FAST throws: Rzero must be < 1

Dear All,

I am trying to get the aerodynamic loads on an 0.5m length blade for structural analysis project and I am only interested in the forces on the blade only. the model I am using is a wind turbine with a tower height of 3.5m and I was going to use YawDyn with AeroDyn to get the loads but since YawDyn is no longer supported or recommended by NREL so I am using FAST instead , and I am getting this error and I can not figure out how to solve it :

Value of Rzero = 1 must be smaller than 1.

Aborting FAST.

thanks in advance,
Ahmed.

Dear Ahmed,

The error regarding Rzero that you are reporting is generated by AeroDyn if a blade node ever extends beyond the radius of the blade tip as defined at model initialization. My understanding is that the error can be generated from some combination of precone, small element length (DR) at the tip, or excessive blade stretching.

If excessive blade stretching is part of the problem, you are likely also getting a warning from FAST regarding a small angle assumption violation. We’ve answered a question regarding how to address this warning in the post dated October 1, 2009 in the topic found here: http://forums.nrel.gov/t/nrel-5mw-controls-dll-interface/183/1.

If excessive blade stretching is not part of the problem, a suitable workaround may be to increase the length (DR) of the outermost element (unless you have to increase DR so much that the discretization is too coarse for sufficient accuracy).

I hope that helps.

Best regards,

Dear Jason,

Thanks alot, I decreased the time step and changed the initial conditions and it worked :slight_smile:

time step = 4e-6
and total time 20 seconds

it took about an hour :slight_smile: will it have any different effect on the results if I have decreased the total time to may be 5 seconds.

thanks again, appreciate your help

Regards,
Ahmed

Dear Ahmed,

A time step of 4e-6 s is extremely small. Are you sure you’ve specified the structural mass and stiffness properties properly in your FAST model? Incorrectly specified mass and stiffness properties may lead to unrealistically high natural frequencies, which require unecessarily small time time steps.

I’m not sure what you’re asking regarding 20-s or 5-s simulation lengths. Typically, it is necessary to run hundreds if not thousands of 10-min simulations ecompassing a range of wind speed and operational scenarios to characterize the response of a given turbine design.

Best regards,

Hi Guys

I also have received a “Value of Rzero = 1 must be smaller than 1 in xphi().” warning. After rummaging around in the code for a little while I found some maths about 3D geometry, which isn’t really my bag. Especially when it comes without pretty diagrams or lots of comments. (Although it is likely that any amount of commenting might not have been enough for me!)

It occurred 180 seconds into a simulation. Previous simulations had been very happy when my precone was -4°. When I changed my precone angles to -7 degrees it fell over.

I figured out that it seemed to be complaining that the node has sort of fallen off the end of the blade, which is what Jason said above, also that it can be linked to precone. Also that the node needs to stay within the blade for mode things and curve matching.

It seems to me that changing the precone from -4 to -7 is an entirely reasonable thing to do, and that the blade node positions on the blades should be quite happy with moving with the blade with its slightly different cone angle.

The sort of mistake that I would probably have made would be to put the nodes on some sort of static plane that doesn’t take account of precone rather than on the blades where they belong or the radius ‘R’ isn’t aligned with the blade.

Please could someone look at the code. I’d be curious to know if there are plans afoot to fix this issue?

Thanks

Alec

Dear Alec,

As I mentioned in my May 25, 2011 post above, the error regarding Rzero is generated by AeroDyn if a blade node ever extends beyond the radius of the blade tip as defined at model initialization (i.e., in the undeflected state). For upwind precone, blade thrust tends to increase the swept area of the rotor disk because the blade deflection counteracts the precone to some extent. For small to moderate precones, increasing precone will increase the change in swept area, so, it makes sense that 7 degrees of precone could cause problems even though 4 degreees does not. In the current version of AeroDyn, my understanding is that the only workaround is to increase the length (DR) of the outermost blade element. I agree that this is problem that needs to be fixed. I will pass this feedback on to the team that is working on the AeroDyn overhaul.

Best regards,