TRANSIENT PIPE COOLING FEA

[ulloap]

Dear All,

I need to figure out how can I set up this FEA, I wanna simulate a plastic pipe cooling process in a extrusion line, so basically, I have a water cooling nozzles over the surface of the pipe, they're fixed in a frame, and the pipe goes in a constant velocity passing for all the water nozzles to get cooler, simulating the conveyor moving the pipe through it, or have the pipe fixed and move the nozzles, I have been played with DFLUX subroutine and SFILM with no success.

I need to find the exactly time to get cool down.

Can somebody guide me please

Regards

 

[corus]

The heat transfer coefficient from nozzles is particularly difficult and a more precise analysis would be to model the pipe in 3D rather than an axisymmetric analysis, though you'd have symmetry at +/- 180 degrees from the vertical. The pipe will no doubt be supported on rollers too, I'd imagine, so you'd also need to include the heat transfer to those too. You'll need a couple temperature displacement analysis so you can move the pipe. To define the heat transfer to the nozzles and roller supports you could explore using fixed parts and gap conductance to those parts as the pipe moves through them and 'fudge' the conductance so that it's zero at a certain distance away and approaches the true value of heat transfer coefficient as it moves closer.

 

[ulloap]

Corus,

I really appreciate your help, I'm agree with your FEA statement.

But @ this time I wanna to simplify the model as far as I can, so that's my idea to keep steady the pipe and axisymetric, and try to move the jet cool nozzle (s) in the axial direction, to cool down pipe, what is not clear yet, is how can I move my SFILM condition when I move my jet nozzle over the surface of the pipe.

Regards

 

[corus]

If I recall, there is a quasi steady state method within Abaqus by which the material can be moved at a constant speed and boundary conditions can then be applied to surfaces in the usual fixed manner. Basically the transient term in the heat diffusion equation is replaced by a term that involves the velocity of the material. I'm not sure if this applies to axisymmetric elements though. You'd have to check in the manuals which I don't have right now. Otherwise you could still try using gap conductance and move the material with a coupled temperature displacement model I described earlier.

 

[IceBreakerSours]

That's "steady state transport".

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083