Cooling a wire to increase its current rating and lifespan

Question

Say I have an 18 AWG wire that can stand 2.7 A. I need to use it for a much higher current than its rating, about 80 A (pulsed current for a few milliseconds).

The wire will get hot more and more with each pulse over time.

So if I made a cooling system for it with a Peltier diode, would this increase its life span and prevent heating?

Can it hold out above its limit with this pulsed current?

Here is my circuit:

This is the coil:

V1 is a step up booster up to 390 V.

Wires used here are 10 AWG except

• wire of thyristor to B2 is 20 AWG
• coil wire is 18 AWG- 400 gram (not sure about length)

The coil wire resistance is 3.5 ohm. The capacitor will send about 80 A to the coil.

Can the coil wire withstand this current with a cooling system to cool the coil?

Answer 1

Answering your original question, a typical way of cooling coils is using mineral oil.

It can not only refrigerate the bobbins but also eliminate corona discharges and act as a general insulator.

An example of a commercial brand for transformer oil is this one.

One key parameter is to keep the oil cool by circulating it in a heat exchanger or similar. Additionally, the oil degrades quickly in the presence of oxygen and thus it must be kept in a sealed container. You might do some tests on your coil by making an enclosure out of epoxy resin, pouring the oil and sealing it properly. A vacuum pump is mandatory to remove all the air in the oil.

Having said all of that, two final ideas:

1. Forget about Peltier cells. They won't make a difference as they aren't cost & efficiency worth it for your application.
2. If your coil will get a high current for some milliseconds only, is overheating actually concerning? Have you measured the temperature?

Additional ideas:

1. If you are ABSOLUTELY sure that you will pass up to 80A through that coil, do NOT use mechanical switches to do that. You will arc and damage the contacts sooner than later. Consider using either a high current MOSFET or an IGBT and send small digital pulses to the electronic switches in saturation to feed the current you need.

2. Wear PPE at all times when running your tests! You don't want a nail or metallic piece flying into your eye.

3. I wouldn't personally use relays, as shown in the linked video (unless they will be actuated a few times only).

4. Make sure that your flyback diode is properly rated at at least 20% above the max current you expect to pass through the coil.

Answer 2

An interesting solution would be if it is not continuous you can use paraffin wax, during phase change it has a the ability to absorb a lot of heat. The heat of fusion of paraffin wax is about 90 - 95 Btu / lb and the heat capacity is about 0.64 Btu / lb / *F. Just for a use, the Lunar Rovers batteries needed a cooling system, and they couldn't add any substantial weight to the design. Their solution was to use paraffin wax as a heat sink. As we now know it worked. With the wax you can melt it and pour it in the coil housing. You will need to allow a small space for expansion and contraction.

There would be no need to change the wax, it should last as long as the product then you could make candles out of it. As it absorbs heat it gets warmer and eventually melts. In the molten state is is a better conductor then when solid. It will also get in into intimate contact with the windings to absorb heat. When It will dissipate its heat through the case even while adsorbing from the coil and eventually the wax go back to a solid as it gives up the heat it adsorbed. It also conducts heat to the case and I have no idea how much that could dissipate. We experimented with this many years ago and used Saber and Ice-pak (thermal cad) to validate our findings, it was close.

Answer 3

In order to ensure that the heat generated by the wire does not exceed the heat generated from nominal operating current, use this formula:

\$A \sqrt{D} \lt A_0\$

A ≡ pulse current (80A)

D ≡ duty cycle (pulse duration divided by total cycle duration)

A₀ ≡ nominal current (2.7A)

If your pulse current multiplied by the square root of the duty cycle is less than the nominal rated current than your wire can handle the pulse current.

Answer 4

Yes, since you have a temperature problem, increasing cooling can definitely solve the problem.

A Peltier module won't help you, since you don't need a temperature lower than ambient, nor do you have a single solid place that you need to suck heat out of.

Consider the simple solution first (KISS) - use a fan to blow air at the coil. If that's not enough, try cutting holes in the plastic ends, and blowing air through the coil. If that's not enough, try a more powerful fan. And if a reasonably powerful fan can't provide enough cooling, then you can try mineral oil or paraffin wax or other ideas.