I would like some help to evaluate the electrical feasibility of a system to heat 35 gram of thermoplastic material (for active stiffness control) in 50 ms, by about 100 degrees Celsius.

The material will be for example a thermoplastic polymer such as poly carbonate, or a composite. The idea is to have some sort of resistive wires integrated in the material for the heating process, similar as done in [1]. But please do not worry about the material part to much.

From a rough calculation i get that the required energy is:

1200 [J/kg°C]*0.035 [kg]*100[°C]=4000[J]

to heat it in 50 ms, the power required would be approximately:

4000[J]/0.050[s]=80 000 W or 80kW.

At this point i would like to know if this is roughly feasible or not from an power and current point of view with a battery set or supper capacitors of a "typical" electric car? What type of Power Draws for short peaks are feasible with capacitors, and batteries which are typically available in electric cars? What is the approximate range? Do i need to go down a factor of 10, or can i go up a factor of 10?

[1] Tridech, Charnwit, et al. "High performance composites with active stiffness control." ACS applied materials & interfaces 5.18 (2013): 9111-9119.

  • 2
    \$\begingroup\$ That would be totally unfeasible with a non thermally conductive object. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Sep 26 '17 at 10:56
  • \$\begingroup\$ This all depends on the thermal mass involved. You use 1kJ/g here, but you may find that the actual materials involved have higher or lower thermal mass. Additionally, while the instantaneous power requirement may be high, it could very well be that a capacitor bank can achieve this. You would need to figure out the voltages and currents involved, and the method of heating. Perhaps something like pulsed-microwave of laser could be used. \$\endgroup\$ – Joren Vaes Sep 26 '17 at 11:00
  • \$\begingroup\$ I guess the "non-conductive" was to be meant "electrically non-conductive". \$\endgroup\$ – next-hack Sep 26 '17 at 11:06
  • \$\begingroup\$ Which material exactly? \$\endgroup\$ – Marko Buršič Sep 26 '17 at 11:52
  • \$\begingroup\$ Sounds like your trying to turn a combustion engine into a steam engine. But whatever, it really depends on the material too. Liquid, solid. gas? Geometry matters too. Whatever it is, I'm going to guess it will be horrendously inefficient. You could possibly use an arc furnace, though the material may be different after you heat it. \$\endgroup\$ – Trevor_G Sep 26 '17 at 12:20

3.5 kJ is something a capacitor bank can cough up in a short time. For example, that could be 48 V on about 3 F. You'd have to make sure the capacitors can handle the high current. Multiple capacitors in parallel, which is what it will take to get to 3 F anyway, helps with this.

However, there is going to be a lot of loss delivering that energy to a load, especially since it isn't "conductive". Its not clear whether you mean it's not electrically or not thermally conductive, but either way, it makes this problem difficult. If you do this electrically, you will need to store several times the energy ultimately delivered to the object.

Heating the object chemically might be easier. For example, a carefully calibrated layer of black powder should be able to deliver the energy in the short time you specify. There are probably much better chemical reactants than black powder. I'm just using that as a example.

  • \$\begingroup\$ Yes i meant electrically conductive. The general idea was to have small restive wires through the material. Chemical reactants are however also an interesting idea, that i did not thought of. \$\endgroup\$ – Hjan Sep 26 '17 at 12:46

Well, the Tesla cars claim a peak power of 581kW (!) which suggests that it might be feasible. In general, high-power batteries have a "C" rating which tells you how fast they can safely discharge. Multiply by the battery's capacity to get a current figure. So if a battery is "10C, 1000mAh" then it can discharge 10A. I would suggest looking at the "18650" round cells.

Stack enough of them together and you then have, say, a 800V 100A pack. You'll probably need a bit more to overcome losses in the cabling - which is going to be fat, Google suggests AWG 6. You will also need some means of switching this thing on and off, a means of charging it up, and a whole load of safety considerations since it's quite capable of starting a fire and killing the user, not necessarily in that order.

Rise and fall times are also a consideration - you can't just turn 80kW on and off, it has to ramp up and down.

Heat transfer considerations may be a problem - you'll need to have a lot of fine wires in your material, in fact it may be mostly wire, and you'll need to work out if they melt before transferring heat to the target.

  • \$\begingroup\$ Thanks for making clear that it is not easy. Don't worry i will not naively try this myself, at this stage (I am still in the concept stage of an idea). What are typical ramping times for this this range of power draw? is it going to affect the 50ms time interval a lot? \$\endgroup\$ – Hjan Sep 26 '17 at 13:58
  • \$\begingroup\$ Why does 80kW have to ramp up? \$\endgroup\$ – Bryan Boettcher Sep 26 '17 at 14:25
  • \$\begingroup\$ At 100A the inductance of the wiring is going to matter. As is whatever you use to switch it, which might have a debounce time or, if it's a big MOSFET, a significant gate capacitance. I don't have a good sense for whether this is miliseconds or not, but it might be a problem for microseconds. \$\endgroup\$ – pjc50 Sep 26 '17 at 15:08
  • \$\begingroup\$ Depending on the material I wonder if there's something like a carbon-loaded version that could be sandwiched between two plate electrodes and act as the resistive element. Years ago I used some conductive plastic cable, for example \$\endgroup\$ – Chris H Sep 26 '17 at 15:44
  • \$\begingroup\$ @ChrisH yes there are some papers that use carbon fibres as the resistive elements. Maybe carbon woven fabric can maybe distribute the heat more equally. \$\endgroup\$ – Hjan Sep 26 '17 at 16:05

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.