Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Temperature decline in autoclave model while rise expected

Please login with a confirmed email address before reporting spam

Hi there,

I've made a model for an autoclave with 2 blocks in it that should heat up during time. They however first cool down to like 200K and after they start to heat up again. That's not what i suspect and what is realistic. Since the model with results is to big i will upload only the model. It isnt that big so should be computable in short time;). (comsol 4.2a)

See the attachted image.

hope all is clear.


Thanks in advance!






8 Replies Last Post 22.05.2012, 11:22 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 20.05.2012, 10:57 GMT-4
Hi

frst of all you have the Conugated HT that mixes NS and HT in a pre-cooked way.

Then for your model, you need at least to define a "heat flux outlet, or open boundary" at the air outflow boundary as quite some heat is escaping with your air, but not now the way you have set up your model

--
Good luck
Ivar
Hi frst of all you have the Conugated HT that mixes NS and HT in a pre-cooked way. Then for your model, you need at least to define a "heat flux outlet, or open boundary" at the air outflow boundary as quite some heat is escaping with your air, but not now the way you have set up your model -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21.05.2012, 04:13 GMT-4
Thanks a lot.

I completely overlocked the outflow of the HT. Model works a lot better now.

I will also try to take the HT and TF togetter.


Thanks again.
Thanks a lot. I completely overlocked the outflow of the HT. Model works a lot better now. I will also try to take the HT and TF togetter. Thanks again.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21.05.2012, 05:44 GMT-4
Dear Ivar,


I've added the outflow condition but is still gives me the unexpected decrease in temperature.. The previous results were deleted. I also Included a results graph and empty model.

I also tried to combine the HT an TF but gave me some errors, Will try that later on once again.

What is going wrong?

Thanks
Dear Ivar, I've added the outflow condition but is still gives me the unexpected decrease in temperature.. The previous results were deleted. I also Included a results graph and empty model. I also tried to combine the HT an TF but gave me some errors, Will try that later on once again. What is going wrong? Thanks


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 21.05.2012, 07:41 GMT-4
Hi

what about the NITF Conjugated Heat Transfer, that is the "one" from my understanding that fits the best your case

And for your temperature ramp, why not use a Temp = rm1(t[1/min]) with rm1() defined as a Definition Functions Ramp at location 0, slope DT, cutoff T1, and smooth at start and stop of 1


--
Good luck
Ivar
Hi what about the NITF Conjugated Heat Transfer, that is the "one" from my understanding that fits the best your case And for your temperature ramp, why not use a Temp = rm1(t[1/min]) with rm1() defined as a Definition Functions Ramp at location 0, slope DT, cutoff T1, and smooth at start and stop of 1 -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22.05.2012, 09:39 GMT-4
Dear Ivan,

I tried both but i now gives me some strange errors.
What is going wrong, i somehow cannot find the error

thanks
Dear Ivan, I tried both but i now gives me some strange errors. What is going wrong, i somehow cannot find the error thanks


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22.05.2012, 11:10 GMT-4
Hi

you need to do some reading and trainig I believe ;)
1) in NITF ther are the solid and the fluid driving approach, you should use the FLUID driving method
2) start simple, use first a laminar flow with no inertialterms
3) use a finer mesh, as you have sharp corners you will get vertex effects which takes longer to solve or use a tube for the airinflow
4) define correctly your functions, often its worth to express the functions fully as analytical ones, then to plot hem and then only to copy the analytical function call into the BC,s
5) you have no V0 for the inflow
6) for diffusion problems, in time series I mostly use power of 2 (or 10) stepping, it follows nicely the assymptotical settling
7) starting a flow with some small velocity and pressure drop often helps the convergence
8) when you include the inertial terms, its important to ramp slowly not to make shock waves, otherwise the solver gets occupied to resolve these and never comes to your flow buildup
9) jump into turbulence only when you mater therest (I havent even tried in thsi case ;)

I have made a certain numebr of changes in the model hereby, I had to reset it and clear the solver settigns, could be an 110-166 patch issue, I can only suggest that you get latest patch up, as there are a few fixes

There is something funny in your file, its stuck in turbulent, something is not correctly reset, I would suggest to restart from scratch

--
Good luck
Ivar
Hi you need to do some reading and trainig I believe ;) 1) in NITF ther are the solid and the fluid driving approach, you should use the FLUID driving method 2) start simple, use first a laminar flow with no inertialterms 3) use a finer mesh, as you have sharp corners you will get vertex effects which takes longer to solve or use a tube for the airinflow 4) define correctly your functions, often its worth to express the functions fully as analytical ones, then to plot hem and then only to copy the analytical function call into the BC,s 5) you have no V0 for the inflow 6) for diffusion problems, in time series I mostly use power of 2 (or 10) stepping, it follows nicely the assymptotical settling 7) starting a flow with some small velocity and pressure drop often helps the convergence 8) when you include the inertial terms, its important to ramp slowly not to make shock waves, otherwise the solver gets occupied to resolve these and never comes to your flow buildup 9) jump into turbulence only when you mater therest (I havent even tried in thsi case ;) I have made a certain numebr of changes in the model hereby, I had to reset it and clear the solver settigns, could be an 110-166 patch issue, I can only suggest that you get latest patch up, as there are a few fixes There is something funny in your file, its stuck in turbulent, something is not correctly reset, I would suggest to restart from scratch -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22.05.2012, 11:13 GMT-4
Thanks again for the quick response.

I am quite new to comsol so there is indeed a lot left to learn!

:)
Thanks again for the quick response. I am quite new to comsol so there is indeed a lot left to learn! :)

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22.05.2012, 11:22 GMT-4
Hi

Well keep up going, do not despair ;)
But start simple you will learn quicker and be less frustrated, build up the complexity phyiscs by physics. and restartng a model is quickly done and you learn a lot in the beginning that way

--
Good luck
Ivar
Hi Well keep up going, do not despair ;) But start simple you will learn quicker and be less frustrated, build up the complexity phyiscs by physics. and restartng a model is quickly done and you learn a lot in the beginning that way -- Good luck Ivar

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.