Problem with massflow BC at inlet

MrNavierStokes · November 3, 2016, 08:41

Hello CFD community,

I am currently working on a simulation of a radial turbine. At the inlet I have a massflow boundary condition for which the massflow is calculated through a "p-regulator" loop I implemented in CFX using expressions. I set the required pressure ratio between two planes in the inlet and outlet section respectively and the p-regulator calculates the appropriate massflow which is then set at the inlet. BUT the "iteratively" calculated massflow mdotIter is only being set after 50 iterations. For the first 50 iterations a constant massflow of e.g. 0.04 kg/s is used.

The loop works fine for one case. After I changed the geometry of the inlet pipe however it stops working. Eventhough the loop should start after 50 iterations, no massflow passes through the turbine (monitor point for massflow equals 1e-016 kg/s) from the beginning. After 50 iterations, when the regulator loop starts, the simulation crashes. My expression for the inlet massflow is the following:

BC mdot = if(citern<50,0.04 [kg s^-1],mdotIter)

I believe the terminology is correct, because I don't recieve any errors from CFX and it works perfectly fine for the first case.
I also suppose there is nothing wrong with the new inlet geometry, because when I change the inlet BC to:

BC mdot = 0.04 [kg s-^1]

it works again.
I hope someone is able to help me, because this problem really confuses me.

Thank you guys very much!

Regards, Alex.

Opaque · November 3, 2016, 13:36

Your simulation crashes/fail, how does it fail ?

The error message might help understand the source of the problem.

omidiut · November 5, 2016, 04:00

Dear Alex
Hi

could you explain more about your problem?
what is 'mdotIter' in your expression?

MrNavierStokes · November 7, 2016, 03:55

Hello guys,

first of all thanks for your help!

@Opaque: Of course you're right, sorry. Should've said something about the error message I receive from the solver. It only says:

Fatal Overflow in linear Solver

after 50 iterations, that ist when my inlet BC switches from a constant massflow of 0.04 kg s^-1 to the regulator-loop massflow mdotIter.

@omidiut: I work on a turbine stage simulation and would like to do a comparison between the CFD and measurement results. In order to do that I have to figure out the correct position of the VTG blades (Variable Turbine Geometry, stationary guide vanes to regulate the turbine massflow) that has been used during the measurements. In order to find the right position and calculate the correct turbine map (they differ dependent on the VTG position), I set a certain pressure ratio (measurement data) between two section planes of my domain and the loop iterates until the massflow at the inlet corresponds to this pressure ratio. Of course I could just set the massflow at the inlet since I have the data, but this wouldn't work for the whole map, because a massflow inlet BC would cause instabilities at high pressure ratios. Now the inlet BC mdotIter is part of a p-regulator loop. It is set as a BC after 50 iterations. Before that a constant massflow is used. Know the strange thing is, that it worked perfectly fine for one case, but after I changed the inlet pipe (it has a slightly different shape) the loop does not work anymore. For a constant massflow the calculation runs normally, but as soon as i write:

BC mdot = if(citern<50,0.04 [kg s^-1],mdotIter)

instead of

BC mdot = 0.04 [kg s-^1]

it crashes with the error message above.

Thank you guys for your time and help.

Regards, Alex.

November 3, 2016, 08:41	Problem with massflow BC at inlet	#1
MrNavierStokes New Member Alex Join Date: Jun 2015 Posts: 22 Rep Power: 0	Hello CFD community, I am currently working on a simulation of a radial turbine. At the inlet I have a massflow boundary condition for which the massflow is calculated through a "p-regulator" loop I implemented in CFX using expressions. I set the required pressure ratio between two planes in the inlet and outlet section respectively and the p-regulator calculates the appropriate massflow which is then set at the inlet. BUT the "iteratively" calculated massflow mdotIter is only being set after 50 iterations. For the first 50 iterations a constant massflow of e.g. 0.04 kg/s is used. The loop works fine for one case. After I changed the geometry of the inlet pipe however it stops working. Eventhough the loop should start after 50 iterations, no massflow passes through the turbine (monitor point for massflow equals 1e-016 kg/s) from the beginning. After 50 iterations, when the regulator loop starts, the simulation crashes. My expression for the inlet massflow is the following: *BC mdot = if(citern<50,0.04 [kg s^-1],mdotIter)* I believe the terminology is correct, because I don't recieve any errors from CFX and it works perfectly fine for the first case. I also suppose there is nothing wrong with the new inlet geometry, because when I change the inlet BC to: BC mdot = 0.04 [kg s-^1] it works again. I hope someone is able to help me, because this problem really confuses me. Thank you guys very much! Regards, Alex.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Inlet problem for turbulent flow in cavity	Sebastian A	FLUENT	0	January 21, 2016 06:54
Problem with mass flow inlet	vainilreb	OpenFOAM Running, Solving & CFD	4	April 8, 2013 07:41
Flux Problem at Inlet	piyushsharma0487	Fluent UDF and Scheme Programming	9	April 19, 2012 03:57
Problem with gas inlet in an external flow	vitor.monjardim	ANSYS	0	January 30, 2012 07:35
Transient simulation (Problem with massflow and Pressure)	ChristianF	CFX	4	August 22, 2011 22:42

November 3, 2016, 13:36		#2
Opaque Senior Member Join Date: Jun 2009 Posts: 1,873 Rep Power: 33	Your simulation crashes/fail, how does it fail ? The error message might help understand the source of the problem.

November 5, 2016, 04:00		#3
omidiut Senior Member omid Join Date: Apr 2011 Posts: 105 Rep Power: 15	Dear Alex Hi could you explain more about your problem? what is 'mdotIter' in your expression?

November 7, 2016, 03:55		#4
MrNavierStokes New Member Alex Join Date: Jun 2015 Posts: 22 Rep Power: 0	Hello guys, first of all thanks for your help! @Opaque: Of course you're right, sorry. Should've said something about the error message I receive from the solver. It only says: Fatal Overflow in linear Solver after 50 iterations, that ist when my inlet BC switches from a constant massflow of 0.04 kg s^-1 to the regulator-loop massflow mdotIter. @omidiut: I work on a turbine stage simulation and would like to do a comparison between the CFD and measurement results. In order to do that I have to figure out the correct position of the VTG blades (Variable Turbine Geometry, stationary guide vanes to regulate the turbine massflow) that has been used during the measurements. In order to find the right position and calculate the correct turbine map (they differ dependent on the VTG position), I set a certain pressure ratio (measurement data) between two section planes of my domain and the loop iterates until the massflow at the inlet corresponds to this pressure ratio. Of course I could just set the massflow at the inlet since I have the data, but this wouldn't work for the whole map, because a massflow inlet BC would cause instabilities at high pressure ratios. Now the inlet BC mdotIter is part of a p-regulator loop. It is set as a BC after 50 iterations. Before that a constant massflow is used. Know the strange thing is, that it worked perfectly fine for one case, but after I changed the inlet pipe (it has a slightly different shape) the loop does not work anymore. For a constant massflow the calculation runs normally, but as soon as i write: *BC mdot = if(citern<50,0.04 [kg s^-1],mdotIter)* instead of BC mdot = 0.04 [kg s-^1] it crashes with the error message above. Thank you guys for your time and help. Regards, Alex.