List of recent 'innovations' of the cycling industry

[MartinC]

https://road.cc/content/tech-news/ratios-conversion-kit-adds-shift-1x11-sram-groupsets-278169
hmmmm

Yes, the last time I really needed and extra sprocket it was between the 17 and 20 on a 9 speed cassette. Apart, of course, from the 11.5 and 12.5 sprockets we really need to make modern road cassettes useful.

[fastpedaller]

I can't see why there's so much dislike of ahead type headsets - I had a (very cheap) Halfords MTB with one for 8 years hard use on road/little off road(and then I gave away the bike but kept the headset as a spare). The only time I've had an issue was with the Spa Tourer supplier with a Tange headset with upper roller bearing which I just couldn't adjust out the drag without ending up with play - a cheap Ebay replacement with clipped balls binned and replaced with grade 10s has been in place trouble-free for the last 6 years - I also have several spacers below the stem.

[Brucey]

...I dislike threadless headsets. I just don't see how you can possibly, apart from by chance, adjust them so that there is no preload and no play....

I'm no big fan of them either. Maybe I'm fussy about headsets but it is perhaps telling that I have changed out more Ahead headsets on my own bikes than any other 'non consumable' part ( so chains, tyres and sprockets excepted). Which to me tells its own story; whilst there are good Ahead headset designs out there, they are not often fitted to bikes as standard and/or they are easily damaged.

I also endorse the view that in threaded headsets Campag and Tange (the latter filled with loose balls) are good ones, along with a few others of similar design (which includes Hatta and old (all steel, ball bearing) stronglight models) .

However I would note that (AFAICT) most threadless headsets are meant to be run with a little preload. The problem is that accurately controlling that preload is something that is difficult to do; an M6 bolt is capable of exerting about x30 times the required load, and the required torque on the bolt (which ought to be about 0.5Nm tops) is insufficient to cause any movement of a wedge ring that has decided to bind....which most of them do... . And then the wedge ring settles in service, leading to free play in the bearings (again). The usual solution is to apply an excessive torque to the preload cap and hope that the extravagant preload is dissipated as the wedge ring settles in service. I agree that it is somewhat hit and miss.

cheers

[iandusud]

...I dislike threadless headsets. I just don't see how you can possibly, apart from by chance, adjust them so that there is no preload and no play....

However I would note that (AFAICT) most threadless headsets are meant to be run with a little preload. The problem is that accurately controlling that preload is something that is difficult to do; an M6 bolt is capable of exerting about x30 times the required load, and the required torque on the bolt (which ought to be about 0.5Nm tops) is insufficient to cause any movement of a wedge ring that has decided to bind....which most of them do... . And then the wedge ring settles in service, leading to free play in the bearings (again). The usual solution is to apply an excessive torque to the preload cap and hope that the extravagant preload is dissipated as the wedge ring settles in service. I agree that it is somewhat hit and miss.

cheers

You are of course correct as It is not possible to adjust a threadless headset to not have play without some (too much?) preload. But that is exactly why I don't like them. It just seems wrong to me, as I know from experience that a loaded headset can affect steering adversely and will inevitably accelerate wear. Out of interest which threadless headsets do you rate, as we are condemned to use them on modern bikes.

[Brucey]

Chris King is good (it should be at the money) so is Hope (but I've not used them much myself) and at more reasonable cost I have had good results with the (now uncommon but briefly popular) 'semi-cartridge' designs, in which the bearings look and fit like cartridge bearings, but you can strip them down to overhaul them and they are filled with much larger ball bearings than you might otherwise find. For example Cane Creek C2 in 1-1/8"; IIRC these have 3/16" balls, full complement. These headsets will clap out quickly if used "as is" (out of the box) because the grease is rubbish, but if they are packed with decent grease they last OK. Tange used to make headsets this way too, but I don't know if they currently do or not.

True cartridge and semi-cartridge headsets need some preload if excessive movement between the bearing units and their angled seatings is to be avoided.
However many 'true cartridge' bearing headsets contain too few balls, too few in number. With these you are hovering on a tightrope, preload wise, between too much and too little. For example many 1-1/8" cartridges contain about 20-odd 1/8" balls; (this sort of size and quantity is considered adequate for kid's bikes normally... ) even though there would be room for many more balls in the raceways if they were designed to be assembled with a full complement instead.

Headsets with a lower roller bearing in theory have a greater tolerance for preload variations. However in practice the angle of the rollers is usually wrong (which means the seals are extremely unlikely to work) and the corrosion which accompanies the slightest ingress of water has to be seen to be believed. So I'm not a fan of those; to cap it all these headsets tend to be slightly draggy even when they are set up correctly.

Right now I struggle to think of a headset I really like that is currently for sale and is reasonable money. 'Choose your poison' time again.... On bikes with mudguards, there is something to be said for using a really cheap headset and doing what we always used to do, which is to assemble it with loose balls, rather than use the clipped balls which are again usually too few in number. On bikes without mudguards you need decent seals on the bottom of the lower race and that seems to be the province of cartridge bearing headsets. If you do end up with these, you can at least choose one which uses a readily available bearing type. If corrosion is the enemy (and it often is, many designs allow water in at the top eg through the wedge ring... ) then using stainless steel cartridge bearings is often better than not, despite their having a lower static load rating in non-corrosive conditions.

Gah!

cheers

[iandusud]

Brucey, I suspected you were going to suggest Chris King or Hope! Totally agree about the use of caged bearings. I always replace with loose balls. Interesting what you say about taper bearings. I always thought the old threaded Stronglight taper bearing headsets looked good but in practice I found them to useless.

Ian

[cycle tramp]

One of the things which folk don't seem to give enough emphasis to is how the steering feels. This is important when you are pushing a motorcycle to its limits and many alternative steering systems are characterised by relatively poor steering feel; if there are benefits to be had, but they come with poor feel, it might be a worse compromise overall.

cheers

Apologies for the length of time in my reply - Yes, you are right, above all, the user of any machine must feel confident using it and would use any expectations of 'feed back' from the vehicle to form part of their confidence.
It was historically noted that when the first safety bicycle was made available many ordinary (penny farthing ) riders did not adopt it, citing that it felt 'wrong' and that they couldn't see the views... certainly when I used a two wheel hpv I found that I had no feed back from the front wheel, and that rather than use my body weight to change direction I had to constantly correct my steering using the handlebars. Something which required more active thought rather than my usual unconscious shift in body weight.. I suspect the quasar may have felt similar to those motorcyclists more used to an upright position.
That is not to say that either the quasar or the hpv was inferior- far from it, it was just that I felt less comfortable using it. No doubt if I had put some serious miles under the wheels of the hpv I would have acclimatised to the feeling, possibly to the point where if I returned to riding a safety bicycle I may have felt that I was 'up too high' and that 'it was a long way to fall to the ground'. I may have also grown to dislike the sense of un-steady ness of the safety bicycle when descending.

[Brucey]

…. I always thought the old threaded Stronglight taper bearing headsets looked good but in practice I found them to useless.

You only need to look at a taper-roller bearing motorcycle headset to realise how it should be done. IIRC there has only been one bicycle headset done that way - Galli I think - with rollers which are nearer to vertical than horizontal, and that (unsurprisingly) had a larger than normal stack height. I have never used one so I don't know that there isn't another problem with them, but at least they got the angle of the rollers closer to something which is sensible.





cheers

[NATURAL ANKLING]

Hi,
Just thinking out load, what about a taper at the bottom and a radial at the top?

[Brucey]

…. above all, the user of any machine must feel confident using it and would use any expectations of 'feed back' from the vehicle to form part of their confidence....

yep, that is a key ingredient alright, and if years of practice have instilled a certain way of doing things into a rider, it can take a long time to develop new skills and new habits appropriate to any new setup.

But nonetheless it is highly instructive to look at how riders have got on with hub centre steering when they have given it every chance. Alan Cathcart (noted one-time racer and motorcycle journalist) has been involved in the Bimota Tesi project for many years, both in development/testing and (I believe) in racing. Even he - arguably one of the most experienced riders in the world, e.g. the only man who has been allowed to ride factory GP bikes of all makes since some time in the 1980s- couldn't get away from the steering feel being different (worse) and during the period when Bimota looked to have shut up shop for good, he commented that 'there were just too many joints in the steering' in that design. IIRC he had some high speed 'offs' where he lost the front end in a highly uncharacteristic fashion. So have Bimota's other test riders too. Now (this year) Bimota have been bought (well 49% anyway in 2019) by Kawasaki and a new Tesi has been unveiled, more expensive than ever, with Kawasaki power.

If Kawasaki don't soon take the opportunity to learn from decades of Tesi development and launch their own (less expensive) bike with hub centre steering, this may tell you all you need to know. After all Bimota's track record prior to the Tesi was for quite a few years to build machines which sold for double the money of Japanese bikes (with the same engines) but out-handled them courtesy of a stiffer/better/better feeling chassis; plenty of race wins backed this up. Their advantage largely disappeared when the approach to chassis design was emulated by the big four. Bimota then decided to take 'the next step' being hub centre steering because there is something to be gained in theory. In practice however that has never been fully realised in terms of either race wins or uptake in the market. To me this says that the primary goals might have been achieved but secondary issues are probably preventing the full benefits from being realised. They have spent at least thirty years trying to iron out the bugs and whilst this amount of effort doesn't guarantee success where it might be possible, it is highly indicative that there are problems that cannot easily be solved.

So for now, telescopic forks are -in reality- still the most practical compromise.

Philosophically teles are a bit like the rear-engined Porsche; a pretty stupid way to design a car but by golly they stuck at it and made it work anyway....

cheers

[speedsixdave]

Hi,
Alternatives to telescopic are heavy, and unsprung weight affects handling.
If not then lateral stiffness, too many moving parts adds to complexity and cost.........then there's maintenance.....
The former may be a compromise but works well..........enough....

Apologies, Moultons again...

Neither of the current production Moulton front suspensions are particularly heavy or complicated. The 'AM' leading link fork - in production virtually unchanged since 1983 - has very little unsprung mass, just a slender stirrup holding the front wheel, and four coulomb-damped pivots and a steel spring, so not very complex. It's completely maintenance free on a day-to-day basis, and mostly maintenance-free in the longer term. It is not without its issues, including diving under braking, and is really fairly primitive, but is a pretty good compromise for a practical bicycle.

The Flexitor front suspension, in production on the New Series since 1998, is amazing, reacting rapidly to bumps tiny and large with aplomb. Again very little unsprung mass but this time eight rubber pivots providing the damping. Completely maintenance free, and anti-dive. It's ridiculously good but also ridiculously expensive, which is a shame.

But one could happily argue that both these 'recent innovations' are rendered redundant by modern high-volume supple tyres, and there's a degree of truth in this.

[Brucey]

the main reason why the flexitor design didn't appear before it did in a Moulton bicycle was because Dr Moulton had sold an exclusive license to a car maker. He had to wait for this to expire before he could use his own invention in his bicycles.

It is very clever of course but it doesn't offer truly stable performance over a wide range of temperatures (but then neither does the power unit, ahem...) and it has all the other qualities (good and bad) of girder forks. You will notice that there isn't exactly a rush to use a similar approach on motorcycles and this is much of the reason why. As far as cost goes the flexitor units ought to be a way of making the whole thing cheaper, not more expensive, hence the interest from car makers in the early days.

cheers

[speedsixdave]

the main reason why the flexitor design didn't appear before it did in a Moulton bicycle was because Dr Moulton had sold an exclusive license to a car maker. He had to wait for this to expire before he could use his own invention in his bicycles.

It is very clever of course but it doesn't offer truly stable performance over a wide range of temperatures (but then neither does the power unit, ahem...) and it has all the other qualities (good and bad) of girder forks. You will notice that there isn't exactly a rush to use a similar approach on motorcycles and this is much of the reason why. As far as cost goes the flexitor units ought to be a way of making the whole thing cheaper, not more expensive, hence the interest from car makers in the early days.

cheers

Indeed, AFAIK the flexitor was invented in 1948 and then appeared on the New Series bikes in 1998, a convenient 50 years later...

I know nothing about motorcycles but I guess there's a difference between a bicycle's couple of kilogrammes of unsprung front wheel mass and whatever a motorbike's front wheel assembly weighs. I've only ridden the flexitor forks a couple of times but what's hugely impressive is the lack of 'stiction' and the fork's ability to track really small bumps really rapidly - it's pretty much constantly in motion on a British road - while the rising spring rate (? terminology) preserves the ability to handle bigger bumps. It's much more responsive than any other suspension fork I've ridden for certain.

Of course the damn thing still bobs up and down when you honk, which is one thing motorcycles don't have to deal with. There is a soft lockout which works quite well but it's actuation could be significantly improved. You're right about the cost, of course. and I suspect it is a commercial decision to only offer it on the most expensive bicycles. And I'm sure you are right about the temperature issues, although how much of a difference this really makes across a normal working range of say -5 to +40C I don't know.

[iandusud]

This is an interesting thread for me as a lifelong motorcyclist and cyclist, who regularly rides an AM Moutlon (and other more conventional bikes) and who appreciates good engineering practice. I am very aware of the compromises of the various front suspension systems mentioned. I have followed the Bimota Tesi since the 70's when I started motorcycling and of course John Britten's wonderful bike development. However if there's one thing that life has taught me it is that a pragmatic approach is often the best one, which generally means compromising one thing or another. When telescopic forks started to replace girders on motorcycles one of the earliest adopters was AMC who produced 350s with "teledraulic" forks for the war dept in WW2. The despatch riders loved in favour of the girder forked 500s of BSA and Norton. The rest, as they say is history. I have a couple of 1950s motorcycles, a Matchless with the teledraulics and a Velocette with their own excellent tele fork. I would however love to have a girder forked machine having ridden a few pre-war girder forked bikes and appreciated they way they seem to float over the bumps. But the proof of the pudding...

As regards the AM front suspension I have not ridden a NS bike to can't comment on it's ride qualities, although I don't doubt that it is very good. As for the original AM front fork the tendency of those who are unfamiliar with them is to set the spring hard (presumably trying to emulate the feel of a rigid fork ) and, hopefully adjust the damping accordingly so as avoid bouncing all over the place. However I soon learned that to get the best out of it it's better to run the spring as soft as you feel you can get away with, again adjusting the damping accordingly to reduce stiction to a minimum, and to ride the bike accordingly, i.e. not jumping out of the saddle every two minutes and trying to power up every hill in a high gear. In other words it works well with a smooth pedalling action and sensible use of gears. If you don't/can't ride like this it's not for you. I do however think this is the most efficient way to ride a bike anyway (you only have to watch the pros doing those big climbs to see that).

[Jodel]

thanks for all that - will try to digest (not as technical as you). I didn't know about the 11-36T 9 speed cassette but no matter, I prefer the 12-36, available at a decent price from Rose. In fact that 12-36T cassette means I can think of no good reason to go beyond 9 speed for general cycling.
I am flat bar so all is more simple, I can easily bodge and give two fingers to the marketing zealots.
About to set out on a build with a rear mech which may be a 34T limit according to official specs - will tread carefully with on-the-stand testing as am hoping I can push it to 36.

I have a 9 speed 12 - 36 on my tandem (26 / 36 / 46 chainrings on the front) and both an Shimano Altus or Deore RD-M591 rear mech work with this set-up. I think both of those derailleurs are only rated to a 34T limit, but they've been fine for me. Chain length is 114 links.