Flow across interface / Negative Moment Monitor Values

aptahaney · August 11, 2015, 20:58

Hi everyone,
I am doing a 2D CFD analysis of a VAWT and thought I was in my last few days of analysis but have ran into a problem.

In order to do the Fluent analysis, I practiced the technique using the "Introduction to Fluent Workshop 7: Using Moving Reference Frames and Sliding Meshes (Version 15)". These were user friendly and do-able and I got the right answers, so I used the sliding mesh method as the basis of my remaining analsys.

However, when I completed my analysis, after several hours of computations, the average Monitors value I got was a negative value (I let the model run for 3 full rotations and took the last 360 degrees to calculate the average Cm value). This would suggest a negative torque value which is clearly wrong.

I do not know the exact reason for this, but I suspect it has something to do with the speed of rotation of the internal rotor, its flow across the interface and its effect on the air within the rotor section (but I can not be sure)....The reason I think this is that when I ran the Tutorial Workshop model, I had excellent flow across the interface, as shown in the attached. However, when I used my own model, there was not much flow across the interface at high rpm.

The wind speed in my problem is just 9m/s, my turbine diameter is 1.03m, and the tip speed ratio I have chosen initially is 1.5, resulting in a rotation at 250rpm by my calculations. This can be seen in the attached "Output at 250rpm".

When I ran my same model at 20rpm, the flow across the interface seems to be slightly improved. (see attached)

So, would the speed of rotation be negatively affecting my Moment Monitor results as I have described? If so, how can I fix it so that I am getting proper results. Is it just a matter of running the model for a longer period (say 7 full rotations) before taking the Monitor values so that it will be fully developed model?

If not, is there any other explanation for my poor results? Or what can I do to rectify them? How do people get around the problem of a low wind speed and a high rotational speed?

I am under quite a bit of pressure time-wise as my project is due in the next few weeks and I need these results before I can proceed any further, so I'd really appreciate any help you can offer.

thanks,

Aidan

Output at 250 RPM.jpg

Output at 20 RPM.jpg

Tutorial Output.jpg

aptahaney · August 13, 2015, 21:40

Hi everyone,
Just wondering if anyone had a chance to consider my query above? I'm not so sure now if it's the flow acriss the interface that is causing my problem. I attach three relevant images from my more basic model, that may inform those in the know.
Any help would be welcome and appreciated.
Cheers,
Aidan

EdFL · October 6, 2017, 08:06

Hi aptahaney,

I'm having almost exact the same problems you had at the time. Have you been able to solve the problem? What should be done when negative torque values are monitored? I've tried many ways but couldn't overcome (refining mesh, domain, reference values, domain etc.) I'd really appreciate it if you could share your comments.

Regards,
Ed

aptahaney · October 6, 2017, 15:26

Unfortunately I don't recall at this stage. It was something I did years ago and haven't looked at since.

I do remember that I had an issue with the interface that was giving me incorrect figures. I think it was something to do with how I had not assigned the second part of the interface correctly as an interface boundary condition... I'm just going on memory here but you may want to look into that as a possibility

There were tutorials online, (Fluent website and YouTube) that I followed and they worked.... Eventually.

August 11, 2015, 20:58	Flow across interface / Negative Moment Monitor Values	#1
aptahaney New Member Aidan Join Date: Jun 2015 Posts: 17 Rep Power: 10	Hi everyone, I am doing a 2D CFD analysis of a VAWT and thought I was in my last few days of analysis but have ran into a problem. In order to do the Fluent analysis, I practiced the technique using the "Introduction to Fluent Workshop 7: Using Moving Reference Frames and Sliding Meshes (Version 15)". These were user friendly and do-able and I got the right answers, so I used the sliding mesh method as the basis of my remaining analsys. However, when I completed my analysis, after several hours of computations, the average Monitors value I got was a negative value (I let the model run for 3 full rotations and took the last 360 degrees to calculate the average Cm value). This would suggest a negative torque value which is clearly wrong. I do not know the exact reason for this, but I suspect it has something to do with the speed of rotation of the internal rotor, its flow across the interface and its effect on the air within the rotor section (but I can not be sure)....The reason I think this is that when I ran the Tutorial Workshop model, I had excellent flow across the interface, as shown in the attached. However, when I used my own model, there was not much flow across the interface at high rpm. The wind speed in my problem is just 9m/s, my turbine diameter is 1.03m, and the tip speed ratio I have chosen initially is 1.5, resulting in a rotation at 250rpm by my calculations. This can be seen in the attached "Output at 250rpm". When I ran my same model at 20rpm, the flow across the interface seems to be slightly improved. (see attached) So, would the speed of rotation be negatively affecting my Moment Monitor results as I have described? If so, how can I fix it so that I am getting proper results. Is it just a matter of running the model for a longer period (say 7 full rotations) before taking the Monitor values so that it will be fully developed model? If not, is there any other explanation for my poor results? Or what can I do to rectify them? How do people get around the problem of a low wind speed and a high rotational speed? I am under quite a bit of pressure time-wise as my project is due in the next few weeks and I need these results before I can proceed any further, so I'd really appreciate any help you can offer. thanks, Aidan Output at 250 RPM.jpg Output at 20 RPM.jpg Tutorial Output.jpg

October 6, 2017, 08:06	Has the problem been solved since then?	#3
EdFL New Member Edward Vismann Join Date: Oct 2017 Posts: 1 Rep Power: 0	Hi aptahaney, I'm having almost exact the same problems you had at the time. Have you been able to solve the problem? What should be done when negative torque values are monitored? I've tried many ways but couldn't overcome (refining mesh, domain, reference values, domain etc.) I'd really appreciate it if you could share your comments. Regards, Ed

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
sliding mesh problem in CFX	Saima	CFX	46	September 11, 2021 07:38
Modeling the fan with a mass flow interface	saisanthoshm88	CFX	6	August 1, 2014 06:11
Modelling valves/orifices using interface models	DarrenC	CFX	5	June 3, 2014 06:14
Radiation interface	hinca	CFX	15	January 26, 2014 17:11
RPM in Wind Turbine	Pankaj	CFX	9	November 23, 2009 04:05

October 6, 2017, 15:26		#4
aptahaney New Member Aidan Join Date: Jun 2015 Posts: 17 Rep Power: 10	Unfortunately I don't recall at this stage. It was something I did years ago and haven't looked at since. I do remember that I had an issue with the interface that was giving me incorrect figures. I think it was something to do with how I had not assigned the second part of the interface correctly as an interface boundary condition... I'm just going on memory here but you may want to look into that as a possibility There were tutorials online, (Fluent website and YouTube) that I followed and they worked.... Eventually.