Aerodynamics / Aerodynamic Cost / Aero parts

Anyone got any handy comparisons between round and aerofoil tubing types and what, if any, aero benefits they might have?

Any idea re: existing tubes though? I don't recall seeing many frames with an airfoil that severe.

Yeah that aerofoil has a fair bit of camber, so it produces lift at zero incidence hence wouldn't be a good choice for a a bicycle.

It wouldn't be surprised if the aerofoil section used on bicycle frames is one of the NACA 4-digit aerofoils, so NACA 00XX range, probably a NACA 0012, where the '12' is the percentage of max. thickness to chord length.

airfoiltools.com has a decent database of aerofoil properties here is the NACA 0012: http://airfoiltools.com/airfoil/details?airfoil=n0012-il
Alpha represents angle of attack, which for a bicycle would be the yaw angle. If you assume the Reynolds number on the headtube of a bicycle is roughly 5x10^4, it should give you a good good idea of the aero properties of the section. You can also assume the CD of a cylinder is around 1 at this Reynolds number (CL is obviously 0).

Should also remember the tubing on bicycles is three dimensions this isn't ideal and distributions of the aero coefficients over the span of the tubes will change. Also once the air reaches places like the downtube the air is already really turbulent and messy, so it's not quite this simple.

If you want to work out the actual drag force that you need to overcome, you just multiply the drag coefficient by 0.5*1.225*velocity^2*surface area

Not as sciencey but maybe more applicable info? Nice bit of analysis here from Specialized's steel Allez vs Venge video (including other aero-road frames).

It wouldn't be surprised if the aerofoil section used on bicycle frames is one of the NACA 4-digit aerofoils, so NACA 00XX range, probably a NACA 0012, where the '12' is the percentage of max. thickness to chord length.

UCI bikes are regulated to a thickness of just over 30% of chord length measured perpendicular to the tube's long axis. Measuring chord the aeroplane way parallel to the direction of travel, that's about 25% for the head and seat tubes and about 15% for the downtube. Because the downtube has a lot of sweep in aeroplane terms, it creates what would be spanwise flow if it were a wing, so thickness:chord measured in the direction of the flow field could be under 10% at some points if the basic tube shape was as thin as is permitted.

Had not thought about the sweep effects that would occur, interesting thanks!

Do UCI rules prohibit airways within the frame? Could take high-pressure air from near the headtube and duct it back behind the seatpost...

Do UCI rules prohibit airways within the frame? Could take high-pressure air from near the headtube and duct it back behind the seatpost...

By implication, kind of. The regulatory model is solid, a duct makes the thickness of the regulated solid the wall thickness of the duct, 25mm minimum. Not that it matters, because the drag in the duct would be so high that it would cancel out any relief of high pressure on the nose cone or filling of low pressure behind the seat tube.

Cervelo published this at some point:

Other sources state that max aero benefit of a superduper airfoil vs a cylinder of the same width is about 1:10 at bike speeds/bike tubing sizes. (with a headwind or no wind)

So the above seem a bit generous (1:24), but cervelo usually know what they are doing.

edit: Also with sidewind forward propulsion is possible. Maybe that is why the numbers are so good. The ratio could be calculated as a average of different wind angles

Other sources state that max aero benefit of a superduper airfoil vs a cylinder of the same width is about 1:10...So the above seem a bit generous (1:24)

Both statements can be true, if the Cervelo graph is for an infinitely long tube with its axis at right angles to a homogeneous flow field and the "other source" is for a finite tube raked at bicycle tube angles in the practical bicycle flow field. The actual section of a circular down tube along the flow line is more like the ellipse represented by the second bar on Cervelo's graph.

the "other source" is for a finite tube raked at bicycle tube angles in the practical bicycle flow field.

No this was just from places like the one linked above http://airfoiltools.com/search/index?m%5BmaxCamber%5D=0&m%5Bsort%5D=5 which shows 2D infinite airfoils. The drag coefficient of airfoils continue to improve at higher speeds/bigger sizes (than bikes)

umop3pisdn 's example is only true at much higher speeds than 30mph

The trek speed concept presentation also has a comparison with different shapes at page 24

1 Attachment

• TK10_SC_white_paper_lores.pdf

Ta, I'll have a read.

Should I build a new TT bike for low yaw (faster) races? n+1 rule...

I'd like to mount a speed sensor to my P3C. I have one of those hub mounted Garmin things, which calculates speed using an accelerometer I believe. The problem is fitting this to a rear disc (as fitting to the front wheel would be a bad idea).

Other than butchering it up to somehow fit to the disc's hub, or to fit inside the valve hole (would the sensor work from that position?), what options do I have?

What's the smallest/most aero Ant+(and BT?) speed sensor I can buy?

Most people glue tiny neodymium magnets to their discs and use whatever sensor they use.

I'll probably need to do the same thing soon so would be interesting to see what you find.

Yeah, I think I'll experiment tomorrow by placing the Garmin sensor thing by the valve stem to see if it still provides a speed reading. If that works well then I'll try and make it as small as possible to wedge into the valve hole.

Also, am I a tart for wanting a set of valve covers from racewaredirect?...

am I a tart for wanting a set of valve covers from racewaredirect

Do they provide any advantage/disadvantage compared to stickers?

Anyone on here got one of the order 1 Aerocoach Aeox discs?
Be interested to know what cassette anyone has managed to fit, at first attempt my ultegra 11-25 clashes with the DS skin both on the top 3 sprocket's carrier and with the chain on the 25t.
Might be missing something - as taking a file to a new 11-28 is the only solution I see currently...

Ask @hippy about running 10 sprockets on an 11sp cassette :)

Cheers,

@hippy presumably drop the 16t between the 15 and 17-19 carrier mounted pair?

Have you asked @xavierdisley about the cassette fouling?

Here’s Hippy’s blog post:
https://thehippy.net/blog/raltech-disc-cover-11-speed-cassette-spacing/

Mega, thanks.

Losing a sprocket from an 11-25 leaves a 2 tooth jump in the middle of sequential sprockets, so not ideal.

Have shot an email over to Aerocoach to see what cassettes they used in testing.

I'm going to see if I have an 11-28 anywhere that'll potentially give me more clearance.

I've still got my custom 10sprocket/11spd cassette and some 11-28s you can test if you're anywhere near west London.

Is it a 10spd freehub body? Do you have the 1.85mm (I think) spacer fitted behind the cassette?

Thanks for the response, read through your blog post as well! Unfortunately I'm in Derby so trying your solutions likely isn't possible, but thanks for the offer.

It's an 11spd freehub, supplied with the 1.85mm spacer (which i shouldn't need). Got an 11-28 on it at the moment which - remedying the problem mentioned in your blog post I had with the 11-25 - no longer fouls the disc 'skin'. I think this was exasperated by the disc skin having no core, and thus running laterally out of true 1-2mm at points across the wheel.

The chain still runs extremely close to the wheel up on the 28t, might run some paste similar to engineer's blue onto an old chain and see if there's any observable contact. If so I'll try your trick of removing a sprocket.