Are there any studies or real experiences with more complex or closed fairings ?

Dreaming about velomobile, but trying to improve/inspire my up-right first by improving particular shapes without significant front area increase.

Started with jacket extended by 1m long triangular prism, seems to work at speeds around 20-35(?) somehow, but less around 50km/h downhill?? where I am trying to compare its effect during commuting.

Then played with small models in water and it seems front part of streamlined shape has some effect too(?).

But have no idea how to easily measure drag/forces or compare them.

First version

Have interesting links collection and some math in my old Q/A here How much resistance comes from each part of a bicycle? (some of maybe mentioned here too?)

Here 13.7% car drag improvement and simple way to check how does it work https://www.carscoops.com/2022/09/how-much-more-aerodynamic-can-you-make-your-car-with-a-trip-to-the-home-improvement-store/

  • What kind of bike do you have to start out with? Velomobiles are rather low recumbent trikes (or quads) There are many recumbents with half fairings, the rear half is the more common one and I bet there are test results for those. But when I see you wear a jacket extended to the rear I do think you do not have a recumbent.
    – Willeke
    Aug 14 at 8:49
  • You measure the drag forces with an aerometer(cda meter). cyclingweekly.com/news/product-news/…
    – Andy P
    Aug 14 at 11:46
  • 1
    There have been some attempts at fairing upright bikes but I haven't seen reliable measurements of the change in drag. In order to make sensible comparisons, eventually you'll want to measure drag rather than just speed. Measuring drag forces is possible, but it's not (yet) easy. See this bicycles.SE question for some approaches: bicycles.stackexchange.com/questions/9938/…
    – R. Chung
    Aug 14 at 17:19
  • 1
    Your jacket seems to have a lot of floppy areas where the skin is free to flap with the wind. This produces an enormous amount of drag! There's a reason why pro riders use skin tight clothes, only. If you want to improve over your bare skin, your fairing must have a hard surface. Aug 26 at 15:38
  • 2
    Of course shape is important. It's just that you cannot expect to see any benefit from shape if you coat that shape with flappy fabric. The drag from the flapping will outweigh any benefits from shape. And yes, accidents happen. And yes, not having the tail fixed to the bike helps reduce side kicks from the wind. I'm just saying that you want to be careful and wary of this effect. I've had many accidents, considering one accident per winter to be perfectly normal. However, those winter accidents are orders of magnitude less dangerous than any high speed accident in summer. You have been warned Aug 26 at 18:59

3 Answers 3


If you have a 50 km/h downhill section, that would be a good place to measure. Aero resistance roughly quadruples with speed. Ted Hohls comment below is better formulated. At 50 km/h you can be sure the major resistance is wind. At 25 km/h rolling resistance still plays a major part. As a % of total it gets less important as speed rises.

Simply rolling down that hill, starting from 0 speed or a fixed speed of the speedo, would give you an idea. It is important to realise that other factors like wind and temperature can mess up the results. If you measure on different dates, it's good to have 1-2 control runs with a setup you know. If setup x does 0-40 km/h and setup Y does 0-50, the later is best.

There are 3 things that factor in on overall aero resistance.

  1. Frontal surface. Less is mostly better. On your upright you are about 180 high, perhaps even 190 from the ground, and about 50 cm wide.

  2. The smoothness of the shape. Even a small edge can cause a disturbance, and vortexes. A Vortex causes more drag. As an example the difference between a modern glued in car windshield or a 1970 one with a rubber strip and chrome trim.

  3. Wetted surface. The surface of the object that is "wetted" by the air. You can get a very elongated shape, that is smoother to the air. But that same longer shape, can create more surface friction, of the air flowing past it. This works together with the previous point. A cylinder upright, has less wetted surface than an aero, symmetrical wing shaped profile around that cylinder. If you elongate that wingshape even more, eventually you will reach a point where the total drag does not go down, but starts going up.

A good place to start with aerodynamics is Julian Edgar, he had several youtube video's on car aerodynamics.

With your upright, you are roughly putting 50x180cm in the wind. That is 9000 square cm. A Snoek velomobile is 80 cm high and 70 cm wide at the bottom, but only about 30 at the headfairing. I would estimate that to have roughly half the frontal surface.

If you want to improve your upright, first look at how they improve UCI approved racing bikes. That is where the money is, and where most research is done.

Mochet experimented with tailfairings on upright racing bikes in the 1930ies, Then went for the recumbent position and eventually came to fully faired recumbents.

This is also a very good tread. https://www.bicycles.net.au/forums/viewtopic.php?t=99693

Fairing an upright is basically like fitting a spoiler on a VW bus to make it faster. It can be done, but there are more efficient vehicles to be found if you want to get the most speed out of limited horsepower.

When racing lowracers, the common experience when adding a lowracer tailfairing out of Fibreglass or Carbon Fibre, was a speed increase of about 10% at speeds over 30 km/h. Those fairings fit very nicely to the riders profile. http://www.m5-ligfietsen.nl/site/EN/Models/CrMo_Low_Racer

I am pretty sure anything shaped less perfectly, and less smooth will provide less performance gain.

If you want to improve speed on an upright bike here are my tips;

  • Drop the touring tyres and mount some fast racing tyres decrease rolling resistance from 50 to 20 watt
  • Fit an Aerobar
  • Wear a onepiece TT suit


  • 5
    You might want to clarify your quadruple statement. Aerodynamic drag quadruples with a DOUBLING of speed. Stated another way, doubling the speed increases drag by a factor of four. Another way is that aerodynamic drag is relative to the square of the speed. Good stuff though.
    – Ted Hohl
    Aug 14 at 15:16
  • UCI is very old school - it bans all best innovations like velos, etc. probably to keep fair competition. I also played with frame or wheels, but front area is probably not significant - maybe more relation or flow between particular parts, that is wind tunnel only(?) And even some cars have excellent coefficient despite they are far from streamlined, etc. But it is complex science/CFD optimisation then...
    – Tom
    Aug 14 at 19:07
  • 1
    IHPVA arranges competitions for many things that are banned from UCI events. If you don't know about IHPVA, it probably shows how well they worked out.
    – ojs
    Aug 15 at 12:25

Adding extra shapes to your body to streamline yourself can certainly help, but its also more mass to move, can catch side and buffetting winds, and could do you a lot of damage/injury in the event of an accident.

In my experience, the quickest aero gain comes from getting your head lower and your hands closer to centerline, and trying to be "small"

A well-sized backpack can help fill in the void behind you when in the upright position on your flatbar bike, but bending down mostly eliminates the area. Hunched-over road bike riders tend to put items in special jersey pockets that hang at the back , also giving a subtle aero gain.

Sadly I don't have numbers for these.

Based on your bike link you're currently riding a hard tail MTB thus:

enter image description here

The first thing to do is swap out any knobby tires for smooth tyres. But given the bike's shape, there's a lot limiting your potential improvements.

You might be best off forgoing tyres, and buy or borrow a drop-bar road bike. Even a steel 80's bike will be more aerodynamic than you current bike could ever be, regardless of how many modifications you make.

The other item you need is accurate timing - a way to measure performance changes. That might be a simple stopwatch, though these days running Strava's app on a modern smart phone will give you a heap of useful data and make it easier to see your best 10 results on a segment for free. No need for an expensive head unit or power meter.

Otherwise committing to riding on the drops will show you the difference between the two styles of bike.

Later on you can invest in a lighter road bike and aero clothes and aerobars, but at this point developing your own power output, endurance, and position are probably the keys to speed.

  • 2
    Well, on the downhill, extra mass is a bonus. "Free" extra Watts. Until you try to get back up the hill, that is. Aug 26 at 15:50
  • Probably 1st upgrade were Schwalbe and Conti Ultra Sport 23mm tyres, great resistance at 6-9bars and bike has Al frame; 13-14kg is comparable or even less than some road bikes... And do not care about top speeds, but low power requirements - commuting 15km and 190m up/down in 45-50 minutes (16-20 average).
    – Tom
    Aug 28 at 19:19
  • @Tom I suggest you find a road bike and try it out. Borrow one if you can.
    – Criggie
    Aug 28 at 19:31

Test road My downhill test show it may have tiny effect above 25km/h:

1st T shirt - quickest speed increase, then slows down

2nd this jacket (quite big size, wide sleeves), speed increase seems to be linear, but acceleration maybe slower

3rd 1/2 (only bottom hazel branches left) of back extension next day, same place

4 tests Data collected by BikeTracker apk's SQL export, time converted to seconds and speed to km/h, had to stop under hill to give way.

Mind extension was temporary test setup made from nonwoven (17g/m2, 2 layers), aprox. 1m long side after my back; not smooth, wind resistant nor perfect stable, anyway it looks like this part is not worth it...

Another day full 2 results seems to be similar, inflated, even more broken structure (wider front part), altitude difference during test by GPS aprox. 11,8 meters.

4th umbrella + jacket + new speed from distance chart

It seems to be clear proven you should not increase front area unless you create excellent shape to compensate that.

Found also interesting mail discussion www.ihpva.org ... fairings_rear.txt:

„Note that the tail fairing changes the drag by .1 no matter
what the rest of the vehicle looks like, but depending on what you compare
it to it either looks like a %10 or a %50 difference.  The upshot of it
is that you should streamline the nose first if you want really noticeable
drag improvement, streamlining the tail is only worth doing after 
everything else has been treated.  Hope this helps.  Nick.“ / Boeing Commercial Airplane

Till now best results were just forward bend without extensions.

Switching to smaller detachable front version test now...


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.