How to calculate turbine RPM using Ansys CFX
ฝัง
- เผยแพร่เมื่อ 14 ก.ค. 2024
- In this video you will learn:
- How to create a rotating domain
- Freeze and unfreeze fluid bodies
- Use parameter set to determine the net torque on a turbine
Get the files: payhip.com/b/Hef0T
If you have any questions, queries or recommendations for future videos, please leave them in the comments section below.
This video was made in response to CFD-Online queries about the error message "Angular velocity can depend only on time for transient simulation" which can be found here: www.cfd-online.com/Forums/cfx...
Useful Books and Resources
Fundamentals of Gas Turbines - W. Bathie
amzn.to/2RaX4S8
Gas Turbine Theory, Sixth Edition - HIH Saravanamuttoo, GFC Rogers, H Cohen, PV Straznicky, AC Nix
amzn.to/35l2vpZ
Firstly, Thanks for the great tutorial.
for those who get the "The solver failed with a non-zero exit code of : 2" error. It may be solution.
(in workbench screen ) right-click to setup and select edit- (Cfx-Pre will open)...in this windows, on left you see "interfaces"..right click on it select edit. And there in the bottom, you will see "frame change /mixing model" selected as none..change it to "frozen rotor" and hit the apply button
thanks for your help
Thank you so much friend
Helpful tip: Follow other comments about angular velocity as OmegaZ, and selecting frozen turbine for frame change/mixing model. After 16 hours of painfully trouble shooting, make sure your mesh sizing element size is large enough that it doesn't exceed the ANSYS Student Version! I tried 0.01, 0.25, and 0.50 inches for element size for meshing, and I kept getting the solver to fail with exit code 2 as seen in other comments. I changed mesh element size to 1.0 inch, and the solver ran successfully (meaning I didn't exceed the ansys student limits for CFD calcs. So yeah, see how many nodes/elements are in your project, and make sure it doesn't exceed the ANSYS Student Version limits! Haven't and probably won't check mesh quality, but at least it ran.
Also thank you Climate CFD, you have one of the only videos on this specific problem of trying to find turbine RPM.
Great advice. Chopped the omega z out in editing and cannot seem to add it back in.
Thank you Sir for putting such great effort into this video. I and my team gained a huge help with this video. Grateful!
That's great to hear.
Thanks.
If you're having the issue of getting the same torque values for the output go back into the rotating domain in setup and instead of putting the angular velocity as a number, set it to 'OmegaZ' also for each section under 'interfaces' in setup, select frozen turbine for Frame change/mixing model, and None for pitch angle.
Yes, I wonder if this was inadvertently edited out?
Apologies for those who were confused!
@stormfront You saved my life!! Thanks a ton!!
Very good logic Sir , Thanks a lot for such a good info
Thanks. I appreciate that.
brilliant idea
Thanks, credit really goes to the OP from CFD-Online.com
Hello there and thank you for this video, you literally saved my life! I would also like to ask a question regarding turbines. I have a wind turbine geometry which has a 90m radius. It's quite big and when I draw a fluid domain around it, I have some problems because I'm supposed to have multiple wind turbines in my analysis(wind farm).
So I decided to scale it down a bit and I decided to follow the rules of similitude which means the Reynolds numbers should be equal for both scaled and original cases for them to be dynamically similar. So 11m/s is the wind speed for a 90m blade and therefore the wind speed should be 22m/s for a 45m blade.
Here is my question. If I use this scaled wind turbine in my analysis, should I divide the velocity values I've found by 2 to find what the velocity would be if it was 90m blade?
Another question is that I know the RPM of my wind turbine for the 90m blade which is roughly 10 and if I scale it down to 45m, do I have to run this analysis of yours to find the RPM or can I use an analytical equation to find it manually?
Hey did u find answer of your question?
great! many thanks.
You're very welcome. If there are other topics that would be beneficial, just leave a comment and this can be uploaded.
@@ClimateCFD thank you again. If you would have an example how to calculate the turbine RPM via DFBI modeling, I mean the air flow impact on the turbine resulting in the rotation... But again, your approach to run some pre-set RPM's is really fantastic.
@@AR-yj4ft thank you. This was recommended by Glen Horrocks on a cfd-online.com post as a means of overcoming numerical instability caused by making angular velocity a function of torque in the CFX solver.
Thanks for the video. It was very informative. I have a doubt though. If you have a radial flow or centrifugal turbine with a nozzle inlet, does this method still work?? Also how good are these results compared to calculations based on momentum transfer by the fluid considering all loads and losses on the turbine shaft??
@Climate CFD, Several years down the line this tutorial is very helpful. Im doing a simulation to determine the rotational speed of a Wells turbine, the visualization results are perfect , however when i compute the torque using the set input and output parameters, the torque values are the same. what might be the problem?
There was a warning in CFX which says frame change option 'None' is only valid if the domains on each side of the interface rotate with the same angular velocity. Shouldn't we change it because one domain is stationary and the other domain is rotating?
Hello thanks for these information
I have question
I am interested in ansys and I am looking to simulate fire in tunnel what should I use Fluent or CFX version
I hope you can answer my question
Hi
1-At the middle of turbine, shaft should be placed but you put it empty.
2- the rotation speed is depnd on the flow speed, turbine weight, bearing friction and shaft tourqe, on based what did you put "4" for it?
The iteration process will only work if you set the expression omegaz as the rotational velocity of the rotating domain! otherwise great tutorial!
Hello sir , i am working on bulb turbine with similar condition except fluid as water , if we use 0 rad/s as input to find torque would'nt it show initial torque ? , What can we do to find torque at which the turbine will spin (from parameter sweep) at so that power can be calculated. Thankyou
I think you miss to put "OmegaZ" variable expression to the angular velocity in the rotating motion domain.
How to do it?
whenever i select the torque expression i created as the output parameter it never takes the value as the design point in the parameters section. How do i solve that?
Hello! this is a great video thank you very much! I am working on finding rpm for other blade designs and somehow, none of them seem to be rotating except when i try this video on your exact design of blades. Could this problem be due to the mass flow rate?
Hi Tala,
Firstly thanks. Glad it helped.
Shouldn’t be to do with mass flow. It’s likely to be a result of the rotating domain setting omega.
thank you sir for simple explanation ,but can we make it as same in Fluent
Should be able to. The two software do essentially the same thing, just with different buttons.
Great one! Btw since the torque is zero (for a mass flow rate), how can we get the power developed by the turbine?
When torque is 0 power is also 0. The way to calculate power is through the frame motion fluent method. Power is given by P= RotVelocity * torque
Amazing video sir!
I´m a little confused about the results, If you input a positive rotation and as output obtains a negative torque, physically, it means that the flow is trying to force the blade surface in the opposite direction of the rotation right?
CFX doesn't know the physics behind what we´ve set up, it's going to solve the math if possible, but it's interesting and very important to understand what's going on...
At the slower rotation speed, the air is forcing rotation, so this comes out as -ve.
In Analysis 'Flow Analysis 1' - Domain Interface 'Default Fluid Fluid Interface': The frame change option 'None' is only valid if the domains on each side of the interface are physically rotating with the same angular velocity. How to solve it?
Hi i did as you have shown, but after i put rad/sec the value does not appear on top after the OmegaZ. Also how to put cel for "rpm" and degrees(angle)? Thanks. i am searching for cel expression file.
Not sure about the first part.
Convert from rad/sec using the 2 *pi conversion.
Hello, on this case turbine weight is ignored or it is somehow determined by material? Because 0 torque means that "air" rotates on that speed in that geometry not turbine, as we know, to rotate certain weight turbine there should be certain value of torque which is enough to spin that turbine.
Hi, thanks for getting in touch. You are correct, this assumes turbine weight is ignored.
Thank you for for such an excellent video. What changes do I need to do for impeller-- flow vane streightener- nozzle simulation of jet pump unit? Need to find jet force, velocity with rpm input.
Hi Andrey,
Glad it was useful for you.
To clarify, is this the jet force on the impeller?
@@ClimateCFD Hi, build up pressure between impeller and nozzle and jet force out of nozzle. I can share exported files from project.
I want to find rpm of savonius wind turbine on given wind speed . I also have values of mass and moment of inertia .How can I do that?
hi there great video, just want to know will it be possible to find the rpm using flow of air in m/s instead of mass
flow rate?
Yes, no reason why not.
Hi everyone, I have a question that consume me...As far I understand, the RPM of a impeller, depends on the load conected to the axle (for example: in steam turbine-generator, if the power used in the generator decreased, the turbine velocity increase) so, if you are calculating the velocity when the torque is zero,¿ it means, that you are calculating the "freewheel rotation"?
Hi Luis,
I think it is best to approximate the power produced using the torque produced by the stationary turbine, and the rotational velocity when the torque is zero.
Hi please can you do step by step exhaust fan simulation and Fan rotation animation for air
We have only rpm ,we don't know how in let air velocity and out let velocity
Hello! I'm trying to perform simulation on an archimedes spiral wind turbine using ansys fluent. My question is how can I determine the turbine's rpm without giving it a pre-defined value? In Fluent, if you want the rotating domain to rotate you must give it a pre-defined value. I'm not sure how to determine its rpm with only the inlet wind condition as an input.
Is the process similar to what you did in this video? Thanks in advance!
It should be a similar process.
My understanding is that CFX (and likely Fluent) cannot have RPM as a variable dependent on air speed in these simulations as it will cause a numerical instability, hence this slightly imperfect approach.
@@ClimateCFD I see, thank you that kind of cleared my question regarding the matter :)
@@erwinphillipcabanban5582 Did you got it ? I have the same problem :(
@@exscootable Hello. I didnt proceed with this matter anymore. I just based my analysis on a fixed tip speed ratio and pre-define the rpm of the turbine.
Sorry I couldnt help :(
Hey, I tried your method but faced with following error:
Message: Error number 1 found in subroutine CHECK_INIT. Interface "Default Fluid Fluid Interface" connects a rotating and stationary domain, but has no frame change model.
Should i change frame change from "NONE" to "FROZEN ROTOR"?
Frozen rotor stator
@@ClimateCFD Thanks for the reply, I've tried it and got the results but I'm facing one problem which I posted in CFD forum www.cfd-online.com/Forums/cfx/231722-marine-propeller-rotating-mesh.html
Can you help me with this😭
@@rakeshchandra1746 I don't currently have the time but can advise to revisit the mesh and BCs.
Use things like growth rate from the interface to ensure economic use of computational resource.
@@ClimateCFD Thank-you I'll try it
I have followed the same procedure and created the expression omega-y in my case. The results are fine the expression for torque generates the value as expected
But when i access parameter set and solve for new values of omega y i always get the same answer. Also i don't know how the expression omega y is linked to any parameter given it is only a number with unit.
Let me take a look at this and I'll get back to you.
I had to put the OmegaZ expression value into the rotating domain as the angular velocity parameter.
good to see you with same problem.
😊
Hello Climate CFD! Great video.
Can u kindly tell me what - (minus) sign represents here in the omegaZ values?
Just to ensure the turbine rotates the correct way.
Just to ensure the turbine rotates the correct way.
@@ClimateCFD thanks for the reply!
I've another query.
In the inlet you select mass flow rate , but if i want to select air velocity that possible? Should i select cartesian velocity and apply the velocity in the Z direction?
@@codexn5716 you can either select normal to the inlet of the vector for flow direction….which I think is a in this case.
hello! i just followed this video and tried by changing the values results are excellent. and I've a doubt that "how to find out the rpm of wind turbine by giving velocity as input? means input: velocity(of air) and output: turbine fan rpm. could you help me with this?
Hi, thanks for getting in touch. For the wind turbine, why is the method different?
@@ClimateCFD
Why because initially we don't know the rpm of the turbine rotor. We know only the wind speed, and main thing in wind turbine is wind makes to rotate the turbine... and based on that wind speed we have to calculate the rpm.
@@phanibhushanm7499 Think about the relationship between torque and angular acceleration. The constant of proportionality between the two is the moment of interia.
So when the angular velocity of the turbine is less than it can rotate due to the wind speed, there will be a positive torque, which will cause it to accelerate towards the angular veloity where torque = 0.
@@phanibhushanm7499 Have u done that? finding the rpm of wind turbine by giving velocity as input?
Hi I have a question, when I try to generate solution I get an error code saying "the solver failed with a non-zero exit code of : 2". Do you know why this is happening and how I can fix this?
Check the .out file. Could be a floating point exception (improve the quality of your mesh) or a license issue.
Let me know if that works.
Climate CFD hi thanks for replying turns out I forgot to enable frozen rotor
Hello, thank you for the tutorial. It is very useful.
I have a questions, I already follow all of the steps. But, when I tried to vary the rotational speed, the value of the torque did not change. Do you have any clues about this? Thank you so much.
Hello,
Thank you. Glad it helped.
It’s hard to say without seeing the case. Have you tried clearing the results and rerunning?
@@ClimateCFD yes, i cleared it but the torque still the same. is it possible to send the image of the case? thank you very much.
@@naufalriyandi7228 sure, climatecfd@gmail
Com
dear hope u are fine. Actually i am doing analysis on vortex turbine blades but the problem came is that i want the fluid to rotate the turbine nor i do give value to turbine balde to rotate for 2d case there are videos present on how to rotate the turbine but mine one is 3d, so i am in serious trouble please help. thanks in advance.
If you need help can you send any relevant information to climatecfd@gmail.com?
Hi, i have to do something like that. Did you solve it?. I really need help
@@santiagoefrainburgoszambra9542 Yes, I think it was a mesh issue.
@@ClimateCFD can you explain me more? I am going to send you a message
@@santiagoefrainburgoszambra9542 I believe the mesh just had poor quality cells, unsure if this is the same for your case.
Send any relevant information to climatecfd@gmail.com and I'll take a quick look.
I followed this video precisely. I'm getting the error "The solver failed with a non-zero exit code of : 2." Looking in further, the error I get says "Error number 1 found in subroutine CHECK_INIT. Interface "Default Fluid Fluid Interface" connects a rotating and stationary domain, but has no frame change model." Any tips?
Frame change model to frozen rotor stator
@@marufmorshed3793 like he said, change to frozen rotor stator model and it fixes the issue
@@marufmorshed3793 OIt should be under frame-change model in interface settings.
@@marufmorshed3793 Should be under the interface and then the tabs at the top.
for a more clear guide .. (in workbench screen ) right-click to setup and select edit- (Cfx-Pre will open)...in this windows, on left you see "interfaces"..right click on it select edit. And there in the bottom, you will see "frame change /mixing model" selected as none..change it to "frozen rotor" and hit the apply button
@Climate CFD sir I have performed the simulation and set the frame change or the mixing model option: Frozen rotor and pitch change option to: None. Now when I change the rotational speed in the workbench input parameter no matter what the torque always seem to come out to be constant. Any suggestions as to how to resolve it
Hi yempada i am facing the same issue. have you resolved it?
@@thejussuresh actually I am using a different software STAR CCM+ in which I found the solution
@@yempadayugesh8195 ok Yempada. Thank you for the reply.
did u find out?@@thejussuresh
@@user-xq9hd8ch9u no
sir how you have determined the mass flow rate and angular velocity please ??
What do ou mean by determined? These were just selected for demonstration.
@@ClimateCFD @Climate CFD sir actually i am working on a wind turbine and want to know what is that angular velocity you have taken, because if know the angular velocity of fan then you can find rpm by simple formula itself so can you please explain me what is that angular velocity you have taken
@@jiraiyasensei5436 It should be in the video.
goood ... I have a question ....how can I find total torque of mixer ???
How do you define total torque? Pressure + friction components?
Hello sir, I'm one of your TH-cam followers. I'm currently facing an issue in running a simulation of a Savonius wind turbine. I'm working on my thesis which needs to be completed, and I'm struggling to meet the demands of my thesis supervisor who requires obtaining RPM directly from Ansys simulation. The variables I'm using include wind speed variations from 1m/s to 7m/s. If you would be willing to create a tutorial video, I would be extremely grateful to you.
Drop me an email at climatecfd@gmail.com
@@ClimateCFD OK sir, thank you. I will send it soon
I keep getting an error saying that
"Interface "Default Fluid Fluid Interface" connects a rotating and stationary domain, but has no frame change model."
Have you ran into this before??
Hi Ben,
In the interface settings, at the bottom there should be an option for frozen rotor stator. Select pitch change to be none. Let me know if that works.
@@ClimateCFD Thank you, yes that worked!!
Now the only issue is that when I input various OmegaZ values, the TorqueZ value doesn't change... does that mean my turbine isn't spinning?
@@benjaminbreeden1181 Hi Benjamin, in the video there is a missed step. This step consist in set the input variable "omega" as the rotating velocity of the impeller. Then you will have solve the problem ;)
@@luisarancibia1869 Thanks man, after some painstaking labour we got it figured out!
hi first of all thank you @Climate CFD for the amazing tutorial and the comment tht let an end to my struggle , now please @luis arancibia would u explain the missing step i still don't get it m new to ansys
hi everyone, i'm working on a hydrokinetic turbine, i am able to calculate the rpm of the turbine rotor at zero torque through the parameter sweep.Now my issue is how do i determine the power produced by the turbine?
If power is torque * angular velocity, just take the torque from omega = 0rad/s and use the omega where T=0.
@@ClimateCFD Sir would'nt it show the initial torque if we find torque at 0 rad/s , what can we do to find torque at the rotation turbine will spin at ?
"I got a code "Update failed for the Solution component in Fluid Flow (CFX). The solver with a non-zero exit code of: 2. Do you have a suggestion?
Check your license
If that fails, clear the solution and retry.
How long is the total running time?
@@congatinh4261 I don’t remember, but I assume only a few minutes with that mesh resolution.
Everything works great, but when I (Update All Design Points) it says failed :(
See other comments. Really need more detail on the error.
@@ClimateCFD Can I send it to you, I'm really wrecked from trying :(
Hi, I followed every step in this video carefully, and finally changes the Frame change to Frozen rotor, then I got a result that the turbine seems didn't rotate. What should I do?
"The turbine seems didn't rotate" means no wake area appears behind my rotor. That is abnormal.
@@ruoyili1884 The rotating domain has an angular velocity about the correct axis?
@@ClimateCFD Yes it has
@@ruoyili1884 I'm not sure what to suggest, if the setting is applied, the domain should be rotating. Try visualising 'velocity in stationary frame' in CFD Post.
@@ClimateCFD Ok! Thanks a lot!
Hi sir. Nice work. I've tried this tutorial with 3 blades HWAT, and got an error "exit code of :2". Then, I tried to fix this error, in "setup". I changed the domain type of rotating part to the "solid domain", then the simulation can run. But....... The result's picture is not the same with your result. Next..... I tried to fix again to change the domain type of rotating part to "fluid domain", with the domain motion option is "Stationary". The result's picture is same with yours, but i'm not sure this is a right solution :). Do you have any suggestion sir? Thank you sir.
It really depends on what the error was. What is the exact error in the .out file?
I think, the error is come from the setup section sir. I already fix it, by keep the "rotating domain" setup is stationary, but the "default fluid fluid interface side 1" boundary setup is rotates. Now i face new problem, the torque parameter does not change like yours. I've read all comments bellow, try to fix it, but still not works.
By the way, if i follow this tutorial. The error happens in solution section. The error says
1. ERROR has occured in subroutine ErrAction. Error number 1 found in subroutine CHECK_INIT. Interface "Default Fluid Fluid Interface" connects a rotating and stationary domain, but has no frame change model.
2. 1 fatal error found in the problen setup.
3. Stopped in routine CHECK_INIT
These all about setup section. I followed this step many times, but still come up with same error. Thank you for your response sir.
@@alfaageng me too facing same issue, do u find any solution
I'm getting an error in Solution part.. it displays update failed .. i followed the same procedure but gave normal speed instead of mass flow rate.. please help me in it
Hi, what's the error/warning text?
@@ClimateCFD it's showing update failed for solution component in fluid flow (cfx),the solver failed with non zero exit code :2
@@rahulbharathwaj6006 check the mesh quality. Most of the time it's that.
@@ClimateCFD one more thing, why are we setting an input parameter as angular velocity and output parameter as torque.! I had set my normal speed as 1 m/s and i'm getting the same values , for example for 5 rad s^-1 i'm getting -0.001582, for -100 also same value.! Please help me in it! I got the solution part done.! thanks
@@rahulbharathwaj6006 this gives the speed at which the aerodynamic torque load is zero...so is the rotational speed at that mass flow rate, assuming no friction.
The torque results are the same How can I fix it?
Thanks
did u find out
The same as what, sorry?
@@ClimateCFD every value of torque regardless of the RPM turns out to be same after we give calculate. Tried for various geomtry, flow conditions and all. Did exactly same as what you did, stll all the torque values during the parametric study is the same.
Is it possible to find the RPM of turbine ?
From the above method, yes.
do you have active email or any social media that u active in it?
Not anymore, no.
@@ClimateCFD so how i can contact you?
@@yeddidyonataneducation comments on here.
What can I help with?
I want to know. The torque and rpm of propeller inside the cilinder, like simulate the cord nozzle propeller
kort nozzle propeller
"The silver failed with a non-zero exist code of : 2 " Please help me out from this
Can you provide more detail from the .out file?
Did this get resolved? Having the same error code...
@@owenmansfield1439 what is the specific content in the. Out file?
Hello climate cfd. Could you please help me in doing cfd analysis of wind turbine for my project..i have my model wind turbine and i just need some guidance. If it possible for you please provide me your email adress and then can discuss with you other details..
Hi, it's Climatecfd@gmail.com
Hai cf climate thank you for the reply. I sent you the mail regarding my project. You didn't replied me there. Please kindly check the mail and let me know what are the possible solutions for my project.