Create a "filled in" Jacobs 0021 dataset and re-run the performance curve

[petebachant]

• [x] Add Sheldahl CSVs for 0 through 180 degrees angle of attack
• [x] Digitize as much of the Jacobs data as possible
• [x] Mirror the Jacobs data about 0 degrees angle of attack
• [x] Use Sheldahl data beyond the angles of attack measured by Jacobs
• [x] Write script to spit out DataFrame in CACTUS/NREL text format
• [x] Allow selecting foil data at run time in run.py.

[petebachant]

Not seeing much difference using the LB model. Even have been recalculating the critical lift coefficients. Haven't been recalculating lift slopes though.

[petebachant]

Running with Boeing--Vertol model looks very promising, but crashes at tsr=3.1:

Timestep: 132
0***** NON-LINEAR ITERATION LOOP DID NOT CONVERGE IN 10 ITERATIONS. PROGRAM TERMINATED. *****

@whophil, do you know what parameters to tweak to fix this?

[petebachant]

Looks like using the default BV stall angles helps a lot. Not sure how these were calculated, but I was trying to recalculate based on the drag coefficient data.

[nearyvs]

These override lift and drag coefficients in lookup table, right?

So curious if we can find out under what conditions we discard these coefficients and visa versa.

Sent from my iPad

On Jun 26, 2016, at 12:39 PM, Pete Bachant notifications@github.com wrote:

Looks like using the default BV stall angles helps a lot. Not sure how these were calculated, but I was trying to recalculate based on the drag coefficient data.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[petebachant]

The Jacobs data is used over its available range of angle of attack, and is extended beyond with Sheldahl data.