Is this a bug in flow simulation? I noticed this issue in 2015 SP4 and upgraded to 2016 SP4 and this version still has the same issue. I really don't get it. Any help would be appreciated!
I should have indicated that this extreme difference occurs when simulating a fanned heat sink at high-altitude (only 17,873 Pa of pressure).
So when I leave "reference density" for the fan definition unchecked, I get a deltaT of around 60 degK whereas if I assign a reference density of 1.2 kg/m^3 and run the exact same simulation, I get a deltaT of around 5500 degK ! According to the help file, the default reference density is 1.2 kg/m^3 so this change should have made no difference at all in the result, but it makes the results extremely different. Does anybody know if the new 2017 version has this issue? Thanks!
I am sure you checked but r u sure it isn't a units thing? it just has that sort of sympton......
Bill - I agree that it does seem to have symptoms of a units thing. I've check everything I know of that I can in that regards (I even changed my model from IPS to MKS to see if it was somehow it was a bug in that conversion, but that had no effect). It could be behind the scenes (i.e. in the code) where the issue resides in which case it's out of my control.
In that case you could try increasing or decreasing it a thousandfold to see if you can replicate the original results closely. SolidWorks is all metric on the inside, so Freedom Units probably won't be the problem.
Through trial and error, if I use 0.23 kg/m^3 as the reference density, I get close to the results that I get when I use the default (which is supposed to be 1.2 kg/m^3). This doesn't make any sense...
Tested what you mentioned and changed the reference density for a fan inside an electronic enclosure -> not that much of difference in results. Can you upload your model/Sim or a simplified version?
Thanks Siavash. This was for an external inlet fan, not an internal fan. And, it certainly becomes more evident at lower pressures (such as the 17,873 Pa referenced above). I'll see if maybe I can upload... is the custom fan data included with the *.sldprt file?
Here's the file. You'll want to concentrate on the project called "al_41000ft" . If the custom fan data does indeed come through, I left the reference density at .23 kg/m^3 because this is what seemed to match the default density better. Thanks !
Does anybody know if this issue has ever fixed this issue? Thanks.
I may be wrong but I think the issue is that putting in a Fan curve without a reference density will cause a certain flow rate for a certain pressure drop regardless of the density of the fluid used in the project whereas specifying a reference density tells the software OK this fan curve data was measured under these conditions so for any other condition you must compensate with fan laws.
I think it essentially adds calculation steps when having that option checked off. Since your fan is being run at higher elevations, with that option checked off, it then compensates for the much lower pressure of the air. Think of an extreme scenario where a fan curve is measured at a density of fluid where the fluid is thick as sludge. The fan has to work really hard to put out the CFM in that scenario. Then this mystery fluid at elevation gets thin as air. For the same amount of pressure difference from inlet to outlet the fan is going to be able to push a lot more fluid through. So in your case you have more fluid being pushed through and the fluid is at a lower temperature so you get a higher delta T.
Can anyone else verify that what I am saying is legit? Specifically the part about the additional calculation steps Flow Simulation would do with the option checked off?
This article gives a little insight although it is related to FloTherm, not Solidworks Flow Simulation, the same principles apply:
How to create high altitude conditions using FloTHERM
Retrieving data ...