150 ft (assuming 10 ft per floor+ ~50ft service entrance). In this case, the worst cable arrangement may not produce excessive ampacity reduction or imbalance impedance particularly if there are load connected at every floor.}
See if the enclose information could help.

http://www.ecmweb.com/mag/electric_simple_calculations_cable/index.html

http://www.kerite.com/catalog/catalogfiles/continuous_support_power.htm

http://www.electrician.com/electa1/traywirepull.html

http://www.peakpwr.com/conduit.htm

 

[BJC]

Three 600 MCM cables of anything aren't enough for 2,000 amps. Check out using plugin busyway, cheaper, faster better. GE, SquareD or Cutler Hammer res can educate you.

 

[davidbeach]

As I understand, teck90 would be a Canadian product, somewhat equivalent to Okonite's CLX, so what the rest of this response will be based on CLX and the NEC, since I don't have a copy of the CEC.

I don't know how you hope to get 2000A out of only three conductors per phase. Assuming you could terminate the cables on lugs suitable for use with the 90^oC rating of the cables, which may not be a valid assumption, it would take three sets of 1250kcmil copper conductors to get to 1995A, based on multiconductor cables. At 1995A you would have to adjust the pickup of your breaker to no exceed the cable ampacity, and most likely that would be 95% or 1900A. To actually get to 2000A in this case would require 3 sets of 1500kcmil cables for an ampacity of 2115A. With aluminum you can't get to 2000A with three sets of multiconductor cables; three sets of 2000kcmil (assuming such a thing would even be manufactured) would only be good for 1890A.

It is a bit better if you can use single conductor cables, but then they probably won't be armored and maintaining sufficient spacing will be a problem. 600kcmil copper, at the 75^oC rating (much more likely to find terminations at this rating), is good for 690A per conductor or 2070A for the set, but again that is single conductors with plenty of spacing. Aluminum would need larger conductors.

Personally, I'd stick with cables that have all three phase conductors, one neutral, and one ground (the ground may be subdivided and distributed - that's ok too) within a single armored assembly and use more conductors. I'd go with six sets of 400kcmil if available (that size isn't available in the CLX product line) for 2010A (at 75^oC). Using the sizes available in CLX (and assuming that the neutral is only for any unbalance from the three phase conductors, not as a current carrying conductor) I'd use 8 sets of 250kcmil copper (2040A) or seven sets of 350kcmil copper (2170A) for 75^oC terminations or, if 90^oC terminations are available, 6 sets of 350kcmil copper (2100A) or 5 sets of 500kcmil copper (2150A). Aluminum conductors are not available in the CLX product line, so I won't go into aluminum settings.

In a vertical run of cables, practical constructing considerations suggest that the cables be installed in a flat configuration, with each cable strapped to horizontal struts on 10 foot (3 meter) vertical spacing. Trying to make fancy patterns will be a construction nightmare. If you don't have the width for a single row, your strut frame will have to have a second (or more) layer of strut and cables with each layer installed separately. Hopefully you have discussed this cable run with the structural engineer on the project, you are talking about a lot of load. Of the combinations listed above the 8x250 would have a combined weight of 33.76lb/ft, the 7x350 39.76lb/ft, the 6x350 34.08lb/ft, and the 5x500 39.1 lb/ft. Assuming 15 feet floor-floor, you are talking a rise of close to 165 feet, or a total load ranging from 5,570 lb to 6,560 lb, all without a floor to set it on.

 

[pithpal]

Thanks David beach. This is called electrical engineering, I dont know why most of the engineers dont even think about these small details which could have such drastic effect. Honestly, I need to take a print out of this post and then I will read it and get back to you, I am not sure I will be able to convince my engineer but will give a try for sure. Thanks to cuky 2000 and BJC also.

Regards,

 

[waross]

It sounds like single conductor Teck cable using the free air rating. Feed end must consider magnetically induced currents. This can be controlled by terminating all cables in an Aluminum plate, or making a hacksaw cut between the holes in each group of A-B-C-N.
At the load end, consideration must be given to both magnetic induction and sheath currents. The conventional approach is to terminate the cables in a sheet of insulating material.
An alternative aproach (which I have seen accepted by two different AHJs in Canada) is to use the same method of termination at the load end to avoid magnetic induction. In addition the concentric ground wires and the armour are cut flush with the overall jacket and wrapped with varnished cambric tape. This avoids sheath currents. A dry type connector is used. All connectors and locknuts must be non-magnetic. The only problem was that on one installation the installers disregarded specific instructions regarding the locknuts, and the inspector required the steel locknuts be replaced with non-magnetic locknuts. That blew a lot of the original saving in labour.

 

[pithpal]

Hi Waross,

I am impressed by ur post on transformers. When I read this cable pulling post, I called one of the manufacturers and asked him how cable termination in 600 V system can help get rid of magnetically induced currents. and he said to me he has never heard about it. But he mentioned though it could be a concern in high voltage. I have worked in 8.3 KV
(L-L), we use cable termination kits and magnetically induced currents as part of concentric neutral and there is where this comes in to picture.
I would be grateful if you could elaborate on this topic a little bit.

Regards,

 

[waross]

I am some what embarrassed by part of my post on transformers and have retracted part of it.
There are several posters here that I look up to.
But the problem at hand.
Two problems; Sheath currents and magnetic induction.
Sheath currents are covered in the Canadian Electrical Code. Hopefully some one versed in the NEC will give us the US references.
I won't quote rule numbers, as my code is so far out of date. I will give you the rule title to facilitate your looking it up in the index.
In my old Canadian code book, in Appendix B, Notes On Rules, Section 4- Conductors Rule 4.xxx Sheath Currents in Single-Conductor Metal Sheathed Cables (SEE APPENDIX b)
Paraphrased. When a current flows in a single conductor cable metal sheathed cable a voltage is induced in the sheath.
The magnitude of the induced voltage is dependant on the current (Not the voltage as the rep supposed).
Below 200 amps sheath currents are not usually a problem.
Between 200 amps and 425 amps, the heating effects of circulating sheath currents can be reduced to the point that the cables need not be de-rated by spacing the cables approximately one diameter apart. That is close enough that field cancellation will reduce the induced shield voltage. It is far enough apart to avoid mutual heating of the cables and allow air circulation (in free air rated installations).
For cables carrying currents greater than 425 amps it will generally be necessary to de-rate the cables or to eliminate the sheath currents.
That’s about what the code says. Now how do we eliminate the currents. Note also that some single conductors such as TECK 90 may have a concentric ground. The lower resistance may increase the current if it is allowed to flow.
The conventional scheme is to terminate the cables in an insulating plate such as micarta or fiberglass. The grounds are not connected at the load end, they serve as ground protection for the cable only. A separate, appropriately sized ground cable is installed parallel to the cables for equipment and/or system grounding.
A cheaper alternative in both material and time is to insulate the concentric ground and the sheath. In a dry location the cables are terminated in dry type connectors. The concentric ground wires and the armor are cut back flush with the outer jacket. If the cable end is insulated with a plastic tape, there is a real danger of a part of the armor or a strand of the concentric ground pushing through the tape under the pressure of the connector. If this happens the isolation of the sheath currents is defeated.
It is best to use a non flowing tape such as varnished cambric or fiberglass tape. There is no objection to using plastic over the varnished cambric to hold it in place until it is clamped in the connector.
The second problem with single conductor cables is magnetic induction. When a single current carrying cable passes through a hole in a magnetic material, the current induces a magnetic flux in the material. The changing flux induces eddy currents in the material. There may also be some hysteresis loss. How much of the heating is due to eddy currents and how much due to hysteresis? I don't know. I came to get rid of it, not measure it.
This is covered in the Canadian Electric Code in section 12.xxx Multi- and Single-Conductor Cables
A single conductor cable carrying a current over 200 A shall be run and supported in such a manner that the cable is not encircled by a ferrous material.
Rule 12.xxx Entrance of Conductors into Boxes, Cabinets and Fittings (See Appendix B)
Precautions to be taken include non-ferrous or nonmetallic box connectors, locknuts and bushings, Non-ferrous metal plates or nonmetallic insulating plates.
Appendix B (Notes on rules) Allows cutting a slot between entry holes, in effect making the cable entry one odd shaped hole.
Imagine 16 conductors in a 4 x 4 array. Each horizontal row contains A - B - C - N. It is not required to join all the holes. you can cut from "A" to "B" to "C" to "N" in each row.
Note particullarly the injunction against ferrous fittings, locknuts and bushings. Use diecast connectors and locknuts and diecast or plastic bushings.
This arrangement was allowed by the Canadian Code. Please post a correction if it is no longer approved. What says the NEC?
yours


If the net current in multiple cables is zero, the magnetic fields cancel each other. That is, all the phases and the neutral of a circuit or a group are passing through the same hole.

 

[pithpal]

Hi waross
Thanks for your time for such an exhaustive explanation.
I will get back after reading this post.
In regard to post on transformers,
If we all be perfect than we would not be human beings,
To err is human but still I have learned some thing from your post and other valuable contribution from davidbeach, jraef etc.

Thanks

 

[BJC]

pithal
I may have been wrong. It is possible to do it with three 600s if: you can get the right terminals, if you can maintain the required spacing, if yo don't ahve to derate while going through floors and/or fire stops. To maintain the spacing in "free air" will require cable tray or an engineered support system. Are you going to tap into the cable anywhere except the to floor?
Things like supports, firestops, taps termination boxes, etc. are things that make busway a more economical installation. I am pretty slure sure by the tiem you walk away from the job busway will cost a lot less. I have penciled that one out a dozen times.
Here are the websites I googled up to check things.

http://www.nexansenergy.com/egy/tecdoc/ElectriciansHandbook/elec-hbk.pdf

http://www.nexans.ca/egy/equip/bw/configuration_1C_ACM_db.htm

http://www.sns.gov/constr/convfac/s...onc/stokescatalogs/General/IPCS/TechInfo4.pdf

http://www.acewirecable.com/pdfs/7615.pdf

 

[cuky2000]

Pithpal, the Nexans catalog posted above answer many of your questions. However, still It’s not clear to me why are your concerned about pulling cables as mentioned in your original post. Appears to me that you first need to select the type of raceway and sizing the cable appropriate for the expected load and satisfy the voltage drop requirement.

If there is any underground section, direct buried from the utility to the basement may be an alternative to be considered. In this case, 3 sets of 1000 kcmil @ LF=100% appear to be OK per CEC shown on page 62 of the Nexans catalog. Smaller cable may be required if the LF < 100%.

On the other hand, ampacity of cable in air inside the building is typically higher than cables for UG installation. However, two different cable sizes in small job sometime may not be practical or cost effective unless the TECK cable is not a special order readily available in stock

 

[resqcapt19]

Personally, I'd stick with cables that have all three phase conductors, one neutral, and one ground (the ground may be subdivided and distributed - that's ok too) within a single armored assembly and use more conductors.

Look out for 250.122(F) if you use cables.
Don

 

[davidbeach]

250.122(F)(2) is not a particularly difficult requirement to comply with. You're right, it does need to be considered, but it certainly isn't a reason not to use cables.

 

[BJC]

"250.122(F)(2) is not a particularly difficult requirement to comply with."
True enough, but I have seen people get in trouble with it. (Actually 122.122 A). The grounding conductors in type C-L-X for example is good for the ampacity of the cable but not necessarly good for paralled cables. I bears checking. When cables are paralled the ampacity of the circuit goes up along with the size of the ground conductor. In some cases it's not big enough to meet requirements of table 250.122.

 

[resqcapt19]

David,
How do you comply with it, without using a custom cable or are you suggesting the use of ground fault protection for the cables? Is you are suggesting ground fault protection, is there any that complies with 250.122(F)(2)(3)? A 2000 amp circuit requires a 250kcmil EGC in each cable or raceway.
Don

 

[davidbeach]

250.122(A) requires the use of table 250.122 based on the overcurrent device protecting the circuit. That's fine and clear.

When conductors are paralleled, the usual method of compliance, and the only method of compliance for individual conductors in raceway, is through 250.122(F)(1) which requires each equipment ground conductor to be full sized based on the table, no allowance for the presence of other ground conductors in parallel.

With multi-conductor cables, 250.122(F)(2) opens another route to compliance. With ground fault protection provided for the circuit, and the other requirements met, the process basically turns around. The sized of the equipment ground conductor now sets the rating of the ground fault setting. So a run of parallel CLX cables in the 250 or 350kcmil sizes, with a 3AWG ground, would be required to be protected by a ground fault device set at not more than 400A. In the example that started all of this, the ground fault setting of the 2000A breaker protecting the circuit could not be set more than 400A if 250 or 350kcmil cables are used. Upping the cable to 500kcmil, with its 2AWG ground, would allow the ground setting to be increased to 500A.

 

[davidbeach]

resqcapt19, your 16:08 post must have appeared while I was working on my 16:14 post. Frankly I don't know what devices may or may not comply with 250.122(F)(2)(3), but I've not had any trouble with an AHJ when using the rest of 250.122(F)(2).

 

[resqcapt19]

David,
It is my understanding that there are no ground fault devices "listed" for that purpose. I have no idea why a special listing would be required, and I see no technical reason for such a listing, but the code rule says that it must be listed for the purpose. The listing issue was the main reason I brought 250.122(F) into this thread.
Don

 

[pithpal]

Hi Everybody,
I guess where I work, ten decisions are made in one day and the size of cable changed from 600 mcm, 3 cable/ phase to 750 mcm 3 cable/ phase and then today 1000 mcm, 2 cable / phase any ways, some times one has to bear with this stuff.

My question is as it will teck 90 cable with armour, free flowing through electrical rooms to 10 th floor, now one manufacturer told me when ur using a single conductor cable, then the clamp has to be non-ferrous. Is he correct.

If yes then if we ground the armour of teck cable both at source as well load, will this problem disappear?

If he is correct then from my basic understanding as an engineer, Is it also not allowed to run a single conductor cable through EMT as then it will form a transformer, so how do we solve this.

My next question, when I visited the site today, there were two new capacitor banks from ABB, and they were suppplied from a 400 A 600 V disconnect but I am not sure it was fused or not. Should I presume that fancy looking cover of 400 kVAR bank had a MCCB built in which was not visible or it was just a fused disconnect that was being used.

Now we want to change that fused disconnect in a MCCB so that we could remove it and use a switchboard where we could hook our new breaker also and the breaker for Capacitor bank also. Do you while selecting breaker,
a phenomena, i read during my studies capacitor switching comes in to picture. Does this has any effect on any selection of protection, if not where this phenomena should be considered.
I am sure I asked many question, but few words for each will save from a headche.

best Regards,

 

[resqcapt19]

You cannot run a single conductor of an AC circuit through any ferous metal. Serious heating and conductor failure will soon result if you do.
Don