0
\$\begingroup\$

I am using 10 AWG wire to power a BLDC motor, and these wires can get quite hot. When an IC is intended for high power uses, it is common practice to put a heat sink on it, but what about for wires? Should I simply use a thicker gauge wire?

\$\endgroup\$
  • 3
    \$\begingroup\$ The question in the title is different from the question in the question. \$\endgroup\$ – immibis Sep 6 '18 at 5:49
  • 3
    \$\begingroup\$ HINT: The insulation of a wire depends on whether it burns or just gets hot. Cheap PVC insulation melts and burns at a low temperature. THHN wire tolerates more heat per given wire gauge. With Teflon insulation you might be able to double the amp rating of the wire, but a lot of other factors come into play. \$\endgroup\$ – Sparky256 Sep 6 '18 at 6:00
  • 2
    \$\begingroup\$ Any heat in the cable is wasted energy... \$\endgroup\$ – Solar Mike Sep 6 '18 at 6:01
  • \$\begingroup\$ Are the wires getting hot along their entire length or just at the connectors? \$\endgroup\$ – Andrew Morton Sep 6 '18 at 13:16
3
\$\begingroup\$

How hot should wires get?

How how wires should get depends on application, but there are benefits to keeping just about any device other than a heater cool. The question ends up being more "How much weight and expense is it worth adding to decrease wasted power?" in the end.

I am using 10 AWG wire to power a BLDC motor, and these wires can get quite hot.

If wires are getting "quite hot" you're wasting a significant amount of electricity. Specifically, wires get hot because of \$I^2R\$ losses, which is to say the power the wire uses to produce heat, \$P\$, is equal to current squared(\$I^2\$) times resistance (\$R\$). If you can measure or calculate current, you can choose a wire size with appropriately low losses. You can also measure the existing wire and calculate its resistance if you don't have a meter. As a general rule, going down 3 AWG sizes doubles wire size and halves resistance. AWG sizes step mostly in increments of 2, but that's beside the point. The calculation of resistance based on cross sectional area measured or from a chart is easy as well. In AC systems, inductive loads, such as motors produce current out of phase with the voltage (reactive current) and in some cases (probably not yours) it is worth using power factor correction to reduce reactive current on transmission lines.

When an IC is intended for high power uses, it is common practice to put a heat sink on it, but what about for wires?

In almost every case short of superconductors it is more practical to use a larger wire and waste less energy as heat than add weight in the form of a cooling system. When space is an object, though, such as in a motor or electromagnet, microprocessor, it becomes more likely that a cooling system might be used, even if what's producing the heat are conductors.

Should I simply use a thicker gauge wire?

Yeppers. If you calculate the current, take into account if they are in free air or a closed space and look at the appropriate DC wire ampacity chart, you can round the current up to the nearest size, apply a derating factor if necessary (a multiplier to decrease the acceptable current for a wire based on factors like ambient temperature) and choose the recommended size of wire for your current. If you have a substantial length of wire you may wish to take voltage drop into account as well.

If you don't have means to measure or calculate the current, you can at the very least massively improve your setup by embiggening or doubling up the wire until it no longer becomes warm.

\$\endgroup\$
  • 1
    \$\begingroup\$ Multiple wires is also a plausible solution, is it not? This might be preferred in certain cases where space/layout is a consideration. \$\endgroup\$ – John Go-Soco Sep 6 '18 at 10:35
  • \$\begingroup\$ "embiggening" nice word. And good advice. I wish you would have say wires should not get hot, Period. \$\endgroup\$ – Misunderstood Sep 8 '18 at 17:12
  • \$\begingroup\$ @Misunderstood yeah, almost everything gets turned into a word these days =). Heating elements are wires that should get hot, as are... Hmmm is there no wire other than a heating element that should get hot come to think of it? \$\endgroup\$ – K H Sep 8 '18 at 18:03
1
\$\begingroup\$

If you are worried about the cables overheating, check the manufacturer's data sheet. PVC insulated cables are usually good for 70°C. Other plastics can handle 90°C. Specialist mineral insulated cables will go higher still.

But if you are running cables very hot, also check the temperature ratings of whatever they are terminated at (the switch, terminal block or whatever).

\$\endgroup\$
1
\$\begingroup\$

How hot should wires get?

Wires should never get hot. Not even a little warm.

No more than a few degrees over ambient.


Should I simply use a thicker gauge wire?

Thicker and or add more wires.


it is common practice to put a heat sink on it, but what about for wires?

No. Use more and or thicker wire.



Here's the thing. If a wire is getting warm during normal operation there is no safety margin for when things go wrong and more current is drawn. Be cool.

\$\endgroup\$
0
\$\begingroup\$

you need to know how much current is flowing in your system, then choose the right wire size.

There are tables like this one to pick the right AWG size. Due to the \$I^2 R\$ a thicker wire will have less power loss.

\$\endgroup\$
0
\$\begingroup\$

There are two limitations to the conductor size. One is the voltage drop, and often it is the limiting factor, but that's outside the scope of your question.

The second is the temperature rise, more specifically the temperature rise at the highest ambient temperature which translates into the maximum conductor temperature.

The temperature rise is affected by many factors besides the wire resistance (which increases at high temperatures a bit so things get slightly worse) and RMS current. The bundling with other conductors, the altitude, the air flow, even the length of the wire for short fat wires. The upper limit is set by the insulation rating or the maximum temperature the designer will permit, if less.

So in aircraft where weight is important we might use a PTFE/Kapton insulated wire that is rated for 200°C and it can actually be allowed to get that hot worst case. Cheap wire is more normally 90°C or thereabouts. So the ampacity of a given gauge of wire can vary wildly. Here is a marine chart from this web page showing allowable amperage varying over a 3:1 range (unbundled, and without correction for altitude, both which can reduce the ampacity).

enter image description here

Guidance can be found in charts like the above that assume some standard operating conditions and have some safety factor incorporated, but using nomographs and working it out completely is sometimes required.

\$\endgroup\$

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.