Heat transfer CHT in ansys CFX

shaheen · August 11, 2013, 23:47

Hi,

I have modelled a pipe in which oil is flowing at 85 deg C with 1m/s speed.There is a heater cable on the pipe to heat the oil as it flows.The heater provides 100 W/m output.The pipe and heater cable is insulated with additional layer of rockwool.

I have created seperate domains to represent fluid and solids and also verified the domain interface properties.

The problem is the oil is not heated and temperature at inlet and outlet is same.But in actual system the oil gets heated to 120 deg C.

Please can any one share their thoughts on this.

ghorrocks · August 12, 2013, 02:47

Have you put interfaces to connect the solid and fluid domains? Also if you have multiple solid domains?

How have you modelled this? Can you show a picture which shows how you have modelled the heating cable.

shaheen · August 12, 2013, 04:22

Hi Glenn,

I have put interfaces to connect the solid and fluid domains with heat transfer option set to conservative heat flux.

1.JPG

2.JPG

3.JPG

I have attached images of the geometry for your viewing

ghorrocks · August 12, 2013, 06:21

You will also need solid to solid interfaces. Did you put them in too?

JuPa · August 12, 2013, 16:38

You also need to enable conservative interface flux at the interfaces.

shaheen · August 12, 2013, 23:30

I have put solid to solid and fluid to solid interfaces and enabled conservative heat flux between them.

ghorrocks · August 12, 2013, 23:32

Can you post an image of the temperature distribution you do get (from the post processor) and your CCL?

shaheen · August 12, 2013, 23:48

The temperature distribution of oil.
4.jpg

The ccl file output
PipeHeater.txt

ghorrocks · August 13, 2013, 00:00

So it looks like heat is being transferred to the water, just less than you think is correct.

From your CCL I see you have a large pipe flowing water at 1m/s. This is quite a lot of water. And your heat is 100W/m. That is not much heat. My kettle at home is 2.4kW and it takes a couple of minutes to boil 0.5 litres of water. So it does not surprise me that 100W/m does not heat things up much.

Are you sure you have your geometry, flow rate and heat input correct? You really need to do some back of the envelope calculations to check you are in the right direction.

For instance, a 0.5m diamater pipe at 1m/s is 0.196m^3/s or 196 kg/s. At 4.2 kJ/kgK that works out to be 1e-4 K temperature difference averaged across the flow. That looks about what you got with CFX

. So I think the simulation is accurately showing what you asked it to model.

shaheen · August 13, 2013, 00:08

Thanks for the nice explanation

.I will recheck every input and get back asap.

selvam2487 · August 13, 2013, 03:29

[For instance, a 0.5m diamater pipe at 1m/s is 0.196m^3/s or 196 kg/s. At 4.2 kJ/kgK that works out to be 1e-4 K temperature difference averaged across the flow. That looks about what you got with CFX . So I think the simulation is accurately showing what you asked it to model. ]

Dear Glenn,

Can you please tell me which formula you used to calculate the 1e-4 K temperature difference across the flow. I am interested in knowing how you included the power input value of 100 W/m in the calculation.

Regards,
Karthick

ghorrocks · August 13, 2013, 05:51

Your thing looks about 1m long so at 100W/m that is 100W. Then you have a power input, a mass flow rate, a specific heat and that defines the temperature rise.

shaheen · August 14, 2013, 00:01

Yes.In the actual system the pipe is long continous and oil is maintained above 85 deg and at intermediate lengths there is a supporting structure to support the pipe which is also partially insulated.At the support due to ambient conditions the oil temp drops.

The heater cable has output of 100 W/m and in operating condition the heater temperature is 92 deg.

August 11, 2013, 23:47	Heat transfer CHT in ansys CFX	#1
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	Hi, I have modelled a pipe in which oil is flowing at 85 deg C with 1m/s speed.There is a heater cable on the pipe to heat the oil as it flows.The heater provides 100 W/m output.The pipe and heater cable is insulated with additional layer of rockwool. I have created seperate domains to represent fluid and solids and also verified the domain interface properties. The problem is the oil is not heated and temperature at inlet and outlet is same.But in actual system the oil gets heated to 120 deg C. Please can any one share their thoughts on this.

August 13, 2013, 03:29	heat flow calculation	#11
selvam2487 Member Karthik Join Date: Oct 2012 Location: Germany Posts: 53 Rep Power: 13	[For instance, a 0.5m diamater pipe at 1m/s is 0.196m^3/s or 196 kg/s. At 4.2 kJ/kgK that works out to be 1e-4 K temperature difference averaged across the flow. That looks about what you got with CFX . So I think the simulation is accurately showing what you asked it to model. ] Dear Glenn, Can you please tell me which formula you used to calculate the 1e-4 K temperature difference across the flow. I am interested in knowing how you included the power input value of 100 W/m in the calculation. Regards, Karthick Last edited by selvam2487; August 13, 2013 at 03:30. Reason: Mistake

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Question about heat transfer coefficient setting for CFX	Anna Tian	CFX	1	June 16, 2013 06:28
An error has occurred in cfx5solve:	volo87	CFX	5	June 14, 2013 17:44
heat transfer in a box in ANSYS CFX 10	Igor Di Varano	CFX	2	November 24, 2006 18:58
STAR, Fluent, CFX and conj. heat transfer	star-user	Main CFD Forum	8	January 21, 2003 00:07

August 12, 2013, 02:47		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Have you put interfaces to connect the solid and fluid domains? Also if you have multiple solid domains? How have you modelled this? Can you show a picture which shows how you have modelled the heating cable.

August 12, 2013, 04:22		#3
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	Hi Glenn, I have put interfaces to connect the solid and fluid domains with heat transfer option set to conservative heat flux. 1.JPG 2.JPG 3.JPG I have attached images of the geometry for your viewing

August 12, 2013, 06:21		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	You will also need solid to solid interfaces. Did you put them in too?

August 12, 2013, 16:38		#5
JuPa Senior Member Mr CFD Join Date: Jun 2012 Location: Britain Posts: 361 Rep Power: 14	You also need to enable conservative interface flux at the interfaces.

August 12, 2013, 23:30		#6
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	I have put solid to solid and fluid to solid interfaces and enabled conservative heat flux between them.

August 12, 2013, 23:32		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Can you post an image of the temperature distribution you do get (from the post processor) and your CCL?

August 12, 2013, 23:48		#8
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	The temperature distribution of oil. 4.jpg The ccl file output PipeHeater.txt

August 13, 2013, 00:00		#9
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	So it looks like heat is being transferred to the water, just less than you think is correct. From your CCL I see you have a large pipe flowing water at 1m/s. This is quite a lot of water. And your heat is 100W/m. That is not much heat. My kettle at home is 2.4kW and it takes a couple of minutes to boil 0.5 litres of water. So it does not surprise me that 100W/m does not heat things up much. Are you sure you have your geometry, flow rate and heat input correct? You really need to do some back of the envelope calculations to check you are in the right direction. For instance, a 0.5m diamater pipe at 1m/s is 0.196m^3/s or 196 kg/s. At 4.2 kJ/kgK that works out to be 1e-4 K temperature difference averaged across the flow. That looks about what you got with CFX . So I think the simulation is accurately showing what you asked it to model.

August 13, 2013, 00:08		#10
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	Thanks for the nice explanation .I will recheck every input and get back asap.

August 13, 2013, 05:51		#12
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Your thing looks about 1m long so at 100W/m that is 100W. Then you have a power input, a mass flow rate, a specific heat and that defines the temperature rise.

August 14, 2013, 00:01		#13
shaheen New Member Shaheen Join Date: Aug 2013 Posts: 8 Rep Power: 12	Yes.In the actual system the pipe is long continous and oil is maintained above 85 deg and at intermediate lengths there is a supporting structure to support the pipe which is also partially insulated.At the support due to ambient conditions the oil temp drops. The heater cable has output of 100 W/m and in operating condition the heater temperature is 92 deg.