BModes : Input parameters about tower support subsystem

Provide feedback, request enhancements, and get help with wind-turbine computer-aided engineering tools.

Moderators: Bonnie.Jonkman, Jason.Jonkman

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Ngoc Ha Tran,

PreComp makes use of thin-walled assumptions. For a solid cross section, you'll need to derive the cross-sectional properties using a different tool/method. If the solid is composed of an isotropic material, the calculation may be straightforward, depending on the shape of the cross section.

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

Ha.Tran
Posts: 11
Joined: Sun May 29, 2016 5:42 pm
Organization: The university of Auckland
Location: New Zealand

Re: BModes : Input parameters about tower support subsystem

Dear Jason,

I really appreciate your response. It helps me so much to simulate my model accurately.

Now, I am having a trouble in determining flapwise and edgewise stiffness of cross-sections of a tidal turbine blade.

I read in BModes 's guide. They said that flapwise stiffness is the section flap bending stiffness about the principal elastic axis (Ye) and edgewise stiffness is the section lag bending stiffness about Xe elastic axis.

As I read in some books, they said that the elastic axis is the locus of shear centers of all cross-sections of blades. It means that the elastic axis is determined for the whole beam, not for any certain cross-section.

However, in BModes 's guide, it seems that the elastic axis is on the cross-section of blade. It makes me confused.

So, can you show me the method to determine the principal elastic axis on each cross-section of blades? I have found the shear center of each cross-section already.

Thank you very much.

I am looking forward to hearing from you.

Kind regards,

Ngoc Ha Tran.

Ha.Tran
Posts: 11
Joined: Sun May 29, 2016 5:42 pm
Organization: The university of Auckland
Location: New Zealand

Re: BModes : Input parameters about tower support subsystem

Dear Jason,

One more question is about blade-tip mass. In my mind, blade-tip mass is mentioned as a tip brake is used on the blade. Am I thinking correctly? Can you explain me the meaning of this parameter and when should I use this parameter? Thank you very much.

I am using Bmodes for the purpose of determining the coefficients of blade shape modes in FAST. However, I realize that natural frequencies I got from Bmodes are too large, even up to 400Hz. Does it seem to have some problems here? I checked the parameters in the auxiliary input file and maybe I used the principal inertia co-ordinate system to obtain flapwise and edgewise stiffness instead of the principal elastic axes. As I mentioned in the previous question, I searched elastic axis and found that elastic axis is the locus of shear centers and is used for the whole beam, not for a cross-section. Can you show me the definition and the method to determine the principal elastic axis location for each cross-section of blade?

The third problem is that as I account for blade-tip mass in BModes input file, the natural frequencies reduced from 400Hz to approximately 20 Hz. It means that blade-tip mass plays an important role to natural frequencies, doesn't it?

These problems make me confused.

I would be really grateful if you can help me out.

Thank you very much.

Have a nice week.

Kind regards,

Ngoc Ha Tran.

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Ngoc Ha Tran,

See my post dated Jan 25, 2013 in the following forum topic for my definitions of the various centers: viewtopic.php?f=4&t=775&p=3076. By referring to the elastic axis as the "locus of elastic centers", this means that a line (or curvilinear line) is passed from the root to the tip through the elastic center of each cross section. Note that some authors refer to the "elastic center" as the "tension center" and other authors refer to the "elastic center" as the "shear center".

The principal axes of bending are defined as the orthogonal axes in a cross section whereby the cross bending stiffness terms are zero. The origin of the principal axes of bending is the tension center (the neutral axis).

A tip brake is one example where there is a large lumped mass near the tip of the blade. In most blades, there is no large lumped mass near the tip, so, the tip-mass inputs in BModes or FAST can usually be set to zero.

I know nothing about the blade you analyzing, but I would be surprised if a tip mass was so large that it reduced a blade natural frequency from 400 to 20 Hz.

I hope that helps.

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

Ha.Tran
Posts: 11
Joined: Sun May 29, 2016 5:42 pm
Organization: The university of Auckland
Location: New Zealand

Re: BModes : Input parameters about tower support subsystem

Dear Jason,

I am really sorry if I make some inconvenience for you, but I am still confused with the coordinate system to determine structural twist angle, flapwise stiffness and edgewise stiffness in BModes.

These following information I read in BModes' s guide:

- Str_tw: for a blade, it is the angle the chordwise principal elastic axis would make with the blade reference plane.

- Flp_Stff: for a blade, it is the second flap bending stiffness about the Ye elastic axis.

- Edge_stff: for a blade, it is the second lad bending stiffness about the Xe elastic axis.

On the page 12 of BModes' s guide: Point E is the section shear center and is the origin for the principal elastic axes Xe-Ye.

As I understand, in order to determine Str_tw, flp_stff and edge_stff, I need to determine shear center and the principal elastic axes Xe-Ye for each cross-section. Am I thinking correctly?

If the answer is yes, can you show me how to determine the principal elastic axes for each cross-section? I am really grateful for this help.

I attach the figure in BModes' s guide for the coordinate systems in BModes.

Thank you very much.

I am looking forward to hearing from you soon.

Kind regards,

Ngoc Ha Tran.
Attachments
axes system at a typical blade section.png (41.62 KiB) Viewed 2887 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Ngoc Ha Tran,

The BModes documentation is a bit misleading because it defines the origin of the principal elastic axes of bending as the shear center. However, this should be the tension center (the neutral axis). In general, the shear center may be at a different location in the cross section. However, BModes contains simplifications relative to the most general 6x6 beam theory, and in essence, BModes assumes that the shear center and tension center are coincident (while the tension center is specified separately in BModes, the input is ignored).

I'm not an expert on sectional analysis, but there are a number sectional-analysis tools e.g. VABS, BECAS, NuMAD/BPE, PreComp, etc. that can be used to calculate the sectional properties (6x6 or reduced mass and stiffness matrices, including the various centers and orientations) for you.

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

Ha.Tran
Posts: 11
Joined: Sun May 29, 2016 5:42 pm
Organization: The university of Auckland
Location: New Zealand

Re: BModes : Input parameters about tower support subsystem

Dear Jason,

Thank you very much for your quick and useful response.

In this case, I will make tension center coincide with mass center and the principal elastic axes of bending coincide with the principal inertia axes.

I hope it will be right.

One more problem is that as I run the example "Test01_nonunif_blade_bmi" in Bmodes, it appears an error.

I attach the image of error and the file of BModes I extracted in C driver.

I hope that you can help me out.

Once again, thank you very much for your help.

Kind regards,

Ngoc Ha Tran.
Attachments
image 3.png (142.77 KiB) Viewed 2874 times
image 2.png (116.81 KiB) Viewed 2874 times
image 1.png (75.96 KiB) Viewed 2874 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Ngoc Ha Tran,

In this case, I will make tension center coincide with mass center and the principal elastic axes of bending coincide with the principal inertia axes.

I hope it will be right.

While this may be a suitable approximation for the blade you are modeling, you should be aware that this assumption is not valid in general. For a general composite cross section, the tension center differs from the mass center and the orientation of the principal axes of bending differs from the orientation of the principal axes of inertia.

In your simulation, BModes cannot find the beam section properties input file because you are not running BModes from the directory containing the primary input file. BModes will run if you do this:

C:\Bmodes_v1.03.01>cd CertTest

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

Ha.Tran
Posts: 11
Joined: Sun May 29, 2016 5:42 pm
Organization: The university of Auckland
Location: New Zealand

Re: BModes : Input parameters about tower support subsystem

Dear Jason,

I run Bmodes according to your guide and it works well.

I also tried some software to calculate blade cross-section properties. However, the blades I am using are solid. They don't have any shear webs. The results I got didn't include all parameters required in BModes.

So, can you suggest me some other trial/free software to calculate sufficient parameters of blade solid cross-section, including tension cente, flapwise stiffness and edgewise stiffness?

Thank you very much.

Kind regards,

Ngoc Ha Tran.

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Ngoc Ha Tran,

I gave you a list of sectional-analysis software, some of which are free, in my post dated Sep 13, 2016 above.

If the solid cross section is composed of an isotropic material, you may be able to calculate most sectional properties by hand.

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

Yingyi.Liu
Posts: 101
Joined: Thu May 01, 2014 3:16 am
Organization: Kyushu University
Location: Japan

Re: BModes : Input parameters about tower support subsystem

Jason.Jonkman wrote:Dear Lei,

The approach you describe to derive mode shapes for FAST from BModes sounds correct.

I'm sure that the reason for the differences you are seeing between the BModes-generated mode shapes and the mode shapes derived by NREL for the OC3-Hywind tower contained in "NRELOffshrBsline5MW_Tower_OC3Hywind.dat" is because NREL did not use BModes to derive the tower mode shapes for the OC3-Hywind tower contained in "NRELOffshrBsline5MW_Tower_OC3Hywind.dat". When NREL derived the tower mode shapes for the OC3-Hywind tower, the ability to derive tower mode shapes for systems with floating platforms was not available in BModes. Instead of BModes, we used the linearization functionality of a full-system ADAMS model to obtain the tower modes for the NREL 5-MW turbine models (both land-based and floating). That is, we built an ADAMS model of the wind turbine (using the FAST-to-ADAMS preprocessor), enabled all system DOFs, and linearized the model. Then we passed a best-fit polynomial through the resulting tower mode shapes to get the equivalent polynomial representations of the tower mode shapes needed by FAST. For the tower modes, there is not only an influence from the floating platform DOFs on the tower mode shapes, but there can also be a great deal of coupling with the drivetrain and rotor DOFs. While BModes can now account for the floating platform DOFs, it still treats the drivetrain and rotor rigidly, so, the BModes- and ADAMS-generated mode shapes will differ.

Best regards,

Dear Dr.Jason,

You mentioned that you are using the linearization functionality of ADAMS model to get the tower modes. Would you mind to tell that under which menu in ADAMS I can find this functionality? (Our university has purchased the license). I would like to analyze the tower modes of a wires-guyed tower by using both ADAMS and BModes, and then compare the two results.

Thank you very much.
Yingyi Liu
Kyushu University

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Der Yingyi.Liu,

The ADAMS linearizations were run using the *_ADAMS_LIN.acf files generated through the FAST-to-ADAMS preprocessor. These ADAMS command files invoke a linearization analysis through the LINEAR/EIGNSOL command. More information is available in the "ADAMS Preprocessor" chapter of the old FAST User's Guide: https://nwtc.nrel.gov/system/files/FAST.pdf.

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

Yingyi.Liu
Posts: 101
Joined: Thu May 01, 2014 3:16 am
Organization: Kyushu University
Location: Japan

Re: BModes : Input parameters about tower support subsystem

Dear Dr. Jason,

I'm running the ADAMS solver from the ADAMS command prompt to calculate the "NRELOffshrBsline5MW_OC3Hywind" example in you FTP folder http://wind.nrel.gov/public/jjonkman/NR ... Bsline5MW/, taking advantage of the A2AD functionality and the ADAMS_FAST preprocessor (I use the file "NRELOffshrBsline5MW_ADAMSSpecific.dat" included in your folder). However, a problem occurs as follows:

Adams_FAST_error_OC3Hywind.png (30.95 KiB) Viewed 2583 times

Before doing this, I have set up the A2AD interface (i.e., ADAMS2015_1_x64.dll) correctly using the CompileLinkA2AD.bat according to the guidance in FAST user's manual. I have tested the Onshore wind turbine benchmark that the calculated time history results between FAST and ADAMS are almost same.

However, for the offshore case it fails to work correctly. Do you know what may cause this problem, according to the screenshot? Have I missed some settings in the ADAMS_FAST preprocessor or the ADAMS software for this offshore wind turbine case?

Best Regards,
Yingyi LIU

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: BModes : Input parameters about tower support subsystem

Dear Yingyi.Liu,

It is been years since I've ran MSC.ADAMS, but I do recall receiving that error on occasion. If I recall correctly, I've had some success with suggestion (3) i.e. playing with the ERROR argument.

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

Yingyi.Liu
Posts: 101
Joined: Thu May 01, 2014 3:16 am
Organization: Kyushu University
Location: Japan

Re: BModes : Input parameters about tower support subsystem

Jason.Jonkman wrote:Dear Yingyi.Liu,

It is been years since I've ran MSC.ADAMS, but I do recall receiving that error on occasion. If I recall correctly, I've had some success with suggestion (3) i.e. playing with the ERROR argument.

Best regards,

Dear Jason,
We have solved this problem. To get rid of the above errors, one need to change the "Tstart" value in the *.fst file into a small number, e.g., zero, and the "error" value in the *acf file into a larger number.

Best Regards,
Yingyi Liu