Hello everyone,
I am using AirfoilPrep and i do have a question regarding the 3D stall file. We contracted a 2D CFD enterprise to provide us the cl,cd,cm coefficients for a rotorblade designed for lower wind speeds.
My objective is to compare different 3D correction(Such as Snel.et.al correction) models as the provided data is just 2D.
I found out that there is a huge impact on the output of the cL/cd paramters in change of min.alpha and max.alpha in the sheet 3D Stall.
In other postings it has be said that the values are used to calculate the slope of the CL curve. That makes sense to me but i cannot see why the slope of the curve results in a huge change of c_L values.
Moreover I am interested in "the best" values, to make the output realistic.
Here I show you the impact for various min.alpha and max.alpha
The used values RotorRPM=75RPM, RotorR=5.75, Wind V= 7.5 m/s, r/R=0.3, chords= 0.54737, alpha end=30°
There is a huge difference between the 3D corrected values, for example if we pick 18° AoA.
cL(alpha min 0, alpha max 5)=1.975
cL(alpha min 0, alpha max 10)=1.9122
cL(alpha min 0, alpha max 20)=1.72
cL(alpha min 0, alpha max 50)=4.26
Can any one of you answer me the question for the best fitting min.alpha and max.alpha?Or even better, what is the physical backround of that?
That would be very helpful
Kind regards
Alexander Breitenbach
ab@enbreeze.com
Impact of min aplha max alpha in 3D stall table AirfoilPrep
Moderator: Bonnie.Jonkman
Impact of min aplha max alpha in 3D stall table AirfoilPrep
 Attachments

 AirfoilPrep_v2.02.03.xls
 My AirfoilPrep
 (349 KiB) Downloaded 223 times

 Posts: 5882
 Joined: Thu Nov 03, 2005 4:38 pm
 Location: Boulder, CO
 Contact:
Re: Impact of min aplha max alpha in 3D stall table AirfoilP
Dear Alexander,
The "Min. Alpha" and "Max. Alpha" inputs in the 3DStall worksheet of AirfoilPrep should specify the lower and upper bounds of the linear range of the CL curve, which is used to determine the slope of the CLalpha curve for use in the 3D correction. From a quick look at your data, I'm not sure why your 2D airfoil data does not have a clear linear region, but certainly I would say that the curve is not linear beyond a range of alpha between 5 and 10 degrees. So, setting "Min. Alpha" and "Max. Alpha" to be outside of this range will yield the wrong slope and inaccurate 3D correction.
Best regards,
The "Min. Alpha" and "Max. Alpha" inputs in the 3DStall worksheet of AirfoilPrep should specify the lower and upper bounds of the linear range of the CL curve, which is used to determine the slope of the CLalpha curve for use in the 3D correction. From a quick look at your data, I'm not sure why your 2D airfoil data does not have a clear linear region, but certainly I would say that the curve is not linear beyond a range of alpha between 5 and 10 degrees. So, setting "Min. Alpha" and "Max. Alpha" to be outside of this range will yield the wrong slope and inaccurate 3D correction.
Best regards,
Jason Jonkman, Ph.D.
Senior Engineer  National Wind Technology Center (NWTC)
National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway  Golden, CO 80401
+1 (303) 384 – 7026  Fax: +1 (303) 384 – 6901
nwtc.nrel.gov
Senior Engineer  National Wind Technology Center (NWTC)
National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway  Golden, CO 80401
+1 (303) 384 – 7026  Fax: +1 (303) 384 – 6901
nwtc.nrel.gov
Re: Impact of min aplha max alpha in 3D stall table AirfoilP
Dear Jason,
thanks for your reply.
So, your advice is to choose a "alpha.min" and "alpha.max" in a linear region of the cl vs AoA curve.
The more linear the region, the more accurate the outcome? Is this the input for the DU and SELIG 3D correction?
From what I found in the internet, DU and SELIG correction use the formular that I attached to that post.
Within that formula I dont see any kind of "slope" like a "[cl(2)cl(1)]/[AoA(2)AoA(1)].
I would be very happy to understand the mathematical and physical impact in order to validate and compare the 3D correction with other model of 3D correction.
Anyways, I will give that a try and use a try and use a linear region for alpha.min and aplha.max
P.S:The nonlinearity of the curve might be due to the special design of the rotor blade.
Thanks a lot and best regards
Alexander Breitenbach
ab@enbreeze.com
thanks for your reply.
So, your advice is to choose a "alpha.min" and "alpha.max" in a linear region of the cl vs AoA curve.
The more linear the region, the more accurate the outcome? Is this the input for the DU and SELIG 3D correction?
From what I found in the internet, DU and SELIG correction use the formular that I attached to that post.
Within that formula I dont see any kind of "slope" like a "[cl(2)cl(1)]/[AoA(2)AoA(1)].
I would be very happy to understand the mathematical and physical impact in order to validate and compare the 3D correction with other model of 3D correction.
Anyways, I will give that a try and use a try and use a linear region for alpha.min and aplha.max
P.S:The nonlinearity of the curve might be due to the special design of the rotor blade.
Thanks a lot and best regards
Alexander Breitenbach
ab@enbreeze.com
 Attachments

 DU_SELIG_Correction.PNG (93.19 KiB) Viewed 6258 times

 Posts: 5882
 Joined: Thu Nov 03, 2005 4:38 pm
 Location: Boulder, CO
 Contact:
Re: Impact of min aplha max alpha in 3D stall table AirfoilP
Dear Alexander,
I'm no expert on the AirfoilPrep spreadsheet or the Du and Selig rotational augmentation correction, but my understanding is that the 3DStall worksheet uses the actual 2D CL slope of the airfoil in place of the idealized slope of 2*pi used in your equation. You can view the Visual Basic source code within the AirfoilPrep spreadsheet to see exactly how the inputs are being used by the model.
Best regards,
I'm no expert on the AirfoilPrep spreadsheet or the Du and Selig rotational augmentation correction, but my understanding is that the 3DStall worksheet uses the actual 2D CL slope of the airfoil in place of the idealized slope of 2*pi used in your equation. You can view the Visual Basic source code within the AirfoilPrep spreadsheet to see exactly how the inputs are being used by the model.
Best regards,
Jason Jonkman, Ph.D.
Senior Engineer  National Wind Technology Center (NWTC)
National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway  Golden, CO 80401
+1 (303) 384 – 7026  Fax: +1 (303) 384 – 6901
nwtc.nrel.gov
Senior Engineer  National Wind Technology Center (NWTC)
National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway  Golden, CO 80401
+1 (303) 384 – 7026  Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

 Posts: 3
 Joined: Mon Sep 28, 2015 8:52 pm
 Organization: Missouri University of Science and Technology
 Location: Missouri, USA
Re: Impact of min aplha max alpha in 3D stall table AirfoilPrep
Dear Alexander and All Members,
I know my answer is quite late but for those still interested i will try to explain. In the model by Du and Selig, delta_CL part has Cl,p term. Cl,p =2pi(alphaalpa0) where alpha0 is the angle at which the 2D Cl =0. To find this angle you need the slop of the linear segment of 2D cl against alpha and also interception of this linear segment (alpha0=interception/slop). The max and min alpha should yield reasonable R^2. if you make min and max alphas out of the linear segment then you have wrong slop and interception. In the other side, if you make the min and max alphas so close to each other say [1 to 1] then this range will yield a high R^2 but it is not a good representative for the linear segment. ( I believe AirfoilProp will give error if the range is small). just check for reasonable R^2 say greater than 0.97 with rang that cover a wide range of data from the linear segment.
Now it comes to my question if anyone can help please
1 I wonder what is the best value of Alpha End ( the end of the rang of the full correction). is there any value should be chosen based on the stall angle?
2 If I use XFOIL to generate the 2D data, should the generated data include the matching point or the stall angle for Viterna extrapolation?
I will really appreciate any suggestion
Best Regards
Aziz
I know my answer is quite late but for those still interested i will try to explain. In the model by Du and Selig, delta_CL part has Cl,p term. Cl,p =2pi(alphaalpa0) where alpha0 is the angle at which the 2D Cl =0. To find this angle you need the slop of the linear segment of 2D cl against alpha and also interception of this linear segment (alpha0=interception/slop). The max and min alpha should yield reasonable R^2. if you make min and max alphas out of the linear segment then you have wrong slop and interception. In the other side, if you make the min and max alphas so close to each other say [1 to 1] then this range will yield a high R^2 but it is not a good representative for the linear segment. ( I believe AirfoilProp will give error if the range is small). just check for reasonable R^2 say greater than 0.97 with rang that cover a wide range of data from the linear segment.
Now it comes to my question if anyone can help please
1 I wonder what is the best value of Alpha End ( the end of the rang of the full correction). is there any value should be chosen based on the stall angle?
2 If I use XFOIL to generate the 2D data, should the generated data include the matching point or the stall angle for Viterna extrapolation?
I will really appreciate any suggestion
Best Regards
Aziz
Who is online
Users browsing this forum: No registered users and 1 guest