0
\$\begingroup\$

I would like to know if using two 2.5mm² cables has a greater current rating than one 6mm² cable. The ratio of CSA is 2.4x and so I would have expected to need three 2.5mm² cables. However, I then looked at the actual current ratings tables for different cable sizes and see that 2.5mm² is rated at 18.5A while 6mm² is only at 32A. Hence presumably two 2.5mm² cables would have a joint rating of 37A and so exceed the 6mm² cable.

I have read other posts that the deciding factor is the amount of copper, which would suggest that three 2.5mm² cables are needed.

However based on the current rating tables, only two cables are needed.

So I don't understand why the tables contradict the advice in posts on this site. I would like to know if I have understood the tables correctly, or if my logic is flawed.

Also, my cable is single strand. Should I use multiple strand, as this seems to be used in the tables?

Source: http://www.diydoctor.org.uk/projects/cablesizes.htm

\$\endgroup\$
  • \$\begingroup\$ Often such tables take into account the surface area that is available for heat dissipation. \$\endgroup\$ – PlasmaHH Dec 11 '16 at 22:16
  • 1
    \$\begingroup\$ The surface area is 2piRL. So that makes sense. eg 2.5mm2 and 6mm2 have ratio of radii of 1.58/2.45 so that two 2.5mm2 cables will have significantly greater surface area than one 6mm2 and so better heat dissipation I think. That said as Vofa says below its safer to work with one larger cable in case of a break, ie loss of power rather than a fire if I understand him correctly. \$\endgroup\$ – rupert Dec 12 '16 at 10:55
  • \$\begingroup\$ Often current rating is limited by voltage drop rather than insulation rating. For the latter, I use the aircraft wiring charts which include cable bundle # and loading, wire size, insulation rating and altitude. \$\endgroup\$ – Spehro Pefhany Dec 12 '16 at 18:19
3
\$\begingroup\$

When comparing single to multiple cables current capability you need to consider that multiple cables will run hotter (for the same current capability) than does a single cable in free air. As soon as you see multiple cable bundles, you will see the current rating being reduced because of thermal restrictions.

Get a good Cable chart like this ...Engineering Toolbox is your friend. And a cable areas conversion like this

If you are in EU then I'm sure there are similar cable company or website that cover mm cabling.

Adding this link to an excellent treatise on the de-rating of bundled cables due to increase in thermal resistivity.

\$\endgroup\$
  • \$\begingroup\$ Thanks on the first table the current ratings are given by the number of cores from single up to 43+ cores. The rating decreases for a greater number of cores for the same cross-sectional area. So AWG14 ie 2.1mm2 for single then rating is 24A but for 43+ cores its only 7.5A. This suggests its much better to use single core. From basics though I cannot see why the rating should drop given same CSA? \$\endgroup\$ – rupert Dec 12 '16 at 7:32
  • \$\begingroup\$ @Rupert. It's a thermal restriction. Assuming that the frequency is low enough to not have skin effect, then a bundle of cables has much poorer thermal transfer from the center to the edge than a single cable. If the cable bundle is of insulated cables the situation is even worse. \$\endgroup\$ – Jack Creasey Dec 12 '16 at 18:34
3
\$\begingroup\$

There is not really such a thing as current rating for wire. The practical limit is wire temperature. At a certain point, the insulation will melt or burn. Extreme overloads can melt the wire, but that's a different problem entirely. Without knowing your application, or how the wires will be bundled and routed, it's not possible to say what wire gauge is appropriate. In certain applications, such as residential wiring, tables exist in regulations to remove guesswork and ensure that no installed wires will get hot enough to start fires (if properly installed). Wire sizing in regulations is typically extremely conservative.

One conductor is preferable to two for power transmission. Think about this: you have power transmitted over two wires in parallel. One wire comes loose. Now the current is only flowing through one wire, and it may get dangerously hot. If you only use one conductor, the system will fail entirely if a connection opens. This is a safe design practice.

You can ensure that wire will never get dangerously hot by protecting it with a fuse or circuit breaker. The size of the protective device will vary depending on the application, available cooling, and normal load current.

Solid-core wiring and stranded wiring will do the same thing, but stranded wire is much easier to work with. Solid-core wire is more susceptible to breaking when flexed.

\$\endgroup\$
  • \$\begingroup\$ Many thanks. So I understand the reason is heat dissipation. Two 1mm2 cables lose heat faster than one 2mm2 cable and so can handle more current. However, one cable is safer per above. I am wiring a cooker which said to use 6mm2 but I only had 2.5mm2. I'll buy some 6mm2 and put that in. I had a 4 slot toaster and dishwasher on a 2.5mm2 and the cable melted. I think I'll try to replace that cable as well but am held back by having to damage walls to feed cables. \$\endgroup\$ – rupert Dec 12 '16 at 7:45
  • \$\begingroup\$ Careful. Heat is generated due to the wire's resistance. 1mm^2 cables have a higher resistance, and more heat will be generated in them than in a 2mm^2 for the same amount of current. It's better if the wire never gets hot, rather than that if it is sufficiently cooled. If you're doing residential wiring, always follow the applicable codes and get it inspected. If a fire starts from wiring that you installed, you're responsible. \$\endgroup\$ – vofa Dec 12 '16 at 15:45
  • \$\begingroup\$ @vofa. If the same current travels over two wires (far enough apart for no thermal interaction) as opposed to one, where does this extra heat come from? eg. 10 Amps over two 1 mm^2 will produce exactly the same heat loss as 10 Amps through a single 2 mm^2 wire. Ohms Law, right? ....the problem comes when wires are bundled and the inner surface of the wires no longer have the same thermal resistance to ambient. This is what causes the bundle to be de-rated. ..ieee802.org/3/poep_study/public/sep05/walling_2_0905.pdf \$\endgroup\$ – Jack Creasey Dec 12 '16 at 22:21
  • \$\begingroup\$ @JackCreasey P = I^2 * R. Thinner wire = more R. For single conductors of equivalent length, 1mm^2 wire will dissipate more heat for the same amperage because its resistance is higher. I was considering single conductors only, not paralleled conductors. I think you are right that two 1mm^2 wires will dissipate the same heat as a single 2mm^2 wire for a given current and identical length, but it is dangerous design practice to rely on parallel conductors for reasons specified in my post. \$\endgroup\$ – vofa Dec 12 '16 at 22:31
  • \$\begingroup\$ @vofa. Almost universally high current cables are made of multiple conductors. The reason for this is cables are designed to provide at least some flexibility. The more flexibility required the more conductors used. The only place you see single copper/Al conductors used is in house wiring where the cables are supported and don't move after installation and in industrial situations with mineral insulated copper covered cable (inflexible by definition). In sub-station wiring (out of scope for the question) where you don't need any flexibility you typically use copper or Aluminum bus bars. \$\endgroup\$ – Jack Creasey Dec 13 '16 at 4:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.