## DrTrDOF impacting power output

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

Moderators: Bonnie.Jonkman, Jason.Jonkman

Drew.Gertz
Posts: 2
Joined: Tue Sep 12, 2017 4:30 am
Organization: Northwind Engineering OÜ
Location: Estonia
Contact:

### DrTrDOF impacting power output

Hello,

I am translating to OpenFAST a HAWC2 model provided by DTU of a 25kW upwind, geared, 3 bladed turbine with asynchronous generator and 13m rotor diameter. I am seeing strange behaviour in the power output when I turn on the DrTrDOF. See the figure below for wind stairs from 5 m/s:

power-wind.png (277.26 KiB) Viewed 119 times

The blue curve is with DrTrDOF turned off, and the red is when I turn it on with the values I calculated, and the black is when GenIner is 100x higher than my calculated value. First of all the response is very slow for some reason for my calculated values of DTTorSpr and DTTorDmp. Secondly the power level seems to scale with GenIner.

I am wondering if it is something about the combination of parameters? I have calculated the GenIner as follows:

GenIner = GenIner_LSS/GBRatio^2 = 190.81/23.25^2 = 0.353

and DTTorSpr, and DTTorDmp according to the equations in the post here: viewtopic.php?t=940

K=G J/L
where
K:DTTorSpr
G: Modulus of rigidity for the shaft material
J=pi*d1^4/32 and
L=l1+(GBratio)^2*l2*(d1/d2)^4
d1 and d2 are diameters of LSS and HSS, respectively;
l1 and l2 are lengths of LSS and HSS, respectively.

G [N/m2] 7.93E+10 for stainless steel
d1 [m] 1.20E-01
d2 [m] 6.00E-02
l1 [m] 7.00E-01
l2 [m] 5.00E-01
gbratio 23.25

J= 2.04E-05
L = 4.33E+03
DTTorSpr (K)= 3.73E+02

DTTorDmp can be calculated as follows:
DTTorDmp =Cc*ζ
where,
Cc is critical damping coefficient =2*sqrt*(K*rotor inertia);
ζ is fraction of critical damping or the percentage of damping.

Rotor inertia = 3448.517 from ED.sum
Cc = 2.27E+03
ζ = 0.6
DTTorDmp = 1.36E+03

fixed-free fn (in Hz) = SQRT( DTTorSpr/RotIner )/(2*pi) = 0.05 Hz

The resulting fixed-free natural frequency also seems to be very low, although I don't have any reference for what it should be for this size and type of turbine, I guess it should be more like ~2 Hz?

Best Regards,
Drew Gertz
Northwind Engineering OÜ
www.northwindengineering.com

Jason.Jonkman
Posts: 3857
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

### Re: DrTrDOF impacting power output

Dear Drew,

Yes, I would guess your drivetrain torsional spring constant (DTTorSpr) is too low by several orders of magnitude (and DTTorDmp is too low accordingly). From the FAST/OpenFAST certification/regression tests, DTTorSpr for the UAE Phase VI turbine (20 kW turbine) is 199000 Nm/rad and DTTorSpr for the AOC-15/50 turbine (50 kW turbine) is 600000 Nm/rad. I would guess your DTTorSpr value should fall in (or near) this range.

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

Drew.Gertz
Posts: 2
Joined: Tue Sep 12, 2017 4:30 am
Organization: Northwind Engineering OÜ
Location: Estonia
Contact:

### Re: DrTrDOF impacting power output

Hi Jason,

Thanks for your reply. I clearly have to revisit the method I used to calculate DTTorSpr .

I tested some higher DTTorSpr values within the range you provided. The following combination turned out to be acceptable:

6e5 DTTorSpr
5.5e4 DTTorDmp (ζ = 0.6)

The resulting fixed-free natural frequency is 2.1 Hz which seems reasonable as well.

My GenIner value of 0.353, however, results in drivetrain resonance. I had to increase it to 1.5 to solve it, but overall I consider the problem to be solved.

Thanks!

Best Regards,
Drew Gertz
Northwind Engineering OÜ
http://www.northwindengineering.com