Fluid Dynamics, Metalurgy, Math??

[todayican]

Im wondering if there is someone out there who can help me do better then just "eyeball" the cooling vent requirements on my homebuilt trike (the Belaero) Here is what Ill need to cool:

People
Ev controller
Electric motor
diesel motor.

Ill be using Naca inlets with varying "pipe sizes" the size of the tube that the inlet builds to just under the skin of the trike.

Heres some things I know to help size these things:

Average speed of airflow across all missions: about 42 mph

Diesel 10hp, most of these are used in stationary applications as generators, water pumps, or dc welders. its 415cc and will be running at 80% of its max load.

E motor rated at 30kw for 3 min 10kw continuous. it will see about 10kw worst case for up to 15 - 20 min at a time (climbing hills) (motor claims 93% efficiency so figure 90 real world.

Controller Sevcon Gen 4 claims 95% efficiency. it will be used in both drive and constant generation mode in conjunction with the diesel, so it will see a lot of action. I will have a heat-sink running the length and width of the bottom of it with 1.5" fins.

I'm thinking 2" toward the diesels cylinder head, 2" towards the electric motor, and 6"x1.5" rectangle aiming down the "fin channel" on the controllers heat sink.

Any input on how these calculations work?
Thanks!

[Zerohour]

There's alot of variables there. Are you going strictly heat sink? It sounds to be the case. At that point we would need pictures, diagrams, some initial data of the devices and information on the enclosure. Also needs to consider loads both given by driver and geography of the area(and all potential traffic patterns), and as well as the climate where the vehicle will be driven.

[RobertSmalls]

If you're already good with math and need some help with a mathematical model of a heat exchanger, check out a copy of this book:

Amazon.com: Introduction To Heat Transfer (9780471386490): Frank P. Incropera, David P. DeWitt: Books

I keep one on my bookshelf.

For loads that will last 15-20 minutes, don't bother trying to store heat. You need a heat exchanger that will get rid of it as fast as you produce it. There are enough numbers in the above post (except diesel engine efficiency - conservatively suppose 50% of the energy going in needs to leave through the radiator, and the rest through the output shaft and tailpipe) to calculate the total amount of heat you need to dissipate.

Given initial temperature and flow rate for air and water (or oil), you should have enough information to guesstimate heat flux per square meter of heat exchanger. I say guesstimate, because fins complicate things a great deal.

Maybe once you compile the above information, you could e-mail it to a company that makes custom radiators and ask them to size you one. I bet they have a spreadsheet for their radiators that's more accurate than anything I could spit out by guessing at the fin effectiveness.

[todayican]

For the record, all the parts are air cooled.
I dont know If I mentioned before, but I will have a 1" gap on either side of the rear drive wheel (wheel is 25" high) so with a 6" ground clearance there will be 38 sq inches of "exhaust" heading out of the engine compartment (which is otherwise enclosed)

[RobertSmalls]

Air cooling, eh? It's not always the most aerodynamic option, but it has its place.

The above text has plenty of formulae for the simulation of heatsinks, but given the complexities of their geometry and the geometry of your air flow, I wouldn't even bother doing anything other than an instrumented build 'n' test.

[todayican]

Yeah, thats what I was thinking. everything in the equasion has a temp probe, so it will be a close watch on the temp readings for a while, perhaps some variable openings based on temp.

Id like to watercool the controller and e motor, but its not that kind of design.

One thing I was thinking was to design the outer skin of the car so the fins of the sink were actually outside the skin of the car and in the airflow.

[aerohead]

Originally Posted by todayican

Yeah, thats what I was thinking. everything in the equasion has a temp probe, so it will be a close watch on the temp readings for a while, perhaps some variable openings based on temp.

Id like to watercool the controller and e motor, but its not that kind of design.

One thing I was thinking was to design the outer skin of the car so the fins of the sink were actually outside the skin of the car and in the airflow.

Michael Hackelman ( sp? ) did this with his HONDA Civic EV conversion.The heat sink was atop the hood below the line of sight,ahead of the cowl where the airstream was highly energetic.
Michael has written a very good book on EVs and it might have some good empirical stuff to contemplate.
Also,in the Sticky on mod-data,you should find a reference to Dr. Walter Korff's low-drag cooling system design and inlet architecture.
If you're going to operate the diesel at 80 %,you may just want to use the original radiator or one of equal heat rejection capability,and then play with the ducting,both sides if you can.
PS, the human body develops about 300-Btu/hr at rest.If you can cool the head and neck it will help temper the whole bloodstream for comfort.Successful Solar race cars might give clues as to NACA submerged inlet minimums for driver survival.

[todayican]

Thanks for the pointers! for the record, the Diesel is air cooled as well

[aerohead]

Originally Posted by todayican

Thanks for the pointers! for the record, the Diesel is air cooled as well

If the engine has a crankshaft-mounted fan which induces air into a shroud,then forcing over cooling fins on cylinder and head,the 'inlet' diameter and its free area on the shroud will be your guide for ducting area.
I don't trust my memory and neither should you,but,say,Korff said a 'radiator' inlet,with full width of the radiator,but only 1/6th the height,at a distance ahead of the radiator equal to the radiator height,then that might give a clue as to the minimum.And this would be for nice,smooth-walled duct,air-tight,so as not to lose any precious ram pressure from the inlet.

[todayican]

Excellent! Thanks