Radiator Validation

[VictorV8]

Hello everybody!!

I just want to know from you guys, the ways that I can validate a radiator, I know that I can do it by putting it in the engine and run it to see the results, but I want something safer, because it's my first radiator project and I don't want to burn the engine if anything goes wrong.
Any idea would be nice!!

I am sorry for any mistake, English is not my native language!!

[onemaniac]

Hello everybody!!

I just want to know from you guys, the ways that I can validate a radiator, I know that I can do it by putting it in the engine and run it to see the results, but I want something safer, because it's my first radiator project and I don't want to burn the engine if anything goes wrong.
Any idea would be nice!!

I am sorry for any mistake, English is not my native language!!

http://en.wikipedia.org/wiki/Laws_of_thermodynamics

[stever95]

Just monitor coolant temp...

[Dunk Mckay]

get a double pass radiator that is 100mm deep, 600mm tall and 800mm wide, attach ducting to a 500mm fan to the back of it and mount it perpendicular to the direction of travel of the car. That's big enough that if your engine overheats it wont be your fault...

..on the other hand if the cooling system ends up weighing 50kg that will be you fault.

At the end of the day you want it as small as possible without risking overheating. The short-cut is to go from experience and observation, what have you used successfully before, what have others used successfully. The more engineer-y approach is to actually figure out how much heat your engine is putting out through thorough thermodynamic calculation. But that is no guarantee as not only the engine that you use but also they way you map it, the way you box it into the chassis, the position, size and length of your exhaust runners will all have an effect, as will the temperature on the day. So now you need to go and do a full thermofluid CFD analysis of your entire car, and simulate a full endurance run, stopping for driver change and all. Of course then some of the hydrodynamic bearings might start to fade in your engine, not causing it fail, but far more heat than normal to be produced.

At the end of the day there even if you calculate what will work in all normal circumstances the fact of the matter is that you will end up with non normal circumstances that you could not have foreseen, some that you can fix/change, others that you can't. So calculate what you'll need, then slap a safety factor onto it for when those unforeseeables go wrong. Want more reliability? Go bigger. Want better acceleration? Go smaller. Want to validate a radiator? Put it on the car and run many many hours of testing. If it works, keep going smaller until it doesn't and go back one. If it doesn't keep going bigger until it does. Can't afford multiple radiators for testing, play it safe and get a massive one.

[Kanishka Panda]

Hello and Namaste!

Even if your car isn't run-ready for cooling system testing and you want to test your cooling system design before finalizing the design to be put in your car, you may carry out the following experiment-
Maintain the radiator inlet temperature at a constant (according to your needs- this can be calculated using a simple code using the power dissipation rate data of your engine/tractive system) using a thermostat
Pump coolant through the radiator at the same flow rate as to be used in your car (this flow rate can be maintained with the help of a flow control valve); flow rate sensors are available to monitor this flow rate
The velocity of the air flowing into the radiator (in case no fans are used) can be maintained using a fan in the setup and the flow velocity of air can be monitored with a pitot tube; if you are using a fan in your car as well, then needless to say, use the fan at the same rpm (as decided from your calcs) for your experiment
The radiator outlet temperature can be measured with a temperature sensor

You can now observe the cooling effect produced and also calculate the actual heat dissipated through the radiator.
This way, you can recreate the actual conditions (when car runs) in an experiment and test your calculations and modelling used to design the radiator.

Regards,

Kanishka Panda
Ex-Cooling System Design Engineer
IIT Bombay Racing
India