Brake cooling simulation

Vejby · February 12, 2016, 11:08

Hello all!
I will do a thesis project about improvement of brake cooling (ventilated disc).
I will use FloEFD (v14) for Catia.
I don't have any previous experience of CFD (even if I have a great interest and been reading up about it), so I reckon it will be a challenge.

My plan is to apply a heat source to the friction surfaces and measure the heat transfer out of a given control volume. And then change the ducting and run the simulation again to see if it is an improvement.

To make the computing time shorter I plan to limit the model to the wheel or maybe only the rim. Then set up a flow past the outside of the wheel and a flow in the lower part of the inside of the wheel and a pressure above that.
And then a flow in the air hose to the brake cooling duct.
This is to get an as realistic model as possible even with the simplifications. The cooling of the disc should be as even as possible, the ducting cools mainly the inboard side of the disc and through the ventilation channels, while the external airflow cools the outboard side of the disc.
So the type of simulation will be external.

The wheel/rim, brakedisc and hub will be modelled as rotating parts.
Since the flow does not enter the rotating region axisymmetrically (does that mean centered along the rotational axis?), I understand that the best way in this case is to use the "local region (sliding)" approach according to the technical documentation?
The cooling flow through the duct enters the rotating region more or less parallel with the rotation axis but at an offset.

I do not understand the following sentence in the technical reference PDF: "The solid body motion is not taken into account in heat transfer analysis"
Does it mean that friction heat due to the rotating parts is not taken into account? I mean the heat from the brake disc (that is applied as a surface source) will still be transferred to the surrounding air in the model?

How much bigger than the rotating solids should the rotating region be? Can I have stationary parts that is penetrating the rotating region (like the brake pads penetrating the rotating region of the brakedisc) ?

Does this approach seem like a good one or is it to complicated or incomplete in any way?

I am grateful for advice

Vejby · February 14, 2016, 14:10

Here is a picture of how I plan to make the rotating region, its a section view at wheel axle.
Not a very artistic picture but I hope you can understand it.

The parts of the rim that is outside the rotating region will be "rotating wall".
Is this ok? Any suggestions?

February 12, 2016, 11:08	Brake cooling simulation	#1
Vejby New Member A. Lindgren Join Date: Feb 2016 Posts: 6 Rep Power: 10	Hello all! I will do a thesis project about improvement of brake cooling (ventilated disc). I will use FloEFD (v14) for Catia. I don't have any previous experience of CFD (even if I have a great interest and been reading up about it), so I reckon it will be a challenge. My plan is to apply a heat source to the friction surfaces and measure the heat transfer out of a given control volume. And then change the ducting and run the simulation again to see if it is an improvement. To make the computing time shorter I plan to limit the model to the wheel or maybe only the rim. Then set up a flow past the outside of the wheel and a flow in the lower part of the inside of the wheel and a pressure above that. And then a flow in the air hose to the brake cooling duct. This is to get an as realistic model as possible even with the simplifications. The cooling of the disc should be as even as possible, the ducting cools mainly the inboard side of the disc and through the ventilation channels, while the external airflow cools the outboard side of the disc. So the type of simulation will be external. The wheel/rim, brakedisc and hub will be modelled as rotating parts. Since the flow does not enter the rotating region axisymmetrically (does that mean centered along the rotational axis?), I understand that the best way in this case is to use the "local region (sliding)" approach according to the technical documentation? The cooling flow through the duct enters the rotating region more or less parallel with the rotation axis but at an offset. I do not understand the following sentence in the technical reference PDF: "The solid body motion is not taken into account in heat transfer analysis" Does it mean that friction heat due to the rotating parts is not taken into account? I mean the heat from the brake disc (that is applied as a surface source) will still be transferred to the surrounding air in the model? How much bigger than the rotating solids should the rotating region be? Can I have stationary parts that is penetrating the rotating region (like the brake pads penetrating the rotating region of the brakedisc) ? Does this approach seem like a good one or is it to complicated or incomplete in any way? I am grateful for advice Last edited by Vejby; February 15, 2016 at 08:26.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Simulation of high pressure diesel injector - all phases compressible with cavitation	fivos	CFX	4	July 30, 2015 06:48
Simulation and Optimisation of centrifugal fan 3D to 2D	eRzBeNgEl	STAR-CCM+	0	January 31, 2013 13:21
Boundary Condition question for Cooling Water Pumps Bay Simulation	Jaimito	FLUENT	0	August 3, 2010 06:58
Bike brake rotor simulation	gnuboard	CFX	3	August 2, 2010 19:02
Truck brake chamber transient simulation - Moving boundary	vatsa	STAR-CCM+	0	February 11, 2010 00:14

February 14, 2016, 14:10		#2
Vejby New Member A. Lindgren Join Date: Feb 2016 Posts: 6 Rep Power: 10	Here is a picture of how I plan to make the rotating region, its a section view at wheel axle. Not a very artistic picture but I hope you can understand it. The parts of the rim that is outside the rotating region will be "rotating wall". Is this ok? Any suggestions?