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 The information provided may be out of date.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.
heat sink on top of thermally resistive layer
Posted Oct 11, 2012, 9:18 a.m. EDT Heat Transfer & Phase Change Version 4.2a 6 Replies
Please login with a confirmed email address before reporting spam
does anybody now if it is possible to make a system like in the attached file, i.e. some structure with a heat source and a boundary with a heat sink, which is supposed to be a resistive layer at the same time.
I'm interested in building this thing without workarounds, meaning only layers and don't include any meshes.
a section through the system could look like:
physical -> comsol:
I) region with source -> reconstructed in detail
II) rather thin resistive layer -> using the feature of the 'thin resistive layer'
III) heat sink -> a boundary that is set to some temperature
if I do this in a naive way it's not allowed to combine II) and III)
any hint is appreciated, thanks in advance.
flo
I'm interested in building this thing without workarounds, meaning only layers and don't include any meshes.
a section through the system could look like:
physical -> comsol:
I) region with source -> reconstructed in detail
II) rather thin resistive layer -> using the feature of the 'thin resistive layer'
III) heat sink -> a boundary that is set to some temperature
if I do this in a naive way it's not allowed to combine II) and III)
any hint is appreciated, thanks in advance.
flo
Attachments:
6 Replies Last Post Oct 12, 2012, 4:02 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
What do you want to get from the model? Which properties are you interested in? If it is just the temperature of the gold wire at a given power level, with all the simplifications you make, this can be done with a hand calculation, no need for COMSOL.
Cheers
Edgar
Cheers
Edgar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I'm interested in the temporal behavior, which is, at least I think, not that easy to determine even though the system is simple indeed.
But please don't hesitate, enlighten me and share your knowledge!
Let's say I'm interested in a stationary value for the wire-temperature for a given power. what would you suggest?
But please don't hesitate, enlighten me and share your knowledge!
Let's say I'm interested in a stationary value for the wire-temperature for a given power. what would you suggest?
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
for me it looks easier to model as a small 2D case, as sketched, then to derived the main parameters and find a simple equation to reproduce these values, and plug that into your larger model. Or even easier, have 2 coupled models, on small 2D as sketched, and a bigger one for the surrounding, and you link in the relevant items.
What about considering some radiation loss on top and perhaps convection cooling too ? (all depends on the final temperature.
Notice "T" is NOT conservative, energy fluxes yes, when expressed corectly
--
Good luck
Ivar
for me it looks easier to model as a small 2D case, as sketched, then to derived the main parameters and find a simple equation to reproduce these values, and plug that into your larger model. Or even easier, have 2 coupled models, on small 2D as sketched, and a bigger one for the surrounding, and you link in the relevant items.
What about considering some radiation loss on top and perhaps convection cooling too ? (all depends on the final temperature.
Notice "T" is NOT conservative, energy fluxes yes, when expressed corectly
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
---- for me it looks easier to model as a small 2D case, as sketched,
ok, so far
---- then to derived the main parameters and find a simple equation to reproduce these values,
I'm not really sure what you mean, sorry.
what analytically is going on in the system is smth I already know about.
1d vertical diffusion
2d half-spaced spread
1d horizontal
all with different timescales.
but I would have no idea what i could use this information for in comsol?!
---- and plug that into your larger model
I guess you're referring to the surrounding. but the surrounding is not really apparent, I implement it with really simple boundary conditions:
isolated at the hole top and the sides and a sink at the bottom.
I already know that radiation plays a tiny-minor role for the powers of interest and the hole thing is in vacuum (UHV), so no need for conductive effects either.
The only thing I would like to do is to replace the copper-block in the sketch by a heat-sink and at the same time add a resistive layer at the bottom, sort of; before the heat gets to the sink it has to diffuse through the layer.
ok, so far
---- then to derived the main parameters and find a simple equation to reproduce these values,
I'm not really sure what you mean, sorry.
what analytically is going on in the system is smth I already know about.
1d vertical diffusion
2d half-spaced spread
1d horizontal
all with different timescales.
but I would have no idea what i could use this information for in comsol?!
---- and plug that into your larger model
I guess you're referring to the surrounding. but the surrounding is not really apparent, I implement it with really simple boundary conditions:
isolated at the hole top and the sides and a sink at the bottom.
I already know that radiation plays a tiny-minor role for the powers of interest and the hole thing is in vacuum (UHV), so no need for conductive effects either.
The only thing I would like to do is to replace the copper-block in the sketch by a heat-sink and at the same time add a resistive layer at the bottom, sort of; before the heat gets to the sink it has to diffuse through the layer.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
Well Edgar is right asking what you are afte? as I misunderstood a few items, if you are not interested in a larger model (as I assumed you wanted to simplify your model to plug it into something bigger).
Well where do you want to know the temperatures ? on which domains ?
If you are not interested in the details of the gold wire, use a perfect conductor layer, the sink ideally the same, but you must ensure your simplification remain valid. As your model is not really complex you could model both and compare the two.
Take a particlular close look to the differences of Tx Ty and not only T (asuming x-y the in plane vector ir is it x-z for you ?) and I asume the figure is staing your model is 2D and infinite in Y (and not x)
--
Good luck
Ivar
Well Edgar is right asking what you are afte? as I misunderstood a few items, if you are not interested in a larger model (as I assumed you wanted to simplify your model to plug it into something bigger).
Well where do you want to know the temperatures ? on which domains ?
If you are not interested in the details of the gold wire, use a perfect conductor layer, the sink ideally the same, but you must ensure your simplification remain valid. As your model is not really complex you could model both and compare the two.
Take a particlular close look to the differences of Tx Ty and not only T (asuming x-y the in plane vector ir is it x-z for you ?) and I asume the figure is staing your model is 2D and infinite in Y (and not x)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
If you are just interested in a stationary situation and do not want to consider losses or geometrical distributions you can do it analytically. Basically it is:
Pin = Pout
Let Pin be the power in your gold wire, Tgold and Tcopper the temperatures, Ct the thermal conductivity of the resistive layer.
Pin = (Tgold - Tcopper)*Ct
Now solve for Tgold and you are done.
Cheers
Edgar