Air Flow results comparison

[johnab]

I was just comparing air flow pressure drop for intake system - analytical results (start CFD- by out side vendor) vs actual testing

For the same flow - Analytical results showing more pressure drop (say 7 kPa) while actual testing shows around 3 kPa?

Does it mean Analytical Flow calculation includes Static & Dynamic pressure drop while actual testing have only Dynamic?

 

[katmar]

If you are using a pitot type device, then it will convert the dynamic head back to static pressure and you would see a lower measured DP than the analytical calc would predict. But a pressure tapping point that is perpendicular to the flow direction will measure both, giving the same result as the analytical calc.

In atmospheric air (101 kPa abs, 25 deg C) a velocity of around 80 or 90 m/s would be required to explain the difference. Is this what you have?

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[zdas04]

You're kidding, right? 7 kPa and 3 kPa are really the same number when you're talking about the difference between CFD and actual (it is 0.6 psi using Katmar's Uconeer). A CFD model will never get the conduit roughness exactly right in every section. Turbulent effects are averaged. Installation effects (like an irregularity in a bend) can't really be predicted. Whenever I run a model I tend to shade my choices of options so that there won't be unpleasant surprises.

You didn't provide enough information to really be able to help much. Since this inlet plenum is a big enough deal for you to be running CFD models on it and apparently the equipment being supplied should work fine with 7 kPa dP, I'd think it would work even better with 3 kPa dP. Looks like your modeler did a great job.

Katmar,
What Reynolds Number are you basing your velocities on? I don't know how to relate a real-world dP to a velocity without a mass flow rate. Remember that Bernoulli assumed zero friction.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[katmar]

First off, I must agree with David that the difference between the theoretical and measured values is very small, and may actually be meaningless. So my theorising must be seen as exactly that - theorising only. Perhaps the CFD engineer gave some confidence limits for his calcs?

I estimated the velocity using the relationship (Darcy Weisbach)
Pressure drop = ( f_DL/D + K_inlet + K_exit ) x ([ρ]V²)/2

I assumed that the first two terms (the friction element and the inlet loss) made up the 3 kPa that johnab measured, and that the third element (K_exit) is the "missing" 4 kPa of dynamic head (i.e. the energy to accelerate the air). This third element is what is usually termed the "exit loss" and has a value of 1.0, although of course the acceleration actually happens at the inlet.

Guessing the density of the air to be 1.1 kg/m³ I had to solve for V in
4000 (Pascal) = 1.0 x ( 1.1 x V² ) / 2
This gives V = 85 m/s

By this reasoning the velocity estimation is independent of Reynolds number, surface roughness or flow length - these are all part of the 3 kPa which I have ignored.

Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[zdas04]

Harvey,
Clever, as always. I'm not sure that you can say that the inlet constant is included in the 3 kPa and the exit constant isn't, but it gives you numbers that are more or less defensible. I was sticking on the idea that the mass flow rate for the two cases has to be the same (which would yield different velocities for the two cases).

I always cringe when I see the Darcy Weisbach equation since they started with Bernoulli (which started with Euler, but that is another story) which explicitly assumes zero friction and constant density. Then they add back in a friction term and a couple of "fudge factors" that are rarely rigorously determined. I don't know a better way, but when you take into account the hoops the Euler and Bernoulli went to before the got to the famous Bernoulli Equation it makes me uneasy to add back the stuff they assumed was gone. Just a quirk.

David