rotating crankshaft in a flow domain

rakeshpe · February 25, 2019, 00:47

Hi,

I have to model a flow domain inside a compressor in which a crankshaft is rotating.

I have extracted the flow volume and my idea is to define a cylindrical domain around the crankshaft (the shape of crankshaft is not symmetric) and give rotation to that domain and keep the remaining flow domain stationary.

anyone think it will work? or is there any better idea to give rotation for the crankshaft.

and how can i physically rotate the mesh?

urosgrivc · February 25, 2019, 01:08

You could put your crankshaft in a rotating domain (cylindrical domain with a cutout of your crankshaft)

Then you are able to make an interface between the rotating domain and a stationary one. (mesh of the same size on both sides of the interface is prefered, but different sizes work too.)

When this is done, the rotating mesh can be rotated via expressions or whatever, and a GGI mesh connection interface will take care of the flow and other variables through the interface.
This type of connection supports a rotating mesh during a transient run.

What I don't yet know is how you will simulate connecting rod if you have one or not.

rakeshpe · February 26, 2019, 03:38

Thankyou Urosgrivc.

i tried that way. That will impose the rotational component along the interface.

As my rotating component is not symmetric (not cylindrical), it may exert some pressure over the flow domain periodically. for that i need a physically rotating mesh i guess. please post your suggestion

urosgrivc · February 26, 2019, 03:46

Yes you are talking about a transient simulation, because steady-state won't be enough for you because of the symmetry issues.
Your values or results will change according to the rotation of the shaft, the mesh can rotate yes.
Be prepared for long computational times though if you need a transient one, but this depends on what you are trying to achieve

I don't like to call it physically rotate as everything is numerical.

rakeshpe · February 26, 2019, 04:06

ok let me try again.. thankyou

Quote:

Originally Posted by urosgrivc

I don't like to call it physically rotate as everything is numerical.

hehe you are right. but my frame of referance is CFX-post

February 25, 2019, 00:47	rotating crankshaft in a flow domain	#1
rakeshpe New Member rakesh p e Join Date: Aug 2016 Posts: 9 Rep Power: 10	Hi, I have to model a flow domain inside a compressor in which a crankshaft is rotating. I have extracted the flow volume and my idea is to define a cylindrical domain around the crankshaft (the shape of crankshaft is not symmetric) and give rotation to that domain and keep the remaining flow domain stationary. anyone think it will work? or is there any better idea to give rotation for the crankshaft. and how can i physically rotate the mesh?

February 26, 2019, 03:46		#4
urosgrivc Senior Member urosgrivc Join Date: Dec 2015 Location: Slovenija Posts: 365 Rep Power: 11	Yes you are talking about a transient simulation, because steady-state won't be enough for you because of the symmetry issues. Your values or results will change according to the rotation of the shaft, the mesh can rotate yes. Be prepared for long computational times though if you need a transient one, but this depends on what you are trying to achieve I don't like to call it physically rotate as everything is numerical. Last edited by urosgrivc; February 26, 2019 at 07:54.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Pressure distribution on a wall	darazsbence	CFX	17	October 6, 2015 10:38
Wrong flow in ratating domain problem	Sanyo	CFX	17	August 15, 2015 06:20
injection problem	Mark New	FLUENT	0	August 4, 2013 01:30
Rotating Domain	Szczepan	CFX	8	January 19, 2013 09:27
rotating domain in rotating domain, different axis	Robert Stringer	CFX	3	December 4, 2006 07:04

February 25, 2019, 01:08		#2
urosgrivc Senior Member urosgrivc Join Date: Dec 2015 Location: Slovenija Posts: 365 Rep Power: 11	You could put your crankshaft in a rotating domain (cylindrical domain with a cutout of your crankshaft) Then you are able to make an interface between the rotating domain and a stationary one. (mesh of the same size on both sides of the interface is prefered, but different sizes work too.) When this is done, the rotating mesh can be rotated via expressions or whatever, and a GGI mesh connection interface will take care of the flow and other variables through the interface. This type of connection supports a rotating mesh during a transient run. What I don't yet know is how you will simulate connecting rod if you have one or not.

February 26, 2019, 03:38		#3
rakeshpe New Member rakesh p e Join Date: Aug 2016 Posts: 9 Rep Power: 10	Thankyou Urosgrivc. i tried that way. That will impose the rotational component along the interface. As my rotating component is not symmetric (not cylindrical), it may exert some pressure over the flow domain periodically. for that i need a physically rotating mesh i guess. please post your suggestion