Question about off-axis drag numbers in aero wheel tests
Moderator: robbosmans
When a wheel test is done with a wind tunnel speed of let's say, 30 mph, and a wheel is tested off axis, the wind speed on the wheel's axis is now lower than 30 mph. Is this fact adjusted for in published results? I.e. for a drag result at 5 degrees at a nominal 30 mph, is that for a forward motion of 30 mph?
Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓ Broad Selection ✓ Worldwide Delivery ✓
www.starbike.com
Boomer wrote:When a wheel test is done with a wind tunnel speed of let's say, 30 mph, and a wheel is tested off axis, the wind speed on the wheel's axis is now lower than 30 mph. Is this fact adjusted for in published results? I.e. for a drag result at 5 degrees at a nominal 30 mph, is that for a forward motion of 30 mph?
Yes, I believe, the data shows power/cda/whatever for the direction of the bike.
This is a non-trivial error at 20 degree yaw.
Of course if there's a 20-deg yaw the wind is probably so inconsistent precision is hopeless anyway.
Of course if there's a 20-deg yaw the wind is probably so inconsistent precision is hopeless anyway.
Boomer wrote:Is this fact adjusted for in published results?
Hopefully someone with tunnel experience will chime in, but my guess is no. If they spin the wheels then they will be at a constant 30mph, the tunnel flow velocity will be at 30mph, and the drag will be measured in line with the item. Usually the units will be in grams. If you see numbers in grams, and nothing is said about corrections, I'd assume that there aren't any. In truth the numbers will also need to be corrected for slight variations in flow velocity, air density, air temperature, humidity, etc. If they wanted to publish easily useful info it would be converted to CdA.
And don't forget that the power to spin the wheel is another drag term that slows you down in the real world, but isn't measured in a standard tunnel test.
formerly rruff...
-
- Shop Owner
- Posts: 1980
- Joined: Sat Jun 13, 2009 4:02 am
- Location: NoVA/DC
I had a conversation with Josh Poertner at Zipp about that a while ago, and I seem to recall that they do, in fact, factor in the force required to spin the wheel. I could be mistaken...
"In addition, the elevated platforms designed to move the bicycle frame and wheel above the
boundary layer on the floor of the wind tunnel will vary, and in many cases, results are normalized to match tunnel
wind speeds with measured axial forces (fn.2)."
www.acusim.com/papers/AIAA10_2010-1431_MNGodo.pdf
""drag" is really axial force (body-axis coordinate system, not wind axis coordinate system) " - biketechreview user after wind tunnel
" the force measurement system is in the body axis coordinate system not the wind axis coordinate system." - Kraig Willet
boundary layer on the floor of the wind tunnel will vary, and in many cases, results are normalized to match tunnel
wind speeds with measured axial forces (fn.2)."
www.acusim.com/papers/AIAA10_2010-1431_MNGodo.pdf
""drag" is really axial force (body-axis coordinate system, not wind axis coordinate system) " - biketechreview user after wind tunnel
" the force measurement system is in the body axis coordinate system not the wind axis coordinate system." - Kraig Willet
thisisatest wrote:I had a conversation with Josh Poertner at Zipp about that a while ago, and I seem to recall that they do, in fact, factor in the force required to spin the wheel. I could be mistaken...
A long time ago Zipp published some "power to spin" data, but it was obviously not corrected... it was the power drawn by the motor that turned the drum that spun the wheel. It was much too high. Kraig Willet mentioned once that the tunnel in San Diego would make that info available if the customer wanted it. I haven't seen any mention of that sort of data since.
It would take a pretty specialized test to get that info accurately. The uniform stream in the tunnel would screw it up a decent amount since on the road the flow would be zero at the ground.
formerly rruff...
If the air velocity is constant, then the wheel forward speed would be:
wheel speed = air speed * cos(yaw angle)
At a constant airspeed of 30mph, the wheel speed would be:
29.5 mph at 10 degrees yaw
28.2 mph at 20 degrees yaw
26.0 mph at 30 degrees yaw
So, not trivial indeed at the higher angles. Surely they are correcting for this.
I'm not an aerodynamics engineer, but as a test engineer I'd run the wheel at various win speeds while changing yaw. Then I'd plot the data with cos(yaw)*wind speed as the X axis and drag on the Y axis. From that, curves of constant wheel speed should be able to be interpolated.
wheel speed = air speed * cos(yaw angle)
At a constant airspeed of 30mph, the wheel speed would be:
29.5 mph at 10 degrees yaw
28.2 mph at 20 degrees yaw
26.0 mph at 30 degrees yaw
So, not trivial indeed at the higher angles. Surely they are correcting for this.
I'm not an aerodynamics engineer, but as a test engineer I'd run the wheel at various win speeds while changing yaw. Then I'd plot the data with cos(yaw)*wind speed as the X axis and drag on the Y axis. From that, curves of constant wheel speed should be able to be interpolated.
Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓ Broad Selection ✓ Worldwide Delivery ✓
www.starbike.com
+1
This was the same potential result that I was thinking of - if they just blast a 30 mph breeze at various angles then in reality (1) the "bike" is at first moving forward at 30 mph with no breeze what so ever, but then as the angle changes you not only seemingly slow down (per the calculations provided using cosine) but you also suddenly pick up a cross wind that didn't exist in any way what so ever in the 0 yaw condition. Almost like you are getting not only 1 negative (slowing down) but a second negative of the increasing cross wind.....or another for instance - let's say I am moving along at 20 mph ground speed into a 10 mph head wind, then that becomes a 10 mph cross wind, my actual ground speed isn't going to slow down, instead it's going to speed up.....if it's true that they simply rotate the fan then it sounds like while they feed us consistent results, they also may not be real world results.............
This was the same potential result that I was thinking of - if they just blast a 30 mph breeze at various angles then in reality (1) the "bike" is at first moving forward at 30 mph with no breeze what so ever, but then as the angle changes you not only seemingly slow down (per the calculations provided using cosine) but you also suddenly pick up a cross wind that didn't exist in any way what so ever in the 0 yaw condition. Almost like you are getting not only 1 negative (slowing down) but a second negative of the increasing cross wind.....or another for instance - let's say I am moving along at 20 mph ground speed into a 10 mph head wind, then that becomes a 10 mph cross wind, my actual ground speed isn't going to slow down, instead it's going to speed up.....if it's true that they simply rotate the fan then it sounds like while they feed us consistent results, they also may not be real world results.............