Thermal Analysis

[ogg418]

Hi, im new to the forums.
I have a quick question. I have some parts made on solidworks and want to carry out some thermal analysis on these parts. this part is a disc and i want to carry out the analysis when a pad comes into contact with the disc over a certain period. how do i go about carrying out this analysis?, do i need to animate the assembly and then apply the thermal analysis? any help would be much appreciated. cheers

 

[gfbotha]

Not sure about what software modules & version you have, what inputs are available and what exactly you want to achieve.

If you want to model the frictional heating of a brake disc as the pad(s) rub against it, you might need an average heat flux as input, ignore rotation and model the disc only for a 1st order approach. My guess about your problem details may not be quite right, but you might also consider modelling the 2 parts in contact all the time but then play around using time curves for properties to mimic break of contact. E.g., at some stage one could make K very large for the pad.

What ever you do, I think you will have to make a few assumptions in order to simplify the problem.

I'm under the impression surface type contact resistance can only be specified as a constant at this stage...?

 

[ogg418]

its all on the subject of brake fade basically. I have a few different designs on disc and want to carry on analysis so i can see what effect different designs have, ie a disc with holes compared to those with no holes, and then number of holes, thickness etc etc.

would i need to mate the brake pad to the disc and then carry out analysis?, i would guess so. but this is what i need help with. how do i go about doing this, becuase i can do the same thing for all the variations in design i have.

also, what do you mean by surface type contact resisitance being constant? if this means the amount of pressure applied when in contact, i can probably simplify this to say its either in contact or not at all? hope this helps.

thanks for the reply.

 

[gfbotha]

Ogg, what you want to do can typically become quite involved. It depends on the variation of possible braking cycles, times in between, speeds, and the latter will even influence your convection coefficient. As said, you will have to simplify the problem quite a lot.

To compare the designs I think you might consider simulating a typical heating cycle followed by a cooling cycle. You would like to note the max temp reached and then how much it cools afterwards. You will have to include simulating the cooling contribution of the holes (radiation & convection heat loss). I tend to think you should start simple - only with the disc; not modelling rotation, apply a spread heat flux over the full 360° for a specified time, then simulate cooling. You could still model only a sector (axisymmetrical). If you do not know typical braking heat flux values, you could try and calculate or estimate it from braking power at a certain rotational speed.

I am referring to the surface type contact one can specify when modelling an assembly; in the version of Cosmos DesignStar I am using, one can currently only specify a constant resistance (not a resistance that say varies via time curve). DesignStar is related to CosmosWorks.

 

[cdennyb]

gfbotha is very accurate. In addition to his procedure you can simulate the wind speed of cooling by positioning a "fan" on one end of your model and then provide the proper data to simulate the wind cooling in addition to convection cooling which woud occur at a full stop.
In order to do the full analysis you already probably realize you'd have to heat soak the disk with the pad to simulate a stopped condition with the pad in contact with the disk, trapping the heat.
this is very complex and assumptions need to be made in order to compare.
you don't need to have ALL the details of the heating and cooling, just the one's that matter between your designs.
like shrouds or not, cold ambient air temp or hot desert ambient.
Sounds like a realy computer cruncher to me.
good luck.

db