Set different viscosities for different areas

eldudereeno · January 22, 2009, 07:19

Hello everybody.

I would like to know weather it is possible in openFOAM to define different viscosities for different areas?

I would like to simulate the flow of magnetorheological fluids which change there viscosities passing a magnetic field.

I am using the HerschelBulkely model, where i have a tau0.
In the areas with magnetic field it would be great if i could increase the tau0 regarding to my field.

Any ideas?

Thanks for possible answers

Best Regards

grtabor · January 22, 2009, 08:31

Should be fine - make tau0 a volScalarField and increase it where necessary. Should work.

Gavin

eldudereeno · January 22, 2009, 09:56

Sounds good!

Thanks a lot, Gavin!

But could you please explain it for me?
I am very new in OF.

Where shall i do it ? In the transport model (HerschelBulkley) or in the transportProperties ?

grtabor · January 26, 2009, 04:36

Ouch. Having had a chance to have a look at it it is a little more complicated. HerschelBulkley is a class derived from viscosityModel making it run-time selectable (which is nice, of course). To change it you would need to create your own class; probably copy the header and class files across to a new directory, rename it, and then alter it appropriately. Having done this you would need to recompile viscosityModels. The alternative would be to start from scratch and implement your own viscosity law into icoFoam (you could probably extract quite a lot of the appropriate code from HerschelBulkley). How is your C++ coding??

Either way, you need to change tau0 to a volScalarField. Instead of declaring it as a dimensionedScalar and reading it from the transportProperties dictionary you would need to create it as a volScalarField and read it in from a file in the timestep directory via an IOobject. To see how to do this, look at the declaration of the field "p" in createFields.H

Gavin

eldudereeno · February 10, 2009, 11:27

Thank you Gavin!

I made tau0 a colScalarField (which indeed took a long time, hehe)
Via setFields i was able to declare at in the neccesary places.

But the transport modell is of course still using the tau0 of my transportProperties.

My C++ Coding is pretty poor.
I copied the HerschelBulkely class and renamed it.
Than i was even able to compile it and make it usable for my calculation.

But i have absolutely no idea how to change:
read tau0 not as a fixed variable but as a field.

Peter

January 22, 2009, 07:19	Hello everybody. I would li	#1
eldudereeno New Member Peter Weinhaus Join Date: Mar 2009 Posts: 5 Rep Power: 17	Hello everybody. I would like to know weather it is possible in openFOAM to define different viscosities for different areas? I would like to simulate the flow of magnetorheological fluids which change there viscosities passing a magnetic field. I am using the HerschelBulkely model, where i have a tau0. In the areas with magnetic field it would be great if i could increase the tau0 regarding to my field. Any ideas? Thanks for possible answers Best Regards

January 22, 2009, 08:31	Should be fine - make tau0 a v	#2
grtabor Senior Member Gavin Tabor Join Date: Mar 2009 Posts: 181 Rep Power: 17	Should be fine - make tau0 a volScalarField and increase it where necessary. Should work. Gavin

January 22, 2009, 09:56	Sounds good! Thanks a lot,	#3
eldudereeno New Member Peter Weinhaus Join Date: Mar 2009 Posts: 5 Rep Power: 17	Sounds good! Thanks a lot, Gavin! But could you please explain it for me? I am very new in OF. Where shall i do it ? In the transport model (HerschelBulkley) or in the transportProperties ?

January 26, 2009, 04:36	Ouch. Having had a chance to h	#4
grtabor Senior Member Gavin Tabor Join Date: Mar 2009 Posts: 181 Rep Power: 17	Ouch. Having had a chance to have a look at it it is a little more complicated. HerschelBulkley is a class derived from viscosityModel making it run-time selectable (which is nice, of course). To change it you would need to create your own class; probably copy the header and class files across to a new directory, rename it, and then alter it appropriately. Having done this you would need to recompile viscosityModels. The alternative would be to start from scratch and implement your own viscosity law into icoFoam (you could probably extract quite a lot of the appropriate code from HerschelBulkley). How is your C++ coding?? Either way, you need to change tau0 to a volScalarField. Instead of declaring it as a dimensionedScalar and reading it from the transportProperties dictionary you would need to create it as a volScalarField and read it in from a file in the timestep directory via an IOobject. To see how to do this, look at the declaration of the field "p" in createFields.H Gavin

February 10, 2009, 11:27	Thank you Gavin! I made tau	#5
eldudereeno New Member Peter Weinhaus Join Date: Mar 2009 Posts: 5 Rep Power: 17	Thank you Gavin! I made tau0 a colScalarField (which indeed took a long time, hehe) Via setFields i was able to declare at in the neccesary places. But the transport modell is of course still using the tau0 of my transportProperties. My C++ Coding is pretty poor. I copied the HerschelBulkely class and renamed it. Than i was even able to compile it and make it usable for my calculation. But i have absolutely no idea how to change: read tau0 not as a fixed variable but as a field. Peter

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SIMPLE: instability at low viscosities	Lionel S.	Main CFD Forum	7	August 7, 2014 18:01
Setting viscosities in Interior zones	philip	FLUENT	8	September 16, 2005 14:07
face areas UDF	Andrew Garrard	FLUENT	0	October 19, 2004 09:55
emerging areas in cfd?	Red	Main CFD Forum	5	January 25, 2003 09:58
areas of no flow	Eve Chapper	CFX	1	December 16, 1999 07:02