Chain "Snapped"

[SkipdiverJohn]

A lot of modern bike tech seems to be a solution looking for an imaginary problem that doesn't really exist. I too, am a big fan of quill stems and simple DIY-serviceable ball-bearing headsets and BB's.
Why faff around with ugly stacks of spacers when all you need to do with a quill is loosen the clamp bolt, raise or lower (by infinitely small increments if you wish) until you are happy, then re-tighten the bolt. Job done!. If you want to experiment, or you're lending a bike to a different sized rider, it's a quick job to change heights, and easy to revert back again. You also can't make an expensive mistake like cutting your steerer too short and being forever stuck with a too-low bar height.

@Rusty Nails is right about "good enough" meaning different things for different purposes. Most cyclists simply want to be able to get from A to B in a reasonable bit of comfort and with acceptably good mechanical refinement and reliability. They aren't interested in, nor prepared to pay extra, for "marginal gains"
They don't care about saving a few milliseconds during a gearchange, they don't care about average speed, they aren't bothered that their BB may absorb an extra two watts or weighs two ounces more than some mega-expensive version that needs a different special tool to remove.

In no particular order, my ideal bike design criteria includes:-
Frame handling characteristics
Comfort & fit
Aesthetic appearance
Ease of DIY maintenance
Cost of bike & spare parts
Mechanical reliability & longevity
Parts interchangeability

Nowhere in my list appears the desire for absolute minimal weight, aerodynamics, or to do zero routine maintenance then take it to someone else and let them fix it when it breaks. I would suspect that everyday cyclists with a practical and pragmatic approach vastly outnumber the speed and weight focussed obsessives who frequent cycling forums and who actually believe that spending several thousand pounds on a push bike is NOT completely insane.

 

[rogerzilla]

Industrial cartridge bearings are superior for practical, everyday use. When they get worn (and IME they last a long time with no attention), you throw all the wear away, fit a cheap new bearing, and the hub/BB/headset is like new again. OK, they don't take side loads as well as cup-and-cone bearings, and they're not infinitely adjustable, but they hold up better in the real world. The only issue is that they take a few simple tools to swap over, which you probably don't have in your toolbox.

Most of my bikes have quill stems but, apart from a bit of extra knee clearance when climbing, they're far more prone to problems (seizing in place) and even I can feel they're not as stiff. I never adjust stem height after I've built a bike up.

Modern headsets are absolutely brilliant compared to what we used to put up with, The problem of "indexed steering" has gone away with cartridge headsets and yes, you can get threaded ones.

 

[SkipdiverJohn]

Modern headsets are absolutely brilliant compared to what we used to put up with, The problem of "indexed steering" has gone away with cartridge headsets and yes, you can get threaded ones.

The only example I've encountered of "indexed steering" was on a battered Raleigh Pioneer frame that had probably been left outside for ten years or more. As I only wanted it for local hack use, I just dismantled and soaked all the rusty bits in diesel, wiped everything with a rag and then put the whole lot back together with as much fresh grease as I could cram in. No new balls or races fitted. After a few miles it loosened up slightly and was readjusted, and has since settled down to be really smooth and nice to steer with several hundred miles on the clock. You would never guess the rusty state the headset was in from the way it rides now.

 

[Yellow Saddle]

Industrial cartridge bearings are superior for practical, everyday use. When they get worn (and IME they last a long time with no attention), you throw all the wear away, fit a cheap new bearing, and the hub/BB/headset is like new again. OK, they don't take side loads as well as cup-and-cone bearings, and they're not infinitely adjustable, but they hold up better in the real world. The only issue is that they take a few simple tools to swap over, which you probably don't have in your toolbox.

Most of my bikes have quill stems but, apart from a bit of extra knee clearance when climbing, they're far more prone to problems (seizing in place) and even I can feel they're not as stiff. I never adjust stem height after I've built a bike up.

Modern headsets are absolutely brilliant compared to what we used to put up with, The problem of "indexed steering" has gone away with cartridge headsets and yes, you can get threaded ones.

Ahmen brother.

 

[Yellow Saddle]

The only example I've encountered of "indexed steering" was on a battered Raleigh Pioneer frame that had probably been left outside for ten years or more. As I only wanted it for local hack use, I just dismantled and soaked all the rusty bits in diesel, wiped everything with a rag and then put the whole lot back together with as much fresh grease as I could cram in. No new balls or races fitted. After a few miles it loosened up slightly and was readjusted, and has since settled down to be really smooth and nice to steer with several hundred miles on the clock. You would never guess the rusty state the headset was in from the way it rides now.

Then you haven't had much experience, and even the one example you cited wasn't indexed. It was just dirty. You're not going to cure a dimpled headset race with diesel and rags. Further, headsets don't get dimpled from standing outside in the rain. They dimple from riding.

 

[Sharky]

Some technical advances have been significant and some marginal.

 

[rogerzilla]

Biggest advances have been in lights (which used to be universally awful), headsets and brakes on cheap bikes; brakes on better bikes were always fine, if often hard to set up. Gears are more of an incremental thing. The slant parallelogram rear mech was probably a bigger gain than indexing; even friction shift is pretty good with a SP mech.

I know most people love it but I don't find STI useful in any of its forms. I really like MTB thumbshifters and bar-end shifters. My carbon road bike has STI and I find it clunky (too much throw of the levers), not very snappy and hopeless for knowing which gear you're in and whether you're cross-chaining.

 

[Salty seadog]

Then you haven't had much experience, and even the one example you cited wasn't indexed. It was just dirty. You're not going to cure a dimpled headset race with diesel and rags. Further, headsets don't get dimpled from standing outside in the rain. They dimple from riding.

Happened once too my old mtb. The headset would centre dead ahead. Can't remember if there was another notching.

 

[Yellow Saddle]

Happened once too my old mtb. The headset would centre dead ahead. Can't remember if there was another notching.

Yes there were other notchings. If there were 36 bearing balls in there, there was a notching every 10 degrees. If there were 18 bearings, then another notching every 20 degrees. The dimples in the race develop underneath each ball, but develop only in the straight-ahead position. By that I mean that they don't develop during turning. Their creation is counter-intuitive. Most people, when shown such a dimpled race, conclude that because the race is dimpled, it was dimpled by impact. But that isn't the cause of the dimpling. It is dimpled by erosion and specific galling (welding).

In the straight-ahead position, the steering system oscillates axially along the steering tube from small road and rider inputs. You have to steer a little bit to ride in a straight line (yes, it sounds like a paradox but it isn't). The fork itself oscillates fore/aft from road roughness. This oscillation is accentuated in thinner steer tubes such as the olden days' 1" steerers and has the effect of pinching the two races around a few balls in the fore/aft position. Imagine a radar signal going forwards and another backwards, that's the approximate position of this pinching. The pinching the restricts the movement of the balls during the high amplitude position of the rough spot. Any rolling movement of the ball during this tiny moment when the ball is pinched is restricted and the ball can only slide. However, it doesn't slide. Reason for that, follows in the next sentence.

With axial oscillation of the steering system (little left and right turns) the balls don't roll around the bearing like in a wheel bearing, but just wobbles a bit in place. This displaces grease between the ball and race. As soon as the grease is depleted, we have steel on steel. Now, back to the pinched ball. Once the ball is devoid of grease and being pinched, it is forced to slide under pressure. This it does, but immediately galls - just one or two molecules weld. The next vibration peak (I'm talking single vibrate cycle here) breaks the weld and those two molecules are now lost to the race. Eventually it dimples. These dimples are smooth, but not polished, they have a milky, satin finish.

Enough of those and your steering is indexed. I like the term indexed, because that avoids the erroneous term brinelling. Brinelling is after the Swedish metallurgist/engineer Brinell who developed a metal hardness test by dimpling materials with a hard steel ball.

This doesn't happen to modern cartridge bearing A-headset style headsets. The reason is partially thanks due to the stronger (bigger diameter) steerer tube but mostly due to the design of the bearing. These bearings have a 35 degree or 38 degree chamfer on the outside race. This chamfer allows the bearing to shift in the frame rather than pinch the balls inside. A truly brilliant design. Inventions like this are invisible to the quill-is-better flat-earth society but have changed the reliability of bicycles enormously.

Edit: the chamfer angle is 45 degree or 38 degree, depending on brand and patent in place.

 

[Smokin Joe]

This doesn't happen to modern cartridge bearing A-headset style headsets. The reason is partially thanks due to the stronger (bigger diameter) steerer tube but mostly due to the design of the bearing. These bearings have a 35 degree or 38 degree chamfer on the outside race. This chamfer allows the bearing to shift in the frame rather than pinch the balls inside. A truly brilliant design. Inventions like this are invisible to the quill-is-better flat-earth society but have changed the reliability of bicycles enormously.

Amen to that. The old style threaded headsets were a pain in the butt needing regular maintenance and replacement. The first two A - Headset systems I used both went seven years without being touched and were still smooth and in adjustment when I sold the bikes. The best thing they did was do away with the crown race which is the major source of the problem - "Crown races" are now not bearing races at all but locating rings for the cartridge. They can be split with a hacksaw for ease of fitting if they don't already come like that and I don't know why manufacturers even bother with them any more, the fork could be designed to take the cartridge direct (Which I believe some do).

 

[Salty seadog]

Yes there were other notchings. If there were 36 bearing balls in there, there was a notching every 10 degrees. If there were 18 bearings, then another notching every 20 degrees. The dimples in the race develop underneath each ball, but develop only in the straight-ahead position. By that I mean that they don't develop during turning. Their creation is counter-intuitive. Most people, when shown such a dimpled race, conclude that because the race is dimpled, it was dimpled by impact. But that isn't the cause of the dimpling. It is dimpled by erosion and specific galling (welding).

In the straight-ahead position, the steering system oscillates axially along the steering tube from small road and rider inputs. You have to steer a little bit to ride in a straight line (yes, it sounds like a paradox but it isn't). The fork itself oscillates fore/aft from road roughness. This oscillation is accentuated in thinner steer tubes such as the olden days' 1" steerers and has the effect of pinching the two races around a few balls in the fore/aft position. Imagine a radar signal going forwards and another backwards, that's the approximate position of this pinching. The pinching the restricts the movement of the balls during the high amplitude position of the rough spot. Any rolling movement of the ball during this tiny moment when the ball is pinched is restricted and the ball can only slide. However, it doesn't slide. Reason for that, follows in the next sentence.

With axial oscillation of the steering system (little left and right turns) the balls don't roll around the bearing like in a wheel bearing, but just wobbles a bit in place. This displaces grease between the ball and race. As soon as the grease is depleted, we have steel on steel. Now, back to the pinched ball. Once the ball is devoid of grease and being pinched, it is forced to slide under pressure. This it does, but immediately galls - just one or two molecules weld. The next vibration peak (I'm talking single vibrate cycle here) breaks the weld and those two molecules are now lost to the race. Eventually it dimples. These dimples are smooth, but not polished, they have a milky, satin finish.

Enough of those and your steering is indexed. I like the term indexed, because that avoids the erroneous term brinelling. Brinelling is after the Swedish metallurgist/engineer Brinell who developed a metal hardness test by dimpling materials with a hard steel ball.

This doesn't happen to modern cartridge bearing A-headset style headsets. The reason is partially thanks due to the stronger (bigger diameter) steerer tube but mostly due to the design of the bearing. These bearings have a 35 degree or 38 degree chamfer on the outside race. This chamfer allows the bearing to shift in the frame rather than pinch the balls inside. A truly brilliant design. Inventions like this are invisible to the quill-is-better flat-earth society but have changed the reliability of bicycles enormously.

Edit: the chamfer angle is 45 degree or 38 degree, depending on brand and patent in place.

I get the gist of that even if I can't picture it all at first reading. Cheers YS.

 

[Yellow Saddle]

I get the gist of that even if I can't picture it all at first reading. Cheers YS.

Next time you go for a ride, lean forward and aim straight down one of your fork legs, You'll see it vibrate in the plane of travel direction. Now imagine what it does to bearing balls sandwiched between circular rails. Picture the 12 'o clock position versus the 3 O'clock position.

 

[rogerzilla]

You can, of course, buy a threaded headset with these same advantages. The Tange Falcon is my favourite. The bearing cartridges are shared with Cane Creek so you can fit spares if they ever wear out.

 

[SkipdiverJohn]

With axial oscillation of the steering system (little left and right turns) the balls don't roll around the bearing like in a wheel bearing, but just wobbles a bit in place. This displaces grease between the ball and race. As soon as the grease is depleted, we have steel on steel. Now, back to the pinched ball. Once the ball is devoid of grease and being pinched, it is forced to slide under pressure. This it does, but immediately galls - just one or two molecules weld. The next vibration peak (I'm talking single vibrate cycle here) breaks the weld and those two molecules are now lost to the race. Eventually it dimples. .

If you use Extreme Pressure greases primarily designed for motor vehicle use, the EP additives will tend to maintain boundary lubrication at the molecular level, even if the lubricant has been physically squeezed out from between the surfaces.

 

[Yellow Saddle]

If you use Extreme Pressure greases primarily designed for motor vehicle use, the EP additives will tend to maintain boundary lubrication at the molecular level, even if the lubricant has been physically squeezed out from between the surfaces.

That's the theory, but the mechanism by which the greased is depleted between the two surfaces in that scenario depletes all lubrication. Proof is that there's no cure for dimpling other than re-design. The extreme-pressure sulfated grease trick has been tried.