Water wheel simulation

Ema40 · November 11, 2014, 03:17

Hi everybody.

I am simulating a model of water wheel (a breastshot water wheel). The water enters the buckets and pushing on the blades make the wheel rotate. The simulation converges and I have been able to simulate the multiphase fluid (air+water, since the wheel is in a open channel) and the rotating domain. The water level inside the buckets is more or less the same as that estimated theoretically (so it is ok), the exiting flow rate is the same as the entering one (so the mass conservation is ok), etc. Despite this, the torque and the moment coefficient are very lower than that estimated theoretically and measured by experiment. I have used different turbulence models (k-eps standard, k-eps RNG, k-eps Realizable, k-omega, k-omega SST laminar, Reynold stress ), different options (low reynolds correction, curvature correction, etc.), different numerical schemes and different mesh sizing, but I have obtained always the same results.

Can anyone of the members help me with some advices?
Thank you

Kokemoor · November 11, 2014, 10:49

If you compare the total (kinetic + potential) energy of the water at the inlet to the outlet, does that correspond at all to either your CFD results or theoretical ones?

(mass flow rate)*(gravity)*(height change) + 1/2*(mass flow rate)*((average velocity in)^2-(average velocity out)^2) = (torque)*(wheel rotation rate) + (energy losses)

This is a good sort of way to check that torque is being measured correctly.

Ema40 · November 12, 2014, 04:22

Thank you. The problem is that I calculate the difference of energy, but this difference is due both to the losses both to the torque and I don't know in what proportion. I compare the numerical torque to that measured experimentally and it is wrong. I can't understand why.

Thank you

Kokemoor · November 12, 2014, 13:25

Quote:

Originally Posted by Ema40

The problem is that I calculate the difference of energy, but this difference is due both to the losses both to the torque and I don't know in what proportion.

In the theoretical case, you should know the proportion. Can you compare your inlet and exit flow velocities to the theory?

Ema40 · November 13, 2014, 02:35

Thank you.
Yes.
The problem is that the numerical torque is very different to the experimental and theoretical torque, despite the other quantities are right (water levels, velocity, etc.)

I have discovered few hours ago that the static pressure inside the wheel, in zones where only air is present, is -5000 Pa and I think it is totally wrong!! The wheel is in an open environment and the air is lightweigh...

November 11, 2014, 03:17	Water wheel simulation	#1
Ema40 Senior Member Join Date: Oct 2014 Posts: 124 Rep Power: 11	Hi everybody. I am simulating a model of water wheel (a breastshot water wheel). The water enters the buckets and pushing on the blades make the wheel rotate. The simulation converges and I have been able to simulate the multiphase fluid (air+water, since the wheel is in a open channel) and the rotating domain. The water level inside the buckets is more or less the same as that estimated theoretically (so it is ok), the exiting flow rate is the same as the entering one (so the mass conservation is ok), etc. Despite this, the torque and the moment coefficient are very lower than that estimated theoretically and measured by experiment. I have used different turbulence models (k-eps standard, k-eps RNG, k-eps Realizable, k-omega, k-omega SST laminar, Reynold stress ), different options (low reynolds correction, curvature correction, etc.), different numerical schemes and different mesh sizing, but I have obtained always the same results. Can anyone of the members help me with some advices? Thank you

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Solid object falling in water tank simulation	cassini83	CFX	3	August 9, 2021 17:05
Vertical axis water turbine simulation	SHIKHA BHUYAN	CFX	1	September 15, 2014 07:40
Simulation of Water Turbine with CFX	hydroturbines	CFX	0	July 23, 2014 15:58
Simulation of tiltable bucket with water	jurij	CFX	2	August 21, 2012 20:49
Mass conservation problem in mixing tank multiphase simulation	rockewan	FLUENT	0	April 6, 2010 12:34

November 11, 2014, 10:49		#2
Kokemoor Senior Member Andrew Kokemoor Join Date: Aug 2013 Posts: 122 Rep Power: 13	If you compare the total (kinetic + potential) energy of the water at the inlet to the outlet, does that correspond at all to either your CFD results or theoretical ones? (mass flow rate)(gravity)(height change) + 1/2(mass flow rate)((average velocity in)^2-(average velocity out)^2) = (torque)*(wheel rotation rate) + (energy losses) This is a good sort of way to check that torque is being measured correctly.

November 12, 2014, 04:22		#3
Ema40 Senior Member Join Date: Oct 2014 Posts: 124 Rep Power: 11	Thank you. The problem is that I calculate the difference of energy, but this difference is due both to the losses both to the torque and I don't know in what proportion. I compare the numerical torque to that measured experimentally and it is wrong. I can't understand why. Thank you

November 13, 2014, 02:35		#5
Ema40 Senior Member Join Date: Oct 2014 Posts: 124 Rep Power: 11	Thank you. Yes. The problem is that the numerical torque is very different to the experimental and theoretical torque, despite the other quantities are right (water levels, velocity, etc.) I have discovered few hours ago that the static pressure inside the wheel, in zones where only air is present, is -5000 Pa and I think it is totally wrong!! The wheel is in an open environment and the air is lightweigh...