Radiosity Solver and RSURF 1

[HVSmith]

Hi - I'm having problems using the radiosity solver.

I am trying to model a thin plate. Heat is generated on the plate and then there is radiative cooling. For a single plate I think I have the model working correctly.

However, I also require to simulate a double plate scenario, where the plates radiate between themselves, as well as the surroundings.

flash3780 kindly sent this useful link:

http://www.ansys.com/staticassets/A...article/AA-V1-I1-View-Factoring-Radiation.pdf

But I am still having trouble in getting radiation between the plates. Presently I am only heating one plate, to see if I can see any heat transfer to the other - but I see nothing

I have attached my macro file in case anyone can see anything that I'm doing obviously wrong? Or can give me any tips for succeeding?

http://files.engineering.com/getfil...7-98c8-4280-9bd0-b9f4994997b8&file=KEKRAD.mac

Thanks for any help you are able to offer, all very much appreciated.

HVSmith

 

[flash3780]

Do you mean that you want to capture the effects of through-thickness temperature variation by tying two sets of SURF152 elements to TTOP and TBOT, rather than the TEMP DOF?

I think that there is a section of the thermal analysis guide that discusses doing just that, actually (though I haven't tried using through-thickness temperature variations in shell elements). Check section 2.4.1 of the thermal analysis guide. I think you'll want to create two sets of SURF152 elements (one for each side). I believe that you can set KEYOPT(11)=1 to tie the SURF152 elements to the TTOP DOF or KEYOPT(11) to tie them to the TBOT DOF. So... you should be able to set different (space node) temperatures on opposite sides of your foil, if desired... I think.

Here's a link to the documentation:

http://www1.ansys.com/customer/content/documentation/130/ans_the.pdf

 

[HVSmith]

Hi - thanks - yes, that's exactly what I wanted to do, converted into ANSYS speak, thanks very much! I'll give it a go!

HVSmith

 

[HVSmith]

So, I gave that a try and read the section of the manual you suggested. It all made sense in my head, then as soon as I transferred it to ANSYS the errors began...

I worked my way through them and now only have one (stubborn one) remaining:

*** ERROR *** CP = 838.766 TIME= 16:56:51
Radiation element 401 has zero or negative absolute temperature. The
offset temperature (TOFFST) and/or the uniform temperature (TUNIF)
must be positive.

*** ERROR *** CP = 838.766 TIME= 16:56:51
Radiation element 801 has zero or negative absolute temperature. The
offset temperature (TOFFST) and/or the uniform temperature (TUNIF)
must be positive.


This is similar to an error I posted above that miraculously went away with an ESEL,ALL followed by IC,ALL,TEMP,300 but unfortunately this didn't seem to solve the problem this time.

I tried to see what the radiation element 401 and 801 were and from going through line by line and analysing the output I think they are the surf152 on the extra nodes.

I've tried swapping TBOT/TTOP and +/- z for the extra node but nothing seems to do the trick.

If I set TOFFST,300 then the solution works and the numbers seem reasonable. But everything else I'm doing is in K, so TOFFST should be 0, right? This has massively confused me. I don't want to start adding in TOFFST,300 in when it is already in K and TOFFST,0 works fine when I use KEYOPT(3)=2 on SHELL131, so no through thickness.

Is there something obvious I'm missing?!

ET,2,SURF152
MP,EMIS,2,0.8
KEYOPT,2,5,1
KEYOPT,2,9,1
KEYOPT,2,11,2
R,2,1,5.678e-8

N,10000,0,0,-0.1

NSEL,S,NODE,,10000,10000
CMSEL,A,FOIL

TYPE,2
REAL,2
MAT,2

TOFFST,0

ESURF,10000
D,10000,TEMP,300

ESEL,ALL
IC,ALL,TEMP,300



ET,3,SURF152
MP,EMIS,3,0.8
KEYOPT,3,5,1
KEYOPT,3,9,1
KEYOPT,3,11,1
R,3,1,5.678e-8

N,20000,0,0,0.1

NSEL,S,NODE,,20000,20000
CMSEL,A,FOIL

TYPE,3
REAL,3
MAT,3

TOFFST,0

ESURF,20000
D,20000,TEMP,300

ESEL,ALL
IC,ALL,TEMP,300

So, as far as I can tell, the TOFFST,0 is OK (in my opinion) and the TUNIF is >0 since I have the IC,ALL,TEMO,300, no?

Thanks again!
HVSmith

 

[flash3780]

Did you successfully run your previous model changing only KEYOPT(11) to "1" or "2"?

 

[HVSmith]

Yes, changing to KEYOPT(11) = 1 or 2 and applying only a single SURF152 works absolutely fine, no problems.

It's only when I try to make two SURF152's (one for the top and one for the bottom) that I encounter the problem described in my post on the 10th of August.

I have still not found a method in which I can successfully apply SURF152 to the top layer and a separate SURF152 to the bottom layer without seeing the error.

Thanks for any ideas you may have!
HVSmith

 

[HVSmith]

Apologies - the above is incorrect...

KEYOPT(11) = 2 ----> and apply a single SURF152 works OK

KEYOPT(11) = 1 ----> and apply a single SURF152 does not work...
Errors are present for radiation elements 421 and 401 (error as described in post from 10th August)

And the errors are still present when I try to apply SURF152 to both sides.

I'm really confused.

I must be doing something really wrong. I think it might be to do with the way I'm defining the amount of layers now?

I used to make a SHELL131 act as a SHELL57, but now to get the TTOP and TBOT as you suggest I must change they KEYOPTS to allow temperature variation through the layer.

ET,1,SHELL131						!THERMAL SHELL
KEYOPT,1,3,1						!LINEAR TEMPERATURE VARIATION THROUGH LAYER

I used to define KEYOPT(4) to be equal to the amount of layers, which is equal to those in SECDATA, this is what I used to have:

KEYOPT,1,4,1						!NUMBER OF LAYERS (EQUAL TO SECDATA)	
R,1							 !REAL CONSTANT
SECTYPE,1,SHELL
SECDATA,1E-6,1

So now, because I don't just want one layer, I'd have to change these keyopts to reflect this, in some way? I have tried KEYOPT,1,4,2 to try and give me a top and a bottom layer? But then I'm not sure what to put in SECDATA, do I have to define the thickness of the further sections?

I've tried very many things over the past days, but nothing is working.

Any further insight appreciated, many thanks, most recent macro file attached (with many lines commented out) if it's more useful to view this way...
HVSmith

 

[flash3780]

It seems like it's having trouble attaching to the TTOP DOF. I wish I could help more, but I haven't really used the layered shell feature before. I think that you're looking at the SECDATA is the right direction.

Worst case, you can always put in a ticket with ANSYS.

 

[flash3780]

Post back with your results when you get it figured out; I'd be interested in knowing where the trouble stems from.

 

[HVSmith]

OK, so I have spent some time talking to ANSYS about this - their advice is to revert to using SFE RAD command on both load keys rather than SURF152s at all - how confusing!
I'll give this another go and see if anything different happens this time...

Thanks again for all of your input - hope these discussions may shed some light for others out there too...!

 

[HVSmith]

OK, so I convinced the guys at ANSYS that SFE,RAD wasn't working at all, they agree and apparently this is being fixed fo v14.0.

Instead of applying SURF152 to TTOP and TBOT, the same result can be obtained by using 2xSURF152's on the TEMP DOF - I'm not sure if this is just because I have such a thin thickness or also true in the general case. But they assure me this is fine for the above purpose, and is effectively radiation twice, so once from each surface. The answer it gives is also reasonable.

The other point I should make is that TOFFST should always be > 0. I guess this depends on how you define 0 as being +ve/-ve etc. I was under the impression TOFFST,0 would allow me to work in absolute temperature units (K). Apparently not. A value > 0 is required. I could have changed all my values to work with TOFFST,273 and work in degC, but instead I chose to use TOFFST,0.00001 so it is > 0 but only slightly (!), so the effect on the results is negligible...

Hopefully these tips will help someone else out there in the future!

 

[flash3780]

Thanks for posting back. I'd imagine that a thin sheet or foil would work fine by tying both sets of SURF152 elements to the TEMP DOF, which is simply the temperature at the midplane. If you were dealing with something thicker, you may be stuck using solid elements to achieve an accurate result.

That's an interesting limitation on the TOFFST command (must be greater than zero); I'll file that away in my brain.

Glad to hear that your model seems to be working for you. Best of luck.

//signed//
Christopher K. Hubley
Mechanical Engineer
Sunpower Incorporated
Athens, Ohio

 

[shr13]

Hello everybody,

I am quite new to the this radiation method(s) in Ansys and hence have some question to you guys...

1) In one of the post earlier, HVSmith mentioned that results of surf152 and different from Radiosity solver method. I could not understand what is the different in these methods.
Please correct me if I am wrong in my understanding about these methods-
Surf152 method: Here you solve radiation between surf152 element (pasted over other surface of a component) and an ambient nodal point such that the view-factor is always 1. Hence no separate calculation is required to calculate view-factor. Here do we need to assign RAD or RDSF tag on these surf152 elements using SFE command?


Radiosiy Method: Used to compute radiation between two surfaces. Here again, I think we need to create surf152 (or surf252? What is the difference?) over these surfaces, specify RDSF tag using SFE, specify TOFFSET, HEMIOPT setting. Now Ansys computes the viewfactor between these surfaces. Also, if the total viewfactor from a surface is less than 1, the remaining viewfactor is assigned to ambient with the ambient temperature specified (Thats why Ansys Mechanical - R13 GUI panel requires ambient temperature in the radiation command even if surface to surface radiation is selected).

Please confirm if my understanding is correct.
Also some additional queries -

1) When we can create surf152 using ESURF and use RAD or RDSF flage on Surf152 using SFE, why do we also have RSURF to create SURF252?
2) What is different between RAD and RDSF? I believe former is only for ambient radiation (Viewfactor always = 1), while latter is for surface-to-surface ( and ambient if Viewfactor < 1)
3) When surf-to-surf radiation for closed chamber is modeled through APDL commands, do we need to use SPCNOD and SPCTEMP? Also are surf152 enough to model this radiation using SFE,..RDSF flag?
And is this how the Radiation object (inserted in GUI panel of Ansys MEchanical 13 for surface-to-surface radiation) works?

Excuse me for sooo many questions, but I am stuck in radiation and any help will be highly appreciated.

Regards,
Shreesh