Cooling a resistor powered at 6 to 10kW

[stube40]

I have some resistors to cool that I presume are going to get rather hot!!

I am using 6x wirewound resistors in parallel to create a 3.75 Ohm load for a 150V source, hence generating 40A and 6kW. Later I may up the voltage to 200 so I need to spec the cooling for a max of 10kW. Each wirewound resistor is 28mm dia and 350mm long.

I was thinking of a open perspex box filled with some oil (maybe olive oil). Then creating a custom lid for the box out of a thermally conductive material that is an electrical insulator. Then mounting the resistors on to the bottom side of the lid on long legs so that they are completely submerged. On the top side of the lid I would put a large heatsink and fan.

The maximum length of time the power will flow through the load is 1 minute, so this will help reduce the overall specs.

If anyone likes this idea then my next challenge is to spec the size of the box, amount of oil and the size of the heatsink and fan. This is the tricky bit obviously and I'm not really sure where to start.

 

[Resqueline]

Specific heat capacities for different oils varies, some numbers are 1670-2100 J/(kg*K).
Enclosing it in perspex limits the upper temperature to 80°C. Starting at 30°C that's a 50°C (50K) temperature rise.
6kW * 60 sec's = 360kJ. Divide this by 50 and by 1670 and you get a need for 4.3kg of oil. Using 5 liters puts you on the safe side.
I'd rather use a tin can and mount the resistors and any cooling devices suspended from a (perspex) lid. How about using copper tubes and water cooling, or computer heat pipes?
How about using a paint can and it's original lid, maybe reinforcing it, for leak-proof containment. Insulated feed-throughs are easy to make or get and are good enough for the job. I't stack the resistor vertically in a circle with a central post for support & connection to their lower part.
Radio amateurs have been making artificial antenna load resistors like this for ages. You'll surely find a design recipe in the ARRL handbook.

 

[stube40]

Thanks Resqueline.

Over the weekend I spent some time on this myself and came up with the following maths for using 20 litres of Olive Oil (as getting tranformer oil is proving tricky). I think we're both in the same ball park:

1 litre of olive oil weighs between 0.8kg and 0.92kg. Assume mid-way == 0.86kg. 20 litres of olive oil thus weighs 17.2kg

Olive Oil has a specific heat capacity of 1.97 kJ/kg °C. 1.97J of energy raises 1g of Olive Oil by 1 degC.

Thus, 1.97W of power in 1s raises 1g of Olive Oil by 1 degC
Thus, 33.884kW of power in 1s raises 17.2kg of Olive Oil by 1degC
Thus, 10kW of power in 1s raises 5.076kg of Olive Oil by 1degC
Thus, 10kW of power in 1s raises 17.2kg of Olive Oil by 0.29degC

So, at 10kW max power the oil bath will rise in temperature by approx 0.3degC per second. At 6kW nominal power, the oil bath will rise in temperature by approx 0.18degC per second. Thus, running the system for one minute at 150V / 40A will result in a 10degC overall rise in temperate in the oil bath. However, this is an average temperature and there may be local hot-spots hence plenty of headroom is the key.

 

[neon]

Are you trying to biuld a heater or what. oil will retain the heat for a long time now a fan will disperse the heat to whatever you point at and away from the source, juswt think about that for a while

 

[Mitchekj]

May need a rather large fan to cool 6-10kW of dissipation. Maybe a souped up wind tunnel?