"Aero-ness": Smooth VS dimpled surfaces (housings)

Discuss light weight issues concerning road bikes & parts.
Post Reply
User avatar
LouisN
Posts: 2628
Joined: Mon Feb 19, 2007 3:44 am
Location: Canada

by LouisN

Zipp wheels would come to mind when it comes to rim surface and the aero benefits of dimpled surfaces.
I was thinking: Can these physics principles be applied to other types of surfaces ? Because I was thinking of assembling and old school time trial bike, with all the wiring being outside the cockpit/frameset) and use Ilink housings for the build....I wonder if there is any aero drawbacks about this type of housings.

Louis :)
Last edited by LouisN on Fri Nov 02, 2018 10:24 pm, edited 1 time in total.

Jugi
Posts: 447
Joined: Sun Jun 24, 2018 8:10 am

by Jugi

If I remember correctly, 10 cm of exposed housing is generally estimated to produce one watt of drag. If the Ilink’s outer diameter is less than the traditional 5 mm and 4 mm, I quess it may have fractionally less drag.

by Weenie


Shrike
Posts: 1526
Joined: Fri Jun 03, 2016 5:08 pm

by Shrike

My brain's fried trying to understand that. Between work and looking at bike parts all day.. think I need to step away from the computer :D

Do you mean this sort of housing? Use it to create the same aerodynamic effect to save you a watt on your exposed cables? Or are you gonna get a pin and stab the hell out of your cable housing :twisted:

Image

That dimple effect you refer to seems to be generally accepted in stopping air from 'sticking' around. Even saw a bike frame covered in it one time. Wonder what happened with that..

User avatar
LouisN
Posts: 2628
Joined: Mon Feb 19, 2007 3:44 am
Location: Canada

by LouisN

:lol: Sorry my written english sometimes loses itself between my french thoughts, the brain, and computer keyboard... :lol:
Yes, this sort of housing :) .

Louis :)

Vanadias
Posts: 2
Joined: Fri Nov 02, 2018 11:44 pm

by Vanadias

Happen to be from Quebec @LouisN?

I feel your pain on the 2nd language barrier my friend.

tabl10s
Posts: 479
Joined: Wed Aug 23, 2017 5:40 am

by tabl10s

LouisN wrote:
Fri Nov 02, 2018 9:37 pm
Zipp wheels would come to mind when it comes to rim surface and the aero benefits of dimpled surfaces.
I was thinking: Can these physics principles be applied to other types of surfaces ? Because I was thinking of assembling and old school time trial bike, with all the wiring being outside the cockpit/frameset) and use Ilink housings for the build....I wonder if there is any aero drawbacks about this type of housings.

Louis :)
You could always email Zipp and ask or the math department of your local university. In pretty sure there are students ready to show how smart they are.
2016 Orbea Orca OMR:

15.0lbs/6.804kg...😐

2016 Rca:

11.07lbs/5.048kg...😭

2015 Pinarello F8:

13.04lbs/5.915kg...😩

2018 S-Works SL6:

12.04/5.625kg... 😥

Skillgannon
Posts: 3612
Joined: Sat Apr 01, 2006 12:17 am
Location: A bigger rock in the Pacific (AUS)

by Skillgannon

There would be no aero benefit. The aero cost of your housings will increase with their diameter.
This board would be a nicer place if everyone would take themselves less seriously.

I almost miss Mr Search...

youngs_modulus
Posts: 575
Joined: Wed Sep 20, 2006 1:03 am
Location: Portland, OR USA

by youngs_modulus

Actually, this is a pretty straightforward question to answer. Dimples only work for a narrow range of object sizes and airspeeds, and cable housing doesn't fall into that range.

Below is a plot of the drag on a sphere, showing how the coefficient of drag on a rough sphere drops suddenly when the Reynolds number reaches about 5*10^5, or 50,000. Those step-like drops in drag show where the flow over the ball goes from laminar (flowing smoothly around the ball like the water around a slow-moving fish) to turbulent.

Image

The dimples on the rough sphere cause the air to become turbulent sooner (at a lower speed) than it does over a smooth sphere. Since the transition from laminar to turbulent flow comes with a drop in drag, the rough sphere has lower drag for some speeds. (Note that the smooth sphere has drag equal to or lower than that of the rough sphere once the Reynolds number gets high enough. This is why I say that dimples only help for a narrow range of conditions).

What is Reynolds number? It's a unitless value that characterizes the transition between laminar and turbulent flow. And since dimples on golf balls attempt to manipulate that transition by inducing turbulent flow, knowing the Reynolds number is really helpful for evaluating dimples on anything, whether golf balls or bikes.

Basically, you can calculate the Reynolds number by using this formula: R = (u*L)/nu, where:

R is the Reynolds number we're solving for
u is the velocity in m/s (11.1 m/s is about 40 KPH or 25 MPH)
L is the "characteristic length," which, for a cylinder perpendicular to the airflow, is pretty much the diameter of the cylinder. Let's call it 5 mm, or 0.005 m.
nu is the kinematic viscosity of air, which happens to be 1.562*10^-5 m^2/s at 20 degrees C.

Solving the above equation, I get a Reynolds number of about 355 for our cable housing. The Reynolds number needs to be over 50,000 for dimples to start reducing drag. You'd need to hit about 1500 m/s (just under 3500 MPH) before dimples in your cable housings started reducing drag, so that's a little Strava challenge for you.

But you'd go supersonic at about 345 m/s (767 MPH), long before your fancy housings started reducing drag. 3500 MPH is about Mach 4.5. Supersonic flow is pretty different from subsonic flow, so our chances our your dimples still wouldn't help, even if you could pedal your bike at Mach 4.5. (Your tires will get shredded from the centrifugal force long before Mach 1, so there are a lot of issues here).

The characteristic length term in the Reynolds number equation is what drives this, so objects with a larger characteristic length benefit from dimples at speeds far below 3500 MPH.


Edit: Altered sentence fragment, minor edits for clarity
Last edited by youngs_modulus on Mon Nov 05, 2018 11:49 pm, edited 1 time in total.

User avatar
bikerdan
Posts: 135
Joined: Thu Dec 02, 2004 4:03 pm
Location: Germany

by bikerdan

That's the most amazing answer I've ever read on this forum

Sent from my BND-L21 using Tapatalk


Shrike
Posts: 1526
Joined: Fri Jun 03, 2016 5:08 pm

by Shrike

Muhahaa brilliant :D

User avatar
LouisN
Posts: 2628
Joined: Mon Feb 19, 2007 3:44 am
Location: Canada

by LouisN

:lol: That was both entertaining and educative Youngs_modulus :beerchug: ...

I saw a kids science program the other day where the teacher demonstrated that a dimpled ( with a special spoon to create a golf ball like shape) watermelon thrown from a catapult would go farther that a smooth surfaced one. That was a pretty slow speed demonstration...can't find it though... :(

Louis :)

youngs_modulus
Posts: 575
Joined: Wed Sep 20, 2006 1:03 am
Location: Portland, OR USA

by youngs_modulus

I'm glad you guys found my post helpful!

I just edited my original post a little. I wanted to highlight how dimples only help in a narrow range of speeds and object sizes. Louis, that watermelon probably had a characteristic length in the neighborhood 300 mm, not the 5mm of brake cable housing, so that would make dimples much more likely to have a positive effect at low-ish speeds.

by Weenie


XCProMD
Posts: 733
Joined: Sat Jan 26, 2008 10:25 am
Location: Cantabria

by XCProMD

Image

Dimples bring back good memories!



Inviato dal mio iPhone utilizzando Tapatalk

Post Reply
  • Similar Topics
    Replies
    Views
    Last post