BikeTechReview.com

  • Increase font size
  • Default font size
  • Decrease font size
Home Forum
Welcome, Guest
Username Password: Remember me

negative watts to spin
(1 viewing) (1) Guest
  • Page:
  • 1

TOPIC: negative watts to spin

negative watts to spin 6 years, 7 months ago #26543

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
pretty cool to see the physical measurements I've been making of "watts to spin" match up with this caveman check:

http://biketechreview.com/images/biketec...w_negative_watts.wmv
-kraig

Last Edit: 6 years, 7 months ago by kraig. Reason: fix tag

Re: negative watts to spin 5 years, 10 months ago #26824

  • MikeMartin
  • OFFLINE
  • Fresh Boarder
  • Posts: 2
  • Karma: 0
[I originally posted this on the BTR blog, but thought this might be more appropriate for forum discussion -- http://biketechreview.com/blog/watts-to-spin.html#comments]

Hi Kraig-

I'm attempting to understand if the "negative watts to spin" effect is merely an artificial remnant of wind speed gradient and stand interactions effects unique to the test setup in a wind tunnel, or if they might occur on the road as well.

In the video, the wind speed is the same on the top and bottom portions of the wheel (neglecting floor effects). All that is necessary for the wheel to begin spinning under those conditions is for the drag on the upper blades to be less than the drag on the lower blades. It is not necessary to have negative drag on any portion of the wheel to cause it to spin. Also, the support stand may have the effect of increasing the drag on the lower blades and contribute to the spinning effect.

On the road with zero wind, or in a tunnel where the wheel is externally spun up to match the wind speed at zero yaw, the apparent wind that the wheel sees has zero speed at the bottom of the wheel, and 2x the translational speed at the top of the wheel. If the blade drag is indeed positive at all orientations, the blades should experience higher drag in the upper portion of the wheel due to the higher apparent wind speed in that area. (Though this might be offset by the asymmetry of the blade shape -- leading edge and trailing edges swap on the upper and lower positions...)

WRT non-zero yaw, if there were a negative drag effect on the blades, I would expect it to occur at higher yaw values. And the on-the-road apparent wind will show a yaw gradient with the lowest value on top and the highest value on the bottom of the wheel, causing a net torque opposite the spinning direction.

I'd welcome any insights you might have on my comments above.

Finally, some of your BTR articles suggest that you've measured "negative watts to spin" under certain conditions. Can you share more details about the test protocol? Were the wheels externally spun so that their tangential speed matched the wind tunnel speed to simulate the on-the-road apparent wind effects? Were interactions between the wheel and support stand taken into account?

Thanks for your help.

-Mike
Last Edit: 5 years, 10 months ago by MikeMartin. Reason: URL garbled

Re: negative watts to spin 5 years, 10 months ago #26825

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
MikeMartin wrote:


In the video, the wind speed is the same on the top and bottom portions of the wheel (neglecting floor effects).


I don't think neglecting "floor effects" is legit. In the tunnel, the velocity at the wall is zero, so there will be a velocity profile that goes from zero at the wall (ground plane), up to free stream as one moves away from the wall. in the video, the wind is coming from the right.

All that is necessary for the wheel to begin spinning under those conditions is for the drag on the upper blades to be less than the drag on the lower blades.


kind of. i think what is necessary to create negative watts to spin is an aerodynamic torque around the model center/hub in the right direction. this model/wheel/body axis torque can be affected by a component of sideforce in the wind axis coordinate system.

Also, the support stand may have the effect of increasing the drag on the lower blades and contribute to the spinning effect.


yes, interaction effects are tricky. I think in this case, however, the struts cause a lower average effective windspeed on the lower half of the wheel. it is tough to do the interaction study thoroughly/correctly when one has a budget that tops out at the beer can and tussle of string level, though!

On the road with zero wind, or in a tunnel where the wheel is externally spun up to match the wind speed at zero yaw, the apparent wind that the wheel sees has zero speed at the bottom of the wheel, and 2x the translational speed at the top of the wheel. If the blade drag is indeed positive at all orientations, the blades should experience higher drag in the upper portion of the wheel due to the higher apparent wind speed in that area. (Though this might be offset by the asymmetry of the blade shape -- leading edge and trailing edges swap on the upper and lower positions...)


it is the torque around the model center that determines the watts to spin. one can have a net negative axial force for a given static wheel orientation if the wind axis side force to drag ratio (as a function of beta) is high enough.

WRT non-zero yaw, if there were a negative drag effect on the blades, I would expect it to occur at higher yaw values.


the video shows this effect, no? typical axial force "buckets" occur in the 10 degree +/- range - the negative watts to spin seem consistent with this, no?

And the on-the-road apparent wind will show a yaw gradient with the lowest value on top and the highest value on the bottom of the wheel, causing a net torque opposite the spinning direction.


I guess so...which means that in order for negative watts to spin to occur, the top half of the wheel needs the right ambient conditions, the right wheel speed, and the right wheel geometry. It's sort of a Goldilocks kind of deal?

Finally, some of your BTR articles suggest that you've measured "negative watts to spin" under certain conditions. Can you share more details about the test protocol?


the tunnel I used for these measurements is equipped with a six component balance - the rest is just math!

Were the wheels externally spun so that their tangential speed matched the wind tunnel speed to simulate the on-the-road apparent wind effects?


yes.

Were interactions between the wheel and support stand taken into account?


strut tares were taken and removed as a part of the data reduction. this is one way to do these sorts of things. is it the best way? probably not!

i think, though, that at the end of the day, the proof of the negative watts to spin pudding is in the tasting...and the video shows the wind coming from the right, and the wheel spinning against the wind under it's own power...couple that with the fact that there are a couple of factors intuitively working against this rotation (wall velocity = zero, struts reducing average axial velocity magnitude on lower half of wheel), and that an independent measurement predicted this behavior, and I'd say things aren't looking like an artificial artifact of the test setup. independent lines of inquiry converging to a consistent conclusion generally means there is something there in my experience.

hope this helps in your understanding,
-k
-kraig

Last Edit: 5 years, 10 months ago by kraig.

Re: negative watts to spin 5 years, 10 months ago #26826

  • MikeMartin
  • OFFLINE
  • Fresh Boarder
  • Posts: 2
  • Karma: 0
Thanks, Kraig.

kraig wrote:
MikeMartin wrote:


In the video, the wind speed is the same on the top and bottom portions of the wheel (neglecting floor effects).


I don't think neglecting "floor effects" is legit. In the tunnel, the velocity at the wall is zero, so there will be a velocity profile that goes from zero at the wall (ground plane), up to free stream as one moves away from the wall. in the video, the wind is coming from the right.

Yah. The floor effect is interesting. In the tunnel, it will certainly have a velocity gradient of some shape from zero at the floor up to windspeed some height h above the floor. I guess the question is: how does this gradient compare to the apparent wind gradient on the road that varies from zero at ground level up to 2x bike translational speed at the top of the wheel?

All that is necessary for the wheel to begin spinning under those conditions is for the drag on the upper blades to be less than the drag on the lower blades.


kind of. i think what is necessary to create negative watts to spin is an aerodynamic torque around the model center/hub in the right direction. this model/wheel/body axis torque can be affected by a component of sideforce in the wind axis coordinate system.

Sure, we're on the same page. I was considering the moments from blades in the upper and lower positions as components of the torque about the hub axis. My reasoning is that the moment from a lower blade is negligible because of the low apparent wind speed in that area, while the moment from an upper blade is much larger (up to 2x larger) on the road than while at rest in the tunnel.

Also, the support stand may have the effect of increasing the drag on the lower blades and contribute to the spinning effect.


yes, interaction effects are tricky. I think in this case, however, the struts cause a lower average effective windspeed on the lower half of the wheel. it is tough to do the interaction study thoroughly/correctly when one has a budget that tops out at the beer can and tussle of string level, though!

True dat.

On the road with zero wind, or in a tunnel where the wheel is externally spun up to match the wind speed at zero yaw, the apparent wind that the wheel sees has zero speed at the bottom of the wheel, and 2x the translational speed at the top of the wheel. If the blade drag is indeed positive at all orientations, the blades should experience higher drag in the upper portion of the wheel due to the higher apparent wind speed in that area. (Though this might be offset by the asymmetry of the blade shape -- leading edge and trailing edges swap on the upper and lower positions...)


it is the torque around the model center that determines the watts to spin. one can have a net negative axial force for a given static wheel orientation if the wind axis side force to drag ratio (as a function of beta) is high enough.

Sure. But again, the component of the model center torque from the lower blade is negligible on the road, and substantial in the tunnel. In order to get a net negative drag on a wheel spinning at wind speed (i.e. on the road or externally spun up to speed in the tunnel), the torque component from the upper blade(s) must be negative at some point in their sweep -- not merely positive but less than the torque component from the lower blade(s).

WRT non-zero yaw, if there were a negative drag effect on the blades, I would expect it to occur at higher yaw values.


the video shows this effect, no? typical axial force "buckets" occur in the 10 degree +/- range - the negative watts to spin seem consistent with this, no?

I agree. But it seems impossible to determine if the individual torque components from blades in various positions are positive or negative with the setup shown.

And the on-the-road apparent wind will show a yaw gradient with the lowest value on top and the highest value on the bottom of the wheel, causing a net torque opposite the spinning direction.


I guess so...which means that in order for negative watts to spin to occur, the top half of the wheel needs the right ambient conditions, the right wheel speed, and the right wheel geometry. It's sort of a Goldilocks kind of deal?

Yah. On the road, the apparent wind vector at the tire contact patch is due solely to the off-axis real wind. At the top of the wheel, the apparent wind is the vector sum of the real wind and 2x the bike's translational vector (which must necessarily be lower yaw than the true wind vector.)

Finally, some of your BTR articles suggest that you've measured "negative watts to spin" under certain conditions. Can you share more details about the test protocol?


the tunnel I used for these measurements is equipped with a six component balance - the rest is just math!

Were the wheels externally spun so that their tangential speed matched the wind tunnel speed to simulate the on-the-road apparent wind effects?


yes.

Thanks! This does indeed strengthen the case for "negative watts to spin" in the real world.

Were interactions between the wheel and support stand taken into account?


strut tares were taken and removed as a part of the data reduction. this is one way to do these sorts of things. is it the best way? probably not!

If you ever get back into the tunnel, it might be interesting to use a different strut geometry to ensure there is no weird interaction with the wheel -- maybe some kind of "L" shaped strut that mounts behind the wheel, or a really long axle between the struts...

i think, though, that at the end of the day, the proof of the negative watts to spin pudding is in the tasting...and the video shows the wind coming from the right, and the wheel spinning against the wind under it's own power...couple that with the fact that there are a couple of factors intuitively working against this rotation (wall velocity = zero, struts reducing average axial velocity magnitude on lower half of wheel), and that an independent measurement predicted this behavior, and I'd say things aren't looking like an artificial artifact of the test setup. independent lines of inquiry converging to a consistent conclusion generally means there is something there in my experience.

Agreed. Thanks again.

One last bug to throw into the soup is the interaction of the blades with a fork on a complete bike. If the drag is higher as a blade passes through the fork, it could seriously skew the rotational drag picture...

hope this helps in your understanding,
-k
Last Edit: 5 years, 10 months ago by MikeMartin. Reason: Clean up formatting

Re: negative watts to spin 5 years, 10 months ago #26827

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
MikeMartin wrote:

I agree. But it seems impossible to determine if the individual torque components from blades in various positions are positive or negative with the setup shown.


from a gradient perspective, things are going in the same direction as the "real world" albeit at a lower magnitude. if i hadn't done this experiment and seen the reversal in wheel rotation direction as a function of beta, my spidey senses would be full of more angst than they are right now.

I've seen one style of wheel produce negative watts to spin (based on memory!) - all others behave as one would expect.

regarding fork/wheel interaction: I've done quite a bit of study on this at a couple different tunnels...and, it could be true that fat rimmed/faired are significantly worse when watts to spin are taken into account.
-kraig

Re: negative watts to spin 5 years, 4 months ago #26957

  • howardjd
  • OFFLINE
  • Senior Boarder
  • Posts: 41
  • Karma: 0
About a year ago I got pretty intrigued when I saw this video and could not really understand what was going on. So naturally I grabbed my tri spoke and sat around looking at it. One I realized that the wheel is still not spinning to match with the wind speed so its really just a little wattage boast, meaning its still going to take some positive input to get the wheel up to the wind speed. Also after having a discussion with somebody about a wells turbine I realized this is the shape of a tri spoke(at least my HED anyway). I then went down to the shop turned on a fan put the trispoke perpendicular to the airflow and sure enough it rotated forward, pretty cool. You'd think that somebody would make a wheel with better wells turbine characteristics for riding a really windy course like say Kona, everytime you got plowed by a side wind you'd get a boast forward!
  • Page:
  • 1
Time to create page: 1.61 seconds

Poll

Which type of tire is more aerodynamic?