Unsteady Multirow computation

vishwasvemra · January 18, 2019, 02:26

Hi all,

I am trying to perform unsteady computations of two stage machine with domains R1, S1, R2, S2. Originally I had different blade counts(22,34,18,26) but to save computational time I have changed them to (24,36,18,24) method suggesting in Fine/Turbo theory guide p.179 i.e. domain scaling. With the modified blade counts and blades in ratio of(4,6,3,4) accounting to 60degree perfect sector for each row, I am trying to generate solution using Fine/Turbo 12.1.
Blade rotations are(-16000,0,24000,0).

The problem I encounter is solver exits with error message of 'SPEEDS OF ROTATION ARE DIFFERENT' (from .std file).
I am not sure what is causing solver failure. Is multistage computation not allowed? I found a discussion on this forum(https://www.cfd-online.com/Forums/nu...imulation.html) where Mr. Ji could perform computations with different rotation speeds.
Also I don't see NOFROT and IDOMSC expert parameters as suggested by Mr. Alain in same discussion forum.

Please help how to fix this.

Thankyou!

Kind Regards,
vishwas

colinda1 · January 21, 2019, 05:33

Hi Vishwas,

Did you define rotor/stator interfaces in between the rows? With the correct definition of rotor/stator interfaces there should indeed not be any problem to run such a domain scaling simulation.

The post you link to is from 2003 so the version used is already very old and FINE/Turbo has evolved a lot since then. For the way to define the number of time steps and the activation of domain scaling, please do take a look at the user guide of your FINE/Turbo release.

Best regards,
Colinda

vishwasvemra · January 21, 2019, 08:32

Hi Colinda,

Thanks for replying!

Yes, the rotor/stator interface is defined correctly by default, as the complete mesh is generated using Autogrid5.

While going through the Fine/Turbo user guide, on page 253, its mentioned ''The FINE™/Turbo solver allows only one rotation speed (combined with a zero rotation speed) when performing an unsteady simulation using the Domain Scaling method.''

From this text it seems like the solver doesn't allow multiple rotation speed.

For the shake of checking above statement, when I made the rotation speed of both rotor same the simulation starts without any issue.

Thankyou

Kind Regards,
Vishwas

colinda1 · January 21, 2019, 10:42

Sorry, I confused one important information in your email with the one in the link: in your case it is not counter-rotating with an equal magnitude of rotation. You have -16000 and 24000 (not 16000). This is indeed not possible.

January 18, 2019, 02:26	Unsteady Multirow computation	#1
vishwasvemra New Member Vishwas Verma Join Date: Mar 2016 Location: Mumbai India Posts: 25 Rep Power: 10	Hi all, I am trying to perform unsteady computations of two stage machine with domains R1, S1, R2, S2. Originally I had different blade counts(22,34,18,26) but to save computational time I have changed them to (24,36,18,24) method suggesting in Fine/Turbo theory guide p.179 i.e. domain scaling. With the modified blade counts and blades in ratio of(4,6,3,4) accounting to 60degree perfect sector for each row, I am trying to generate solution using Fine/Turbo 12.1. Blade rotations are(-16000,0,24000,0). The problem I encounter is solver exits with error message of 'SPEEDS OF ROTATION ARE DIFFERENT' (from .std file). I am not sure what is causing solver failure. Is multistage computation not allowed? I found a discussion on this forum(https://www.cfd-online.com/Forums/nu...imulation.html) where Mr. Ji could perform computations with different rotation speeds. Also I don't see NOFROT and IDOMSC expert parameters as suggested by Mr. Alain in same discussion forum. Please help how to fix this. Thankyou! Kind Regards, vishwas

January 21, 2019, 10:42		#4
colinda1 Senior Member Colinda Join Date: May 2012 Location: Brussels Posts: 153 Rep Power: 14	Sorry, I confused one important information in your email with the one in the link: in your case it is not counter-rotating with an equal magnitude of rotation. You have -16000 and 24000 (not 16000). This is indeed not possible. Hamidzoka, vishwasvemra and eng.abdul like this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
use unsteady computation results to other compute	turbinesv	CFX	2	October 12, 2008 09:55
Unsteady turbulence model computation	K.S. Cjang	Main CFD Forum	0	January 6, 2006 22:06
blade interface for unsteady computation	wang	CFX	1	June 11, 2003 19:33
blade interface for unsteady computation	wang	Main CFD Forum	2	June 11, 2003 13:02
Unsteady computation using sliding mesh	Ales Skotak	FLUENT	0	September 7, 2000 08:03

January 21, 2019, 05:33		#2
colinda1 Senior Member Colinda Join Date: May 2012 Location: Brussels Posts: 153 Rep Power: 14	Hi Vishwas, Did you define rotor/stator interfaces in between the rows? With the correct definition of rotor/stator interfaces there should indeed not be any problem to run such a domain scaling simulation. The post you link to is from 2003 so the version used is already very old and FINE/Turbo has evolved a lot since then. For the way to define the number of time steps and the activation of domain scaling, please do take a look at the user guide of your FINE/Turbo release. Best regards, Colinda

January 21, 2019, 08:32		#3
vishwasvemra New Member Vishwas Verma Join Date: Mar 2016 Location: Mumbai India Posts: 25 Rep Power: 10	Hi Colinda, Thanks for replying! Yes, the rotor/stator interface is defined correctly by default, as the complete mesh is generated using Autogrid5. While going through the Fine/Turbo user guide, on page 253, its mentioned ''The FINE™/Turbo solver allows only one rotation speed (combined with a zero rotation speed) when performing an unsteady simulation using the Domain Scaling method.'' From this text it seems like the solver doesn't allow multiple rotation speed. For the shake of checking above statement, when I made the rotation speed of both rotor same the simulation starts without any issue. Thankyou Kind Regards, Vishwas