Are FLO wheels really this slow...??

[ed72]

An Aerospace engineer named Hambini with access to specialized wind tunnel facilities that can create turbulent effects has created a protocol ostensibly representing real world wind flows. The results shockingly suggest FLO cycling wheels are slow in real world conditions where wind is much more turbulent with many transient conditions. He is showing 35 watts disadvantage for the FLO at 50 km/h. Yikes!!

I found the study interesting for all the angst it has created on other forums but it seems to answer why my FLO60 carbon wheeled bike has had trouble in cross winds compared to a Zipp 808 wheeled bike, on a relative basis. I also have found the transient behavior at the higher speeds of a recumbent to be quite surprising compared to slower uprights. Maybe I just need new wheels?

Curious, I purchased a 2-spoke wheel and compared using the Chung field testing protocol. I compared it to my FLO60 using a Continental 23 mm Supersonic tire with latex tube on both. The 2-spoke was a little faster (5-7 watts) but there was next to no wind and there was no traffic or anything to disrupt flow because I use a half pipe in an essentially deserted neighborhood. The FLO60 with the Conti usually tests at the top of the class in static conditions, so, not much to see here although 5-7 watts isn't bad.

The crosswind stability is different between the FLO60 and 2-spoke and hard to explain.

I spend a lot time riding along a relatively flat river road. 10 miles up and 10 miles back. Scores or probably hundreds of runs. I had a modest 90 degree cross wind yesterday using the 2-Spoke front and covered wheel in the rear (as usual). There was vehicle traffic. Trees. Roadside cliffs funneling the wind. Real world stuff. My power/speed numbers in both directions were the lowest I have ever recorded by far, to date. Improvements in line with the linked study. I was just in my regular kit.

I suspect Hambini is onto something. I thought I would share this with the racers here. YMMV.

https://www.hambini.com/blog/post/bicycle-wheel-aerodynamics-which-one-is-fastest/

 

[Lief]

Half way through the article and I like the way this is going. Feels like real experience and knowledge being applied to a complex problem with fresh perspectives / relatively little cultural baggage.
Thanks @ed72!

 

[ed72]

You're welcome, Lief.

One of curiosities that I have been chasing on a recumbent is the apparent differences in CdA from static testing to on the road testing and variation in general. If I run 10 miles up the river on say 140 watts at say 20 mph and then 140 watts down the river at 20.5 mph, I can correct for atmospherics (temperature, pressure, and RH) and 0.02% gradient and come up with an estimated CdA. This CdA varies from day to day; however, field testing using the Chung method on a calm day is very consistent. I think it was Larry who posted a lot of CdA data and there was variation during the day, I see similar "unexplained" behaviour but maybe not quite as much. I dunno.

I thought I might be "wrestling" the bike or changing positions, so, I focused strictly on good form but was still seeing 5-10% differences from day to day (even correcting for the atmospheric) using the same gear and kit with same tires, wheels, air pressure.

What else varies in the real world? Traffic and wind direction.

Hambini performed 5 replicates and averaged for each position. The only question to me is whether air is really that turbulent and whether his yaw distribution represents real world.

What I have noticed since taking up recumbents is how wind affects me much more than on an upright but I am going much, much faster compared to my days on an upright. I am generally going 31-32 mph on the flats while time trialing on a bent and would only be doing around 26 mph on an upright. When a pickup truck passes close doing 55-65 mph, it gets my attention on a bent if going fast whereas on an upright, it is generally not noticeable. If I have too large a front tire, the effect is down right scary and it is a bit of a snap in the bars. (The transition is very abrupt as the wheel stalls as the airflow separates)

I was going to spring for the uber expensive Zipp 858 NSW but there is not enough reports or reviews to support the alleged crosswind improvements and I don't like their warranty service when their stuff inevitably breaks, which is why I have been a FLO fanboy for many years-their wheels cannot be destroyed, I have tried!!

 

[Balor]

What I have noticed since taking up recumbents is how wind affects me much more than on an upright but I am going much, much faster compared to my days on an upright. I am generally going 31-32 mph on the flats while time trialing on a bent and would only be doing around 26 mph on an upright. When a pickup truck passes close doing 55-65 mph, it gets my attention on a bent if going fast whereas on an upright, it is generally not noticeable. If I have too large a front tire, the effect is down right scary and it is a bit of a snap in the bars. (The transition is very abrupt as the wheel stalls as the airflow separates)

Well, it has little to do with actual aerodynamics... ok, your frame is bigger, hence your lateral A is a bit larger - but I presume effect is nearly negligible.

Next, your CoG is much closer to your later CoP (given disk/high profile wheels) due to you being nearly horizontal - hence you must do much more leaning (with accompanying steering to accomplish it) to compensate.
Plus, high-profile rims or disk upfront require a zero offset at the very least not to give you 'negative aerodynamic "trail" force'.
At higher speeds this effect is much more apparent because this 'windvane effect' increases cubically like all aerodynamic effects, while actual trail force increasing linearly.

http://recumbents.com/wisil/brown/airdragformula.htm

Negatively offset wheel will actually provide extra stability at higher speeds not unlike extra trail - and ONLY at high speeds where you need it most. And it is much easier to have on a bent.
And the last, but not the least - short bars give much MUCH less leverage to exert control over your steering. John Brauchler always runs dual disk on his (conventionally offset) MBB because he feels wide bars gives him enough leverage for adequate control in all conditions.

 

[ed72]

Well, it has little to do with actual aerodynamics... ok, your frame is bigger, hence your lateral A is a bit larger - but I presume effect is nearly negligible.

Next, your CoG is much closer to your later CoP (given disk/high profile wheels) due to you being nearly horizontal - hence you must do much more leaning (with accompanying steering to accomplish it) to compensate.
Plus, high-profile rims or disk upfront require a zero offset at the very least not to give you 'negative aerodynamic "trail" force'.
At higher speeds this effect is much more apparent because this 'windvane effect' increases cubically like all aerodynamic effects, while actual trail force increasing linearly.

http://recumbents.com/wisil/brown/airdragformula.htm

Negatively offset wheel will actually provide extra stability at higher speeds not unlike extra trail - and ONLY at high speeds where you need it most. And it is much easier to have on a bent.
And the last, but not the least - short bars give much MUCH less leverage to exert control over your steering. John Brauchler always runs dual disk on his (conventionally offset) MBB because he feels wide bars gives him enough leverage for adequate control in all conditions.

Kyle says, "People generally avoid using disk or covered wheels on the front, as they make the bike hard to control in a crosswind. However, as this varies not only with wheel size but with distance from the center of pressure to the steering axis, so a 406mm disk wheel would be vect affected only about 1/3 as much as a 700. Sensitivity to cross winds depends strongly on steering angle"

I am not following you entirely. I believe the torque I feel in the bars with gusts or with the pickup truck example is precisely due to aerodynamics of the front wheel (I get that bents are harder to ride). The 2-spoke, FLO60, Zipp404, and FLO30 each behave differently in wind (also different tires on the last two). There is certainly a lateral force component that one must lean and/or steer into with a side wind. What Hambini says is these ground winds are highly turbulent and what I am suggesting is in addition to the obvious changes in side forces, the wheel stalls (sail effect lost) because the air over the rim has separated and this abruptness causes a jerk in the handlebars. If you read what little Zipp publishes, they try to have the pressure vortex off the leading edges to be centered with the steering axis in order to minimize this effect. I can't remember if Bram Moens wrote it in an e-mail to me when inquiring about the 2-spoke or if I read it online; however, two and four spoke NACA designed foil wheels balance the wind turbulences and the vortex does not shift abruptly. He and others have said how stable the 2-spoke is in wind. I have only ridden in mild wind so far. I remember my old tri Spokes (Specialized who then sold to HED) were terrible in the wind. Maybe I am only feeling the movement of air on the broad side profile of the bent and nothing from the 2-spoke but for sure, there is a stability difference due to just the wheels. I already know the tailbox is not good in terms of aerodynamics in crosswinds (bluff body, slow) and for sure, lower body position one has on a bent makes balance harder compared to an upright.....just basic physics.

Wheels and tire interface are just one of those marginal, transitory thingies. Who cares about a meager 35 watts lost.

Your design is too hard to wrap my brain around at the moment. I did study advanced control systems. I have tried to accept and understand various transfer functions written up describing bicycle handling behaviour. The low angel front end geometry and relatively flexible fork combined with my chubbiness could also be a factor in the dynamics. In the end, I'll just go with Einstein.

“Life is like riding a bicycle. To keep your balance you must keep moving.” – Albert Einstein

One of the things I learned to do and could be contrary to the above Einstein law is to stop pedalling and brace. So, lets say I am doing 35 mph and a huge pickup is coming up behind me at 60 mph, more often than not, I am going to be hit with a large pressure wave that requires correction. At 20 mph, not a problem. The apparent wind angle is a factor, I just never know how much of a veer is coming. I have the perfect venue for side to side testing of the wheels, I just need a day with a strong crosswind. The road is downhill (easy to make speed) and goes from shielded by forest to partially shielded with small interspersed pine trees to partially blocked by houses to open field. I always have to correct when coming down this hill when windy. Maybe I should just "Damn the torpedos, full speed ahead"......

 

[ed72]

This shows the abrupt transition that I feel.....

http://biketechreview.com/index.php/blog/519-overtaken-by-a-car

Page 23 explains that anecdotally, cyclists think aerodynamic torgue on the front wheel is very small but this study shows otherwise.

http://www.ilight.com/wp-content/uploads/2012/08/AIAA2011_1237.pdf

Some may wonder why this is important. Getting up hills is hard even little ones, braking or even just sitting up on the descent kills momentum. More practically, I personally would like to be able to bomb down hills with the same impunity on my bent as I used to do on my uprights. So, it is a quest for understanding and optimization.

 

[Balor]

I think we are talking about different things here. I'm talking more about *stability* than that aerodynamics per se... 35 watts is definetely not something you want to lose, but I cannot afford a 1000$ wheel either, 20$ monokote works for me .

As for steering angle - the article is indeed vague here. It is not the angle that is important, but resulting positive wheel offset you need to prevent from having way too much wheel flop.
Let's assume 55 deg steering, 20" racing wheel with full faring and low-trail geometry people often use on LWBs.

You need about 130! mm of positive offset to keep trail and flop to manageable levels (22mm trail, 10mm flop).

That way, you don't *just* get CoP massively shifted forward, but the wheel acts kind of like a windwane and wants to turn 180 degress to align with air flow at the slightest opportunity - that's what I mean by 'negative aerodynamic trail'.
In a way, this effect is already present on MBB bikes due to boom being, well, 'positively offset' from steering axis, but flat plate drag of disk wheel (laterally) is MUCH higher than that your legs anyway - though you might want to think hard before installing huge TT chainrings with no holes on MBB... but your control over the bars on MBB (given 'Cruzbike cockpit' or wide superman bars, not gunners) is much greater than on most recumbents anyway and while you cannot have too much 'aero', you totally can have too much control if it hinders your performance in other areas - you need just *enough*.

 

[Balor]

More practically, I personally would like to be able to bomb down hills with the same impunity on my bent as I used to do on my uprights.

Fortunately, I have solved that problem for myself Maybe not as good with high-speed cornering but I'm working on that as well. I'm very seriously considering building a 'streetliner' around my MBB chassis, actually, I'm feeling *that* confident:

What is unfortunate there is simply no hills large enough to try, say, 80+ kmh speeds, I can barely reach 70.

Ed, do you think it is even worth trying to fair (Monokote, like I said) rims with about 25mm outside diameter given 40mm wide tires?

 

[ed72]

Ok, I get it. I was only interested in transient steering behaviour due to aerodynamic perturbations.

Yes, I think fair would help but I do not know. You have a very complex solution there.

I like the bungee cords. One of my fears about MBB on long brevets is falling asleep. I have never fallen asleep riding but many riders do. This why I like slacker head angles and a lot of trail. I like self correcting high speed stability. I want to plow thru potholes at 50 mph at oh dark 30. Not a fan of planing, thin tubes, and low trail thoughts out of Seattle.

 

[ed72]

Sorry for off topic question but does the rear shock compress under pedal power? I would think not. If it does, how much is lost inefficiency. On bumpy roads it has to help Crr a lot. Have you ever measured? My roads are terrible and the vibrations kill my neck, so, I do most of my riding on newly paved road. I considered dampeners to mount the seat (the kind used to mount pumps and so forth) but someone convinced me that too much power would be absorbed but in retrospect, I am not so sure. During steady state riding or cruising along at say 22 mph and 120 watts, I can't imagine any changes in the rubber mounts compression. Bumps would compress and restore energy. It is not like rubber mounts would heat up but I am making Jan's planing argument for him.

I never really tried to quantify Crr on real roads until recently. I am surprised how much higher my Crr numbers are compared to the steel drum figures posted on the internet. I am not confident in my measuring techniques to say what they are other than much higher than the 0.003 type figures one reads about.

 

[Balor]

I like the bungee cords. One of my fears about MBB on long brevets is falling asleep. I have never fallen asleep riding but many riders do. This why I like slacker head angles and a lot of trail. I like self correcting high speed stability. I want to plow thru potholes at 50 mph at oh dark 30.

*Scratches head* I really don't think I come up with a design that will slow safely tuck you to sleep if you fall asleep riding You need something like Honda assist for that - way beyond "homebuilder" level.

As for "plowing throw potholes" - that is something you need suspencion for.

People been monting their seats on vibration mounts and reported good results, but I suspect wider tires are much better at it. Will experiment with that myself eventually.

If you use really thick and soft ones it sure will interfere with proper bracing AND make you seat rock side to side a bit.
Brown mentioned that it degrades steering, and he has 40 or so years of building recumbents...
(He also mentioned how flexible seat made him unable to climb hills, and I tend to believe him).

Shock bob - that's a pretty complex subject. On non-driven wheel only weight transfer affects it, and on a bent with long wheelbase and low CG (especially compared to a rider standing on pedals) it is nearly an order of magnitude less, so just a bit of platform damping (or simple stiction) should be enough to eliminate it entirely.
Driven wheel is much more complex - you need to consider chain forces and tire contact patch forces, that ideally require solving a freebody diagram and I'm not the person to ask - I think Geoff Bird is a better choice
In my case I just got massively lucky to be fair, it does not bob except a little bit in small ring uphill.

 

[ed72]

*Scratches head* I really don't think I come up with a design that will slow safely tuck you to sleep if you fall asleep riding You need something like Honda assist for that - way beyond "homebuilder" level.

As for "plowing throw potholes" - that is something you need suspencion for.

People been monting their seats on vibration mounts and reported good results, but I suspect wider tires are much better at it. Will experiment with that myself eventually.

If you use really thick and soft ones it sure will interfere with proper bracing AND make you seat rock side to side a bit.
Brown mentioned that it degrades steering, and he has 40 or so years of building recumbents...
(He also mentioned how flexible seat made him unable to climb hills, and I tend to believe him).

Shock bob - that's a pretty complex subject. On non-driven wheel only weight transfer affects it, and on a bent with long wheelbase and low CG (especially compared to a rider standing on pedals) it is nearly an order of magnitude less, so just a bit of platform damping (or simple stiction) should be enough to eliminate it entirely.
Driven wheel is much more complex - you need to consider chain forces and tire contact patch forces, that ideally require solving a freebody diagram and I'm not the person to ask - I think Geoff Bird is a better choice
In my case I just got massively lucky to be fair, it does not bob except a little bit in small ring uphill.

It is all about marginal gains.

I bet the cords would help riding no hands with only one eye closed.

 

[benphyr]

Back to the original discussion, the bigger question is What happens when real world interferes with steering and what can we do about it? Let's chop the question up into as many variables as I can think of:

Overall environmental :
-wind direction,
-wind speed,
-turbulence,
-trees, open fields, etc. effect on wind direction and speed,

Specific aerodynamic effects of:
-tire,
-wheel,
-pedaling,

Passing vehicle:
-pressure wave (and how it changes as vehicle passes, ie. beginning and ending effects) - I don't have a clue how you would separate this from the following,
-affect on wind direction,
-affect on wind speed,
-affect on turbulence,
-ground wave (when a heavy vehicle passes it creates a wave in the ground

steering input and reactions
-self correct - gyroscopic interaction,
-offset/trail,

This specific question is only looking at the difference between different wheels. With all of these variables it is going to be very, very difficult to definitively make any conclusions. But the question is certainly worthy of exploration, because maintaining control is of high importance. More power to you. Even subjective experience can help so keep the info coming.

 

[Balor]

-ground wave (when a heavy vehicle passes it creates a wave in the ground

That's very interesting, can you please elaborate?

 

[tiltmaniac]

It creates a wave in the air near the ground, for sure.

 

[Balor]

-self correct - gyroscopic interaction,

Gyroscopic effects, unless your rims are made of lead or something are perhaps the least important in bicycle dynamics if my data is of any indication.

Trail, camber thrust, camber torque and weight distribution are much more important for self-stability.
Studies by Delft show that you can manipulate any of those to achieve self-stability *without the rider*.
http://bicycle.tudelft.nl/stablebicycle/StableBicyclev34Revised.pdf

It would be very interesting if a design can be had that achieve total self-stability with rider (as in - with rider passively riding the bike...), at least at certain speeds - so you, indeed, can have your bike that will safely slow down and THAN dump you over.
On the other hand, it may actually be more dangerous because self-stability does not equal lane control and simply promptly falling to the side might be preferable to veering into oncoming traffic or hitting a light post at full speed.

Plus, since your legs are connected to the steering on MBB, I find it REALLY unlikely. Once you slump in your seat, you will inevitably disturb equillibrum, unless you add so much bungees that you'll find usual steering next to impossible I guess .

 

[Balor]

It creates a wave in the air near the ground, for sure.

Makes sense I guess. So, a fully faired lowracer is NOT an optimum solution when it comes to dealing with 'truck suck'?

 

[tiltmaniac]

It is scary enough without a big horizontal sail.

 

[tiltmaniac]

Oh, and https://nullwinds.com/pages/spoke-fins

I actually like the concept as having the aspect ratio change as a result of lateral forces makes a lot of sense...

 

[benphyr]

Oh, and https://nullwinds.com/pages/spoke-fins

@tiltmaniac
It sounds good but I wonder if it might be at least in part a unicorn. Many of the claims sound logical but the logic being used and the testing being used have at least some flaws that I see from a scientific perspective. For example, if the spoke fins work so well against head wind and cross wind then the test should show it pulling away and continuing ahead. If anything, when slowing it would continue to pull away because the extra mass of the spoke fins would have the wheel carrying more momentum. They don't address flutter of the spoke fins which I am sure would loose plenty of the advantage when changing direction - nothing has no friction...

Healthy doubt. I'd like to try them, the idea is highly intriguing but requires some rigorous testing it seems to me.