Folding Bikes - Ride quality

People may or may not be interested but in the on-line magazine I write for, AutoSpeed, I have posted two articles on measuring ride quality.

AutoSpeed is primarily an automotive technical magazine, but we have also covered the build-up of the three recumbent suspension trikes I have designed and built.

The two articles

Part 1 - http://www.autospeed.com/cms/A_111836/article.html
Part 2 - http://www.autospeed.com/cms/A_111837/article.html

describe how to use low cost software and an iPhone or iTouch to measure real-time ride accelerations. I measure all accelerations at the head.

In Part 2 there is data showing how a Brompton (equipped with sprung seat) compares in ride quality to a full-size bike and also one of my suspension recumbent trikes.

Esspecially in fore-aft pitch accelerations, the Brompton is very poor.

If you're interested in the ride quality of small wheel human-powered machines, it may make for relevant reading.

Get that man to test a Moulton!

pretty good article.

I would like to see at least one larger wheeled bike with fat, slick tires.
a hardtail MTB with fat slicks, perhaps?

Excellent comparison. Walking is at last, dead.

Someone must ship Jules a Downtube FS to compare.

For anyone who wants to cut to the chase, the following pic shows actual recorded back-to-back accelerations recorded at the head for the three pedal machines. The accelerations are: X (accelerations in roll), Y (accelerations in pitch) and Z (up/down accelerations).

Brompton = folding Brompton with sprung seat
Gazelle = full size Gazelle Medeo Plus – 28 inch wheels, sprung seat post, sprung front forks
Chalky = self-built recumbent pedal trike, 16 inch wheels, mesh seat, pneumatic / hydraulic front and rear suspension

If you get glazed eyes, the shown RMS values for X, Y and Z give a magnitude comparison between the machines that feels about right.

http://i638.photobucket.com/albums/uu102/JulianEdgar/RRGazChal.jpg

I'd like to see results for an unsuspended carbon fiber racing bike.

Interesting. If I followed the article, the measurements were from going slowly over a "bumpy grass paddock."

Off the top of my head ...

Would speed change the order of Gazelle and Chalky? I suspect that speed would affect Chalky the most since the advantage of uprights would be the body's natural suspension. It also appears that the rider might be particularly tall. So some of the results probably have little to do with the suspension, small wheels, and so on. Instead it might be more with the body's center of gravity.

Anyway, the iPhone technique is informative. From the perspective of an experiement to be repeatable and comparable, things such as tires, PSI, rider height, and rider weight (anything else?) might be worthwhile. For instance it would be interesting to see the profile of the rider on the three bikes overlaid.

Thank you for sharing.

Would be cool to see this for a range of different wheelbases, frame materials, tires, suspensions, etc. Also for different riding positions. Would allow comparison of different ways to tune the same end-feel -- and would have helped guide some choices I've made by trial and error.

Thanks

Would be cool to see this for a range of different wheelbases, frame materials, tires, suspensions, etc. Also for different riding positions. Would allow comparison of different ways to tune the same end-feel -- and would have helped guide some choices I've made by trial and error.

Thanks

I would guess that's beyond the scope of the article.
All you really need is an accelerometer that does XYZ and software that records it.
iPhone makes it easier.

Excellent article, well written. :)

I have some criticism of the article:

1. Nomenclature:
Ride quality is hard to define. It certainly is not solely determined by XYZ acceleration. Comfort might be closer to the mark when referring to acceleration. But ride quality is determined by hard to define issues contributed by the assemblage of parts on a bike. Riding position also has a huge part to play.

For example, you might ride a very soft and squishy bike with soft bouncy suspension, flexing a lot around various parts, and not experience much high frequency acceleration at all, but it would not be very nice to ride. OTOH a very stiff but light bike with careful high frequency damping in the frame would more probably contribute a lot to overall riding pleasure (and therefore quality), harshness compared to the wet noodle notwithstanding.

Efficiency has to feature somewhere in ride quality as well. A bike which saps the energy by being heavy and having poorly designed squishy suspension that absorbs pedalling effort is not a good quality ride, regardless of how well filtered any road vibrations might be. A heavy Kmart special would not feature in a list of quality bike rides for me.

So I would perhaps call it comfort but even that is not exactly it.

2. Frequency response
I could not see a reference to frequency response of suspension - please forgive me if I missed it. :) But frequency response surely is one of the main determining factors in discomfort. High frequency vibration is far more annoying than low frequency. So comparing plain amplitude does not cut it - it has to be qualified by frequency content, so an FFT is the thing to employ.

Coupled closely with this is riding speed: A suspension system even as modest as the Brompton would filter out road vibrations when riding above a given speed. For example my laden Birdy would filter out gravel road corrugations when going above 20km/h. Conversely, riding at such a speed that you manage to hit the resonant frequency of the bike would be very bad.

1-2Hz should about a cutoff frequency when considering vibrations that cause discomfort.

3. Tire pressure
It can be readily shown by some straightforward simulations that tire pressure is one of the dominant factors when considering ride comfort. A wide tire with flexible carcass such as the Schwalbe Big Apple pumped to say 40psi is stiff enough to avoid bad wallowing but soft enough to absorb the majority of high frequency road vibrations. These tyres are one of the best things money can buy when wanting to improve riding comfort - essentially no unsuspended mass, and readily tunable by selecting pressure.

Alongside tyre pressure, is carcass construction to some extent.

I didn't see any reference to tyre pressure in the testing (again please forgive me if I missed it :) ). At least, all test vehicles should have the same tyre type pumped to the same pressure.

4. Choice of test ground
By selecting a paddock, you may have inadvertently chosen the very worst ground for the Brompton wheel size. If a depression matches the wheel diameter closely, that wheel would respond the worst way when rolling through it. This would have at least two effects:

If you consider a pothole of the same size and shape of the wheel, then a wheel rolling through it would roll over the edge while descending into the hole, then the wheel would hit the hole at all points of the touching tyre simultaneously. Because the support is now no longer an oval patch but rather a long line, the equivalent tyre pressure is magnified many times. The resultant high frequency vibration transmission would be many times worse.

The second effect is that the vertical descent followed by ascent is worst case: there as a shard direction change as the wheel hits bottom. With either a smaller or a bigger wheel, this would be less severe. You could argue that a tiny wheel would actually be better off, rolling smoothly over whatever profile there is. (Just avoid any sharp rises!)

5. Choice of accelerometer location
This a bit subjective but it feels worse to me if the body is jolting around. The head is quite removed by its own suspension system. Seems to be the saddle would perhaps be a better choice followed by the handlebars. The saddle position would be greatly influence by riding style as well: a rider automatically taking more weight on the legs when handling rough ground would remove inertia from the saddle, allowing the seatpost and saddle rails to flex less.

So you see there are so many factors at play here that the choice of test would make or break any particular case.

Nevertheless, this article is an excellent starting point for more study.

For more information, see for example here
http://lustaufzukunft.de/pivit/fahrrad/my_bike_engl.html

As a retired salesman of high tech communication products, I like the colors on the charts!!! :D

FYI: This much data would totally confuse the average buyer.

Ride quality is hard to define

Sure, but I am not referring to ride quality as in the bicycle vernacular "it's a good ride". I am referring to it specifically within a vehicular context, ie ride and handling. In terms of this definition of ride quality, that is, the degree of comfort felt when sitting there, measuring magnitudes of acceleration is a very effective way of assessing ride quality. It's certainly not the only way, but it does provide good agreement with subjective discomfort levels. In Part 1 of the series I cite an engineering paper where X, Y and Z accelerations are the primary way in which the authors compare ride quality of different vehicles.


Frequency response
I could not see a reference to frequency response of suspension - please forgive me if I missed it.

The frequency response is primarily determined by the resonant (natural) frequency of the suspension systems. This can be directly calculated from static suspension deflection, ie how much the suspension squishes when the rider gets on. The greater the static deflection, the lower the resonant frequency. (see http://www.autospeed.com/cms/A_108167/article.html for more on this.) In the case of variable rate springs, it gets more complex but the basic idea remains valid: the softer the suspension, the lower the resonant frequency (and therefore the less vertical accelerations that get passed to the rider).

In addition, Part 2 of the series list the frequencies at which most vertical accelerations were measured as occurring (the software does this via a FFT I think?). The Brompton had the greatest proportion of vertical accelerations occurring at 5Hz, the Gazelle at 4.8Hz and Chalky at 2.6Hz. Again, you can see how bad the Brompton is.


Conversely, riding at such a speed that you manage to hit the resonant frequency of the bike would be very bad.

Actually, the outcome depends very much on the suspension damping - you want max damping at the natural frequency of the suspension, because that's when it is most easily excited.

(Incidentally, the reason the front suspension of the Birdy does so well when it has only limited suspension travel is I think because the wheel moves backwards as it rises, so lessening impacts and harshness.)


3. Tire pressure

I typically run 50 psi.


4. Choice of test ground
By selecting a paddock, you may have inadvertently chosen the very worst ground for the Brompton wheel size. If a depression matches the wheel diameter closely, that wheel would respond the worst way when rolling through it

Well, it would have been even worse for the trike, wouldn't it? Same size wheels (16 inch) but 50 per cent more wheels to be hitting bumps! And think of the impact of a trike wheel configuration on roll accelerations, which you don't get at all on two wheel machines when you hit bumps. By any measure, the trike was potentially the most disadvantaged.


Choice of accelerometer location
This a bit subjective but it feels worse to me if the body is jolting around. The head is quite removed by its own suspension system.

The head was picked because it is the most telling. The head experiences pretty well the same vertical accelerations as the body but will emphasise roll and pitch accelerations, as it is effectively a mass suspended on a springy pole. Again, the previously cited engineering paper used a head-mounted accelerometer.


For more information, see for example here
http://lustaufzukunft.de/pivit/fahrr...bike_engl.html (http://go.bikeforums.net/?id=42X129555&xs=1&url=http%3A%2F%2Flustaufzukunft.de%2Fpivit%2Ffahrrad%2Fmy_bike_engl.html&sref=http%3A%2F%2Fwww.bikeforums.net%2Fshowthread.php%3F653291-Ride-quality)

Yes I have seen the link before. It looks an interesting design but without any test data, it makes it hard to assess how well it works.


So you see there are so many factors at play here that the choice of test would make or break any particular case.


I don't agree. The relativity in ride comfort between the three machines in this test is pretty typical of what I have found riding each in many different conditions. I cannot think of any riding situation where the ride comfort (as defined in the first part of this post) of the Brompton would be remotely as good as the air / hydraulic suspended trike. (Edit: except on dead-smooth concrete or bitumen without any joins; there, suspension makes no difference at all to ride quality!)