Fan Curve Model with CHT simulation

mariconeagles96 · April 21, 2015, 02:55

hi all!

Im doing some CHT on one model project and was thinking of using a fan curve (momentum source) and compare it with my actual fan blade model project. I already have the results and it was in good relation with the actual thermal test.

So now, I will use the fan curve using the momentum source technique and had found some good points here in this thread. However I did encounter some problems on my own. First I compared the fan curve and simulated model using the momentum source option and had good relation with the fan curve data. I also made a simple blockage to check if the fan curve reacts to whatever resistance that is in front of it and results were ok.

fansub: CEL
x=0 [kg m^-2 s^-2]
y=0 [kg m^-2 s^-2]
z=fanQ*rho*velF/volF

where fanQ = fancurve(delP)
delP = abs(fanOUT-fanIN) <-- these are the faces of the fan
velF = volumeAve(Velocity)@fansub
volF = fanArea*FanLength

As for the fan curve:
Argument: Pa
Result Unit: m^3

Now using these expressions and put it on my actual project, my components temperature were very high and flowrate was about the same.. maybe 10% lower from my previous actual blade simulation.

Can you guys help me out? thanks a lot! =)

ghorrocks · April 21, 2015, 07:08

Are you saying that the momentum source approach appeared to generate pretty much the same flow, but the temperature of the CHT body you are interested in 10C lower?

Have you compared the flow field around the CHT body in detail? Had a look at the differences in heat transfer rate over the CHT body surface?

Have you also checked that your simulation is accurate - mesh resolution, convergence and time step?

mariconeagles96 · April 21, 2015, 21:16

hi Ghorrocks!

yes pretty much i've tried everything on the CHT side (mesh, boundary layers) and the first one i modelled (with the actual fan blade) gave me good results compared to the thermal test data. When I tried replacing the fan blade model with a simple cylinder and use momentum source, then my airflow whent down a little by about 10% and temperature went very high about +100C.

Strange thing was that when I use the momentum source option in a windtunnel fan model (using the fan curve) it gave me good airflow results versus the fan curve data. Although I tried onlyt 2 points - open delivery and aorund 50% airflow. So I think the equations previously mentioned were ok. I just dont know if its applicable when you put that into a big and very high resistance system. Do you have any suggestion?

Thanks! =)

ghorrocks · April 21, 2015, 23:32

It sounds like the first thing you should do is a careful comparison of your fan model to the momentum source model. There is no point looking at the model until you have confirmed that.

mariconeagles96 · April 22, 2015, 01:21

yes glenn, i am doing it right now and trying to check the difference between the actual 3D fan model and momentum source model only.

One question though, in comparing the model...lets say i have a fan diameter of 120mm and have its fan characteristic curve (based on spec sheet). But in reality, the actual fan blade geometry diameter is only 110mm. Shall I make a comparison using the 110mm diameter simple cylinder model and use the fan curve or use the spec sheet diameter of 120mm?

Thanks!

ghorrocks · April 22, 2015, 01:31

You will have to make that decision. I have no idea what the fan manufacturer is talking about when he says "120mm" - it is not necessarily the diameter of the rotor.

At the end of the day obviously you want to compare your performance curve against the fan geometry which produced that performance curve.

April 21, 2015, 02:55	Fan Curve Model with CHT simulation	#1
mariconeagles96 Member Join Date: Jun 2011 Posts: 32 Rep Power: 14	hi all! Im doing some CHT on one model project and was thinking of using a fan curve (momentum source) and compare it with my actual fan blade model project. I already have the results and it was in good relation with the actual thermal test. So now, I will use the fan curve using the momentum source technique and had found some good points here in this thread. However I did encounter some problems on my own. First I compared the fan curve and simulated model using the momentum source option and had good relation with the fan curve data. I also made a simple blockage to check if the fan curve reacts to whatever resistance that is in front of it and results were ok. fansub: CEL x=0 [kg m^-2 s^-2] y=0 [kg m^-2 s^-2] z=fanQrhovelF/volF where fanQ = fancurve(delP) delP = abs(fanOUT-fanIN) <-- these are the faces of the fan velF = volumeAve(Velocity)@fansub volF = fanArea*FanLength As for the fan curve: Argument: Pa Result Unit: m^3 Now using these expressions and put it on my actual project, my components temperature were very high and flowrate was about the same.. maybe 10% lower from my previous actual blade simulation. Can you guys help me out? thanks a lot! =)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Water subcooled boiling	Attesz	CFX	7	January 5, 2013 03:32
Fan simulation divergence	ayothicfd	CFX	18	July 20, 2012 05:53
Fan heater model: what turbulence source to use?	andy20	CFX	7	March 3, 2008 16:42
Fan slection--ystsem pressure curve	MHK	FLUENT	0	September 12, 2007 13:46
Desperate ! Fan model !!!	Boris	FLUENT	0	January 30, 2003 00:31

April 21, 2015, 07:08		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Are you saying that the momentum source approach appeared to generate pretty much the same flow, but the temperature of the CHT body you are interested in 10C lower? Have you compared the flow field around the CHT body in detail? Had a look at the differences in heat transfer rate over the CHT body surface? Have you also checked that your simulation is accurate - mesh resolution, convergence and time step?

April 21, 2015, 21:16		#3
mariconeagles96 Member Join Date: Jun 2011 Posts: 32 Rep Power: 14	hi Ghorrocks! yes pretty much i've tried everything on the CHT side (mesh, boundary layers) and the first one i modelled (with the actual fan blade) gave me good results compared to the thermal test data. When I tried replacing the fan blade model with a simple cylinder and use momentum source, then my airflow whent down a little by about 10% and temperature went very high about +100C. Strange thing was that when I use the momentum source option in a windtunnel fan model (using the fan curve) it gave me good airflow results versus the fan curve data. Although I tried onlyt 2 points - open delivery and aorund 50% airflow. So I think the equations previously mentioned were ok. I just dont know if its applicable when you put that into a big and very high resistance system. Do you have any suggestion? Thanks! =)

April 21, 2015, 23:32		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	It sounds like the first thing you should do is a careful comparison of your fan model to the momentum source model. There is no point looking at the model until you have confirmed that.

April 22, 2015, 01:21		#5
mariconeagles96 Member Join Date: Jun 2011 Posts: 32 Rep Power: 14	yes glenn, i am doing it right now and trying to check the difference between the actual 3D fan model and momentum source model only. One question though, in comparing the model...lets say i have a fan diameter of 120mm and have its fan characteristic curve (based on spec sheet). But in reality, the actual fan blade geometry diameter is only 110mm. Shall I make a comparison using the 110mm diameter simple cylinder model and use the fan curve or use the spec sheet diameter of 120mm? Thanks!

April 22, 2015, 01:31		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You will have to make that decision. I have no idea what the fan manufacturer is talking about when he says "120mm" - it is not necessarily the diameter of the rotor. At the end of the day obviously you want to compare your performance curve against the fan geometry which produced that performance curve.