Supply Installation Problem

I don't quite understand your cable arrangement. AWG vs mm^2. (Canada, eh?)

However, running cables in parallel is quite common. Take care they are all the same length. Easier to bend etc.

CSA Table 1 and 3 give copper and Aluminum wire in free air.

Table 5A corrects for ambient.

Table 5B corrects for 2 to 4 conductors present and in contact.

Table 5C corrects for bundles of wire in raceway.

Your local code books should give you the derating factors.

In general, where parallel conductors are used you protect the group as if it is one large (derated) conductor.

Glenn

What means 14 x 25 mm^2?

Is your service a single - phase? e.i. 2 x (7 conductors in parallel)? or three - phase with smaller neutral? Or other combination?

In USA is used AWG system, Square millimeters indicate out of USA application so 1.25 factor may not be used. It could be more questions e.g. selectiveness of protections.

Each question must be very clear so you would avoid misleading tips.

Is this a UK installation? - if so BS 7671 does provide factors for derating parallel running of cables but this is extreme.

Generally speaking - provided all the cables are the same type, length, etc. the laws of parallel resistance will dictate that each cable carries an equal load, so the 800A circuit breaker would cover that.

The individual cables would need to satisfy the fault current rating of the 800A circuit breaker which I fear would not be the case.

That said the proposal may be OK in terms of a theoretical exercise, but is not a very practical application and suggests a misguided theoretical attempt to address volt drop.

If I read you correctly, you've got 1400A. of cable and 800A. protection here.
(Capacities of cables in parallel are added together.)
The cable is probably oversized for voltage drop and/or temperature considerations.
(For multiple cables in a single duct the later can be quite a large factor).

If your concern is that failure of one or more cables might overload the remaining sets,
there are cable limiter lug / monitoring systems that are available, but mostly
they've been found not worth the costs except in exceptional cases.

provide me the data of the cable ie, %resistance in ohms per meter and % reactance if fault current excides above the rated breaker then there is protection for cable, if there is earth fault there is protection and phase to phase there is no protection , that to weather it is alluminium armed cable are copper cable ? M.Yogindranath

You are right. In the USA, the National Electrical Code requires that paralleled conductors be a minimum of 1/0 AWG (approx. 54 mm-sq) each--see section 310.4. The conductors you mention are equal to about a #3 AWG--far too small to be permitted in parallel. To ensure that the current is evenly split among all the conductors, the code requires that they all be the same--in construction, type, length, termination, and routing. In practice this requires considerable care, and is not easy to do.

Reject it, tag it as unsafe, throw it out. John M--master electrician.

I agree that the installation is unusual, but how long has it been installed & working?

There is an old dictum; "If it works, don't fix it.", that might apply, (provided there are no local code issues.)

It might be advisable to check the run to see if there was some physical condition that dictated the odd cable configuration.

You say: There is an old dictum; "If it works, don't fix it.", that might apply. . . I believe that the situation is better described as "an accident waiting to happen". You parallel conductors because larger conductors have more trouble radiating their built-up heat than smaller ones, because the very large conduits and conductors are harder to handle and take more space to install and terminate, because you can't buy conductors rated for the circuit's current (unless you go to busbar). The fairly small sized conductors described may work very well for a very long time, but I would not want to sign-off on such an installation. I have seen an installation where minor differences in terminating parallel conductors resulted in significant differences in the currents they carried--the repair of this put the main building of a hospital on back-up power for one entire shift.

Thanks--John M.

Canada is the same for 1/0 AWG as minimum in parallel.

There is a small section on #14 AWG in parallel, but the protection must be for a single strand. Basically it allows for panelboard wiring of devices with home runs from external devices.

I am still not clear what the original question's cable was. Do I understand it as 14 parallelled cables with each conductor 25mm^2 per phase. That would seem a bit extreme and creative.

Does anyone have European to North American AWG cable descriptions?

That was how I interpreted the post, as 14 paralleled cables, each having four conductors of 25 sq-mm cross section. In the 2002 National Electrical Code, table 8 in Chapter 9 gives the following (I will list only some of the sizes)

AWG Sq-mm Circular Mils

14 ---2.08 ----4110

12 ---3.31 ----6530

10 ---5.26 ---10380

8 ----8.37 ---16510

6 ----13.3 ---26240

4 ----21.15 --41740

2 ----33.6 ---66360

1/0 --53.5 --105600

4/0 -107.2 --211600

A #3 AWG conductor is 26.7 mm-sq.

Hope this helps--JMM

Hmmm,

I wonder where the original author is?

Sorry for taking so long to reply. I was away for some time and sincerely appreciate your comments.

These 14 cables are indeed 4 core each and 25mm² copper per conductor. The figure I mentioned of a current carrying capacity of 100A per phase should make it unnecessary to have to convert it back to AWG, but with the handy table supplied, that should no longer be a problem.

My concern is that, when running 14 such (exactly similar cables) in parallel, the 800A circuit breaker may not offer protection to an individual cable.

Being bundled together on a cable tray, failure of one cable due to external influence or manufacturing fault, is not going to be covered by such an enormous circuit breaker and chances are that a blowout is going to damage adjacent cables until the fault level goes high enough to trip the 800A breaker.

Like someone mentioned, It is also highly unlikely the the kA rating of the individual cables will be an even close match to that of the circuit breaker.

This is a new installation and I have in more than thirty years of experience never come across such a controversial situation. So controversial that there is no statutory reference or ruling to this.

My concern is strong technical/design orientated. I mean one could for example run 800 x 1A rated cables on that circuit breaker if ther are no limits.

Remember these are individually separately insulated cables, each prone to external damage, not bare copper wires in a single strand, twisted together and in contact with each other.

I assume you are not in North America.

Here it is clearly covered by code. Minimum cable size to parallel is 1/0.

Where flexibility is a problem I have used braided straps with heat shrink or with higher temperatures glass with silicone sleeving. There are also other equipment wires that have much finer stranding. (The traditional trade cable is usually a compact stranded hard drawn copper that takes bending tools to even approach a minimum bend---or a gorilla of an installer).

Oh dear - I was assuming he was in North America and hopng he is not in the UK!

I did not think it relevant at first but if it is going to be of any value to you, I'm from South Africa.

We converted to the SI units in the nineteen sixties somewhere and it still fascinates me how economical giants like the US and others can in this day and age still hobble along on units to which they don't even mutually agree. I know that this is besides the point but I await with much interest the day when the whole world has moved away from 'hours on horseback' or 'spitting distance'.

Don't think I'm ignorant about your systems as I grew up and was schooled under the British imperial system. I am lucky enough to be fully 'bilingual' and have no problem converting kg to pounds or millimeters to thousands of an inch on the fly. I was however privileged to be able to experience the advantages of SI units above the Imperial system.

I have since been in contact with the cable manufacturing company and the local Bureau of Standards and I've sent them some photographs. They seem to agree with me and are very much interested in what seems to be a loophole in the regulations here.

If you people are interested in the outcome, I can report back as soon as there is more light shed on the subject. Thanks for your support anyway.

I'll be interested. It seems that in South Africa, as in the USA, the codes are driven by experience. Measurement system use is curious, as you point out. The USA first investigated switching to the metric system over 200 years ago, when it was mostly a French idea. I suspect this was partly because we had just concluded a war of independence from the English with the help and cooperation of the French. At that time, French was the international language for diplomacy and commerce--literally the "Lingua Franca"--and the USA was still (seen as) a backwoods and unlettered intrusion. Unfortunately we struggle along with dual systems in use (even within the same machine), and many people are unable to work with any of them.

As promised, I'd like to report back on this subject.

The last word was from the cable manufacturers, Aberdare Power Cables. Their guy emphatically declares that using a bigger protection unit on cables in parallel is not on. There is no advantage to be gained by using cables in parallel as you are not allowed to increase the short circuit protection value to more than the load bearing capacity of one cable. (The properties of cables in parallel must of course be identical.)

Chances are better of coming off even worse when bundling derating factors have to be applied. So far I could not find any derating tables for parallel conductors but educated opinion is that it should be 0.5 for two cables, 0.33 for three, 0.25 for four etc.

Are there any official derating tables for the permissible loading of parallel conductors available in your countries?

Sorry for my delay in replying to your last question. In the USA, the nearly universal electrical code is the "National Electrical Code", sponsored by a non-governmental organization, the National Fire Protection Association. The NFPA is made up of manufacturers, technicians, engineers, insurance people, and many other types of people all sharing a concern for reducing the risk of fires and injury from the use of electricity (and more recently other and related risks). The association began over 100 years ago and is going strong. They have a cycle of reviewing current injury and fire reports, new technology, trends in the industry, and suggestions or comments they have received, then drafting all this into reports which are voted into a new "model" code every 3 years or so. This then is available for any interested governmental agency to adopt (with various amendments or exceptions being frequently used). This is about the only law I know of which is not drawn up by lawyers or politicians.

It is a minimum standard, not an instruction book. It is a little like the ISO 9000 approach, because it doesn't tell you how to do the work, just tells you what the finished work should be like.

In the code, as in real life, the main concern is heat. Directly, it can cause fires. Indirectly, it damages insulation and makes future catastrophe's more likely. Therefore, the code says quite a bit about installations which can cause excessive heat. The more cables or conductors in a single bundle or wire (likewise the more cores in a single cable), the less current each one can be allowed to carry. Thus, 4-7 means a reduction to 80%, and so on as the count increases. The reason for this is the ability of the heat to be dissipated to the surrounding environment. There are similar rules for reducing the allowed current because of the temperature rating of the insulation versus the local temperature the cable or wire is in. If your paralleled cables or conduits are separated from each other by a few diameters, they are considered to be separate, and there is no reduction in their individual current carrying ratings because of their use in parallel.

Other sections of the code give exceptions to the general rule that a conductor is supposed to be protected against overloads at its source, according to the capacity of that conductor or cable. These sections permit taps of different lengths, depending on the type of installation, the equipment connected, etc. Others recognize that a larger overcurrent protection is needed handle the inrush current of a motor's starting. For conductors or cables in parallel, the main rule is that they must be a fairly large size (much larger than the size mentioned in the question starting this post), and installed in such a way that the current is equally split among all the cables in parallel. Because such an installation is usually not disturbed, and is normally in a facility with more competent maintenance personnel, there is no requirement for individual protection for each cable, based on its own current carrying capacity.

However, many fuse manufacturers sell "cable limiters" which are fast-acting fuses with a crimp ferrule on one end, designed to be installed on each end of each conductor in a set of parallel feeders. These would then act to isolate a faulted wire or cable from the rest, limiting the amount of damage and the risk of a catastrophic fire or other hazard.

Finally, the NEC also specifies in its opening sections, that all equipment and materials must be installed in accordance to the manufacturer's instructions and the approval the equipment has received from the listing agency (such as CSA, UL, FM, and many others throughout the world). Thus, for you, if the manufacturer says this is not a proper use of its cable, then the installation is automatically illegal, and must be corrected. Yes, in theory it would carry the load without problems, but in practice there is sufficient risk that the manufacturer does not want to suggest it (for fear of liability, or because it has not been tested for safety under the proposed use) or because some problem has arisen in the past.

Thanks--John M.