Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Cooling down a barrel of wort

Please login with a confirmed email address before reporting spam

Dear everyone,

For a school project, I have to model a barrel of hot wort cooled down by cold water flowing trough a spiral. Since the barrel is pretty decent in size and there will be some convection in the wort, calculation times are pretty high. Since I have to build the model to later design ways to shorten the cooling time, this long calculation time is not what you want.

Anyone that has an idea of how to reduce this time as much as possible by using (not so common) functions in Comsol?

By myself I already came up with the idea of just taking one quarter of the full barrel and then coupling all pieces of the spiral. But by doing this, there will have to be some kind of rule applied to the coupling. Something I have no idea about at the moment.

I hope some of you can help me further.

Sincerely,

Guus



3 Replies Last Post Nov 14, 2019, 8:42 a.m. EST
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Nov 12, 2019, 5:05 p.m. EST
Updated: 5 years ago Nov 12, 2019, 12:05 p.m. EST

Hello Guus,

One approach that could potentially save you quite a bit of computational resources is presented in this model of cooling in an injection molding operation. The idea is to replace the cooling flow in the spiral by a single line. It's a good approximation if the temperature distribution is fairly uniform in any cross-section of your spiral, which I would guess to be the case here. Note that this methodology will require the Pipe Flow Module. Another example along the same lines is this tutorial.

Best of luck for your project. It's a fun one!

Best regards,

Jeff

-------------------
Jeff Hiller
Hello Guus, One approach that could potentially save you quite a bit of computational resources is presented in [this model](https://www.comsol.com/model/cooling-of-an-injection-mold-12371) of cooling in an injection molding operation. The idea is to replace the cooling flow in the spiral by a single line. It's a good approximation if the temperature distribution is fairly uniform in any cross-section of your spiral, which I would guess to be the case here. Note that this methodology will require the Pipe Flow Module. Another example along the same lines is [this tutorial](https://www.comsol.com/model/ground-heat-recovery-for-radiant-floor-heating-18529). Best of luck for your project. It's a fun one! Best regards, Jeff

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Nov 12, 2019, 6:06 p.m. EST

Dear Jeff,

Thanks for the fast reply, I indeed already checked out this method but here the convection currents will not be generated in the same way as with the fully modelled pipes since the model will see it as a line and not as an 3d object. Therefore the results will differ from the tested measurement.

Dear Jeff, Thanks for the fast reply, I indeed already checked out this method but here the convection currents will not be generated in the same way as with the fully modelled pipes since the model will see it as a line and not as an 3d object. Therefore the results will differ from the tested measurement.

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 5 years ago Nov 14, 2019, 8:42 a.m. EST
Updated: 5 years ago Nov 14, 2019, 3:47 a.m. EST

Hi Guus,

You indicate that the convection currents will not be generated in the same way if you replaced the 3D spiral with a line spiral. Is it because the lines that would replace the spiral would take less space than the 3D spiral would, and therefore the wort domain would be "too big" between the coils?

If that's the concern, you could try using a hybrid approach where you simultaneously draw a 3D spiral, thereby accounting for its real cross-section, and draw a line representation of the spiral at the center of the 3D spiral. You would use the Pipe Flow Module on the line spiral and a component coupling to pass the averaged temperature from each 3D spiral cross section perimeter to the corresponding point on the line spiral.

I haven't tried it, but conceptually it should work. If the temperature in the wort does not have a strong gradient along the axis of the spiral, so that the averaging effect is small, it could be an elegant solution.

Best regards,

Jeff

-------------------
Jeff Hiller
Hi Guus, You indicate that the convection currents will not be generated in the same way if you replaced the 3D spiral with a line spiral. Is it because the lines that would replace the spiral would take less space than the 3D spiral would, and therefore the wort domain would be "too big" between the coils? If that's the concern, you could try using a hybrid approach where you simultaneously draw a 3D spiral, thereby accounting for its real cross-section, _and_ draw a line representation of the spiral at the center of the 3D spiral. You would use the Pipe Flow Module on the line spiral and a component coupling to pass the averaged temperature from each 3D spiral cross section perimeter to the corresponding point on the line spiral. I haven't tried it, but conceptually it should work. If the temperature in the wort does not have a strong gradient along the axis of the spiral, so that the averaging effect is small, it could be an elegant solution. Best regards, Jeff

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.