Author Topic: Are all front mechs created equal?  (Read 8093 times)

rogerzilla

  • When n+1 gets out of hand
Re: Are all front mechs created equal?
« Reply #25 on: 03 September, 2017, 09:35:32 pm »
Even a Sturmey-Archer hub is indexed at the shifter.  Rohloff is a real exception.  One disadvantage of havng the indexing at the hub is that the shifts will always feel mushy and remote.

Pre-stretched gear inner cables and constant-length SIS housing solve the problem of correct cable pull most of the time.  Careful squaring and de-burring of the housing ends before it goes in the ferrule are important as you don't want to inadvertently introduce squish into the system.
Hard work sometimes pays off in the end, but laziness ALWAYS pays off NOW.

Re: Are all front mechs created equal?
« Reply #26 on: 03 September, 2017, 09:57:03 pm »
as LWaB says Shimano's first indexed derailleur was a thing called a 'Positron' which had the index detents built into the mech.

 It was (joy of joys) controlled with a solid wire instead of a cable, that could push as well as pull. I forget the details  but there was either a weak spring or no spring in the mech itself, so the shifter needed to push the mech to make upshifts.

To allow accurate pulling and pushing, the cable housing was like modern SIS housing (i.e. reinforced with near-parallel strands, lined etc) and was itself clamped tight at either end of a long continuous housing run.

Within the constraints of the shifting technology of the time, (I think UG style twist teeth were the coming thing) the shifting was OK for a pressed steel mech made to a price. I think the same approach could give better results with modern cassettes and chains.

 However there was

a) no real scope for overshifting, which at the time was a common requirement on downshifts and
b) the solid 'cable' was still wrapped around a pulley at one end and still flexed at the other; this made it vulnerable to breakage.

When the 'cable' failed it turned out to be a part that wasn't readily available.  You can (just about) buy NOS cables to this day, if you look carefully enough.

SIS came later with shifting that was at least as good, if not better because there was scope for an overshift if required. 

IIRC we were beguiled by this technology and if we didn't want it, or it stopped working, there was a little switch on the lever that allowed you  to revert to friction shifting. I wish we still had that switch...

BTW Sturmey Archer had an elegant (barely any larger than the smallest friction type)  set of stem mount or DT mount levers which were intended to control their 5s hub. I don't know for sure but these may have been the first that were sold (by anyone) for bicycle use with a ball-detent mechanism in the lever to give an indexed action.

cheers

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #27 on: 03 September, 2017, 10:22:10 pm »
I'm using 105 road mechs* on my MTB build.  If you don't have front indexing, most triples will do rhe job si long as you get the right fitting and pull direction.

And for those who do have front indexing, if you want to use a MTB shifter to shift a road front mech, then FD-R453 is your magic number.

ElyDave

  • Royal and Ancient Polar Bear Society member 263583
Re: Are all front mechs created equal?
« Reply #28 on: 03 September, 2017, 10:28:17 pm »
I'm using 105 road mechs* on my MTB build.  If you don't have front indexing, most triples will do rhe job si long as you get the right fitting and pull direction.

And for those who do have front indexing, if you want to use a MTB shifter to shift a road front mech, then FD-R453 is your magic number.

Can't remember what I used on the M5 build, but it was MTB levers, shifters, cassette and rear mech, but front road compact double and road derailleur, seems to be working ok 1000 miles on.
“Procrastination is the thief of time, collar him.” –Charles Dickens

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #29 on: 03 September, 2017, 10:38:30 pm »
It's presumably less critical for doubles, as they only have two positions:  Pulled up to the big ring, and dropped to the lower limit screw.  Excess cable slack in the lower position shouldn't affect how well it works.

JennyB

  • Old enough to know better
Re: Are all front mechs created equal?
« Reply #30 on: 04 September, 2017, 08:38:46 am »
as LWaB says Shimano's first indexed derailleur was a thing called a 'Positron' which had the index detents built into the mech.

 It was (joy of joys) controlled with a solid wire instead of a cable, that could push as well as pull. I forget the details  but there was either a weak spring or no spring in the mech itself, so the shifter needed to push the mech to make upshifts.


When the 'cable' failed it turned out to be a part that wasn't readily available.  You can (just about) buy NOS cables to this day, if you look carefully enough.


Maybe someone will do a hydraulic version?  There was also the old-fashioned way, with a lever on the mech. A bit hard to reach, but as positive as you could get.
Jennifer - Walker of hills

Re: Are all front mechs created equal?
« Reply #31 on: 04 September, 2017, 08:54:06 am »
a problem with having the indexing in the mech is that (esp with derailleur gears) the indexing can't be adjusted on the move.

  By contrast with indexing in the shifter, a tweak can be made if necessary by twirling a barrel adjuster, without stopping; this is a good feature to have whilst racing.

Every system has its plus points; needless to say, perfection has yet to be achieved....

cheers

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #32 on: 04 September, 2017, 12:44:32 pm »
Sure, but most of the need for adjustment is because there's a load of sloppy cable between the indexing and the mech.  Unless something gets bent, I suppose.


I'm coming round to the doing indexing in software way of thinking...

Re: Are all front mechs created equal?
« Reply #33 on: 04 September, 2017, 01:04:51 pm »
Sure, but most of the need for adjustment is because there's a load of sloppy cable between the indexing and the mech.  Unless something gets bent, I suppose.
it happens often....


Quote
I'm coming round to the doing indexing in software way of thinking...

the thing is that you are essentially having a very complicated system to do something that is, at heart, rather simple, which is lining up a mech with some sprockets and pulleys. 

 Indexing of any kind has some benefit for racers (who want to make shifts under duress) and newbies (who find it simpler) but derailleur gears had been around for about eighty years before anyone added indexing; think of a the millions and millions and millions of miles that were happily cycled without indexing. Shifting gears with friction shifters isn't difficult but it is a skill that has to be learned.

  If I was wanting to sell bikes to newbies I'd spec indexed shifters; they will choose indexed shifting because they will find it easier.  Is it 'better'...? Well in the long run I don't know; I'd like to be able to turn it off if necessary. It is the first thing that goes wrong on an awful lot of bikes and I don't think making the system more complicated is going to help that.

I recently witnessed the same bike in the LBS (to fix different things) about five times in the space of as many weeks. The bike had Di2 and hydraulic disc brakes, amongst other things. It may have had tubeless tyres as well. I was astonished to discover that its owner was preparing to embark upon a ride across Europe in the near future on that bike. He had zero chance of fixing any of these parts if they started to give trouble on the road.  I thought that since the bike had clearly not gone more than a few hundred miles between problems leading up to the ride, the chances of it not giving trouble on the ride he planned to do was about zero. 

In fact it represented the very antithesis of the kind of bike that I would have taken on that kind of trip. Only suitable for straying a taxi-ride's distance away from home in my book....

cheers


fruitcake

  • some kind of fruitcake
Re: Are all front mechs created equal?
« Reply #34 on: 04 September, 2017, 01:30:41 pm »
It's presumably less critical for doubles, as they only have two positions:  Pulled up to the big ring, and dropped to the lower limit screw.  Excess cable slack in the lower position shouldn't affect how well it works.

This, basically.

A system with two rings up front offers the benefits of a choice of ring size, and 'chain catching' functionality within the front mech itself, without (many of) the component compatibility issues.

Re: Are all front mechs created equal?
« Reply #35 on: 04 September, 2017, 01:57:56 pm »
Quote
In fact it represented the very antithesis of the kind of bike that I would have taken on that kind of trip. Only suitable for straying a taxi-ride's distance away from home in my book....

Cycling and audax is at an interesting crossroads or perhaps slight junction.  If we applied Bruces comments to cars then I would be better off driving my Triumph Herald or MG midget (on which I could and did change everything) to the South of France rather than my new Volvo which has no user serviceable parts.

However I know which I and everybody else would generally take.  Whilst some iterations of Di2 and hydraulic brakes are a problem, I suspect the MTBF for Di2 is substantially greater than for cable actuated systems.

I may be wrong but my understanding is that the majority of TCR riders are now using Di2 or eTap for this reason.

We may like cable actuated but experience suggests that Di2 is more reliable and less likely to fail in real life use.

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #36 on: 04 September, 2017, 02:05:15 pm »
Di2 indexing is also easier to tweak.  Put it in the settings mode and frob the gear change buttons until the whatever is perfectly aligned, save the settings.  Barrel adjusters are simple, low tech, unreliable and awkward by comparison.

The reliability of hydraulic brakes is well proven.

As for the average car driver, the ease of roadside repair is equivalent for all systems when you have no tools and don't know what you're doing.  The average bike user might as well pick the more reliable (or better performing) one.

Mechanical skills aren't actually a prerequisite for being able to ride a bike a long way.

mattc

  • n.b. have grown beard since photo taken
    • Didcot Audaxes
Re: Are all front mechs created equal?
« Reply #37 on: 04 September, 2017, 02:34:02 pm »

Quote
I'm coming round to the doing indexing in software way of thinking...

the thing is that you are essentially having a very complicated system to do something that is, at heart, rather simple, which is lining up a mech with some sprockets and pulleys. 

 Indexing of any kind has some benefit for racers (who want to make shifts under duress) and newbies (who find it simpler) but derailleur gears had been around for about eighty years before anyone added indexing; think of a the millions and millions and millions of miles that were happily cycled without indexing. Shifting gears with friction shifters isn't difficult but it is a skill that has to be learned.

  If I was wanting to sell bikes to newbies I'd spec indexed shifters; they will choose indexed shifting because they will find it easier.  Is it 'better'...? Well in the long run I don't know; I'd like to be able to turn it off if necessary. It is the first thing that goes wrong on an awful lot of bikes and I don't think making the system more complicated is going to help that.

...
Indexing is probably a good thing with 10-11-12speed systems (and 9?). So riders who "need" those systems probably want indexing. And perhaps Di2 - once mature - will be better than cable-indexing.

But front derailleurs aren't 10-speed - they are at most 3speed, and the fashionable cash-rich rider generally just has two-speed upfront. So I remain convinced (for now)  that Campy micro-index shifters at the front, or friction levers are the best system for most of us.
 [and friction may be "best" for the rear upto 8speed. Not sure ... ]



Do beginners "need" indexing? Well, a minority of drivers need automatics - but I would say that manual car gear-changing is far harder to learn than adjusting a friction lever to line up your chain on a sprocket.
Has never ridden RAAM
---------
No.11  Because of the great host of those who dislike the least appearance of "swank " when they travel the roads and lanes. - From Kuklos' 39 Articles

ElyDave

  • Royal and Ancient Polar Bear Society member 263583
Re: Are all front mechs created equal?
« Reply #38 on: 04 September, 2017, 04:27:16 pm »
This has gone way, way OT and become very interesting. But let's take a look at my study of one.

Bike is now > 10 yo. 

Upgrades = 1, 8sp to 10sp, requiring new cassettes and 105 shifter.  No problems with this
Failed parts = 3
- rear derailleur just got sloppy and couldn't handle the 10sp, replaced, no issues since
- freehub, water ingress, races pitted, replaced, no further issues
- front mech - will be replaced.
Replaced on wear - chainset, sora triple upgraded to ultegra 'cos a mate had an unused one going cheap.

NO indexing failures, failures at the other end of the cable. MTBF looks pretty good on mechanical there. 

I have no experience of DI2 or hydraulic discs

“Procrastination is the thief of time, collar him.” –Charles Dickens

Re: Are all front mechs created equal?
« Reply #39 on: 04 September, 2017, 08:08:15 pm »
modern cars are an interesting case in point.  They often contain the same kinds of mechanical parts that older cars had, but have an additional layer of electronics smeared over the top.  This almost inevitably makes the system more complicated and less reliable than it could be.

Cars are presently (by weight) about 1/4 electronics and by value about 1/2 electronics.  Past crash damage, electronic failures are the main reason for vehicles being 'uneconomic to repair'.

I don't think bicycle hydraulic brakes are very reliable either. I have seen lots break or leak for no good reason.

The idea that Di2 has a longer MTBF is not borne out; as I mentioned previously since it has come into common use I have not seen a race where at least one rider has had gears that have mysteriously stopped working. It has happened to Froomy several times. If Sky's mechanics can't make it reliable what chance does the average punter stand exactly...?

cheers

Re: Are all front mechs created equal?
« Reply #40 on: 04 September, 2017, 09:34:34 pm »
modern cars are an interesting case in point.  They often contain the same kinds of mechanical parts that older cars had, but have an additional layer of electronics smeared over the top.  This almost inevitably makes the system more complicated and less reliable than it could be.

Cars are presently (by weight) about 1/4 electronics and by value about 1/2 electronics.  Past crash damage, electronic failures are the main reason for vehicles being 'uneconomic to repair'.

I don't think bicycle hydraulic brakes are very reliable either. I have seen lots break or leak for no good reason.

The idea that Di2 has a longer MTBF is not borne out; as I mentioned previously since it has come into common use I have not seen a race where at least one rider has had gears that have mysteriously stopped working. It has happened to Froomy several times. If Sky's mechanics can't make it reliable what chance does the average punter stand exactly...?

cheers

Electronics fail in cars for two simple reasons; the first one is that at least on low and mid range vehicles the components are bought by the kilo from the cheapest suppliers; expensive high-end vehicles get the components that have been properly inspected for quality. the second reason is the insistance on integrating a lot of over sophisticated systems into one giant mesh. Failure of a 10 cent sender in one part (door locks, window winders, temperature sensors etc) can lead to the whole integrated set-up refusing to function. If the correct diagnostic approaches are not built in and the garage refuses the manufacturer's approach of replacing large bits of electronics with new then the vehicle ends up immobilised while someone tries to work out which 10 cent piece  is faulty.

Re: Are all front mechs created equal?
« Reply #41 on: 04 September, 2017, 09:42:27 pm »
Sure, but most of the need for adjustment is because there's a load of sloppy cable between the indexing and the mech.  Unless something gets bent, I suppose.
it happens often....


Quote
I'm coming round to the doing indexing in software way of thinking...

the thing is that you are essentially having a very complicated system to do something that is, at heart, rather simple, which is lining up a mech with some sprockets and pulleys. 

 Indexing of any kind has some benefit for racers (who want to make shifts under duress) and newbies (who find it simpler) but derailleur gears had been around for about eighty years before anyone added indexing; think of a the millions and millions and millions of miles that were happily cycled without indexing. Shifting gears with friction shifters isn't difficult but it is a skill that has to be learned.


cheers

Osgears (or Super Champion derailleurs) had indexing, even though they needed overshifting to work (and a spoke guard to counter the absence of limit screws!). That was the 1930's. earlier than that I don't know although the Cyclo gears from the 20's didn't have AFAIK.

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #42 on: 04 September, 2017, 09:48:45 pm »
The problem with electronics in cars isn't so much that it's difficult to repair, as that it's all proprietary vendor-specific secret sauce.  Diagnostics are limited and prohibitively expensive - you take it to the dealer and they plug in their test computer that says there's a problem with the foo sensor, so you replace that and then it starts moaning about the bar valve servo, so you replace that and it doesn't fix the problem and then the next logical thing to try is a new ECU but that's more than the car is worth so you scrap it.

Same principle applies to most e-bikes, and indeed Di2.  It's just that they're an order of magnitude less complicated than running a modern internal combustion engine, so I expect the mean-time-to-uneconomical-repair to be somewhat better.  I expect that batteries dying of old age will be the usual failure mode, with newer better systems looking much more attractive than an expensive replacement battery.

My parents had a B&W telly from prehistoric times (or at least the early 1970s).  It fell off a shelf in a caravan once when I was small, failed to work, and gathered dust in the loft for years.  As a curious teenager I dug it out, opened it up and discovered two things:  Firstly, a broken inlet fuse (looked to be the result of mechanical trauma rather than overload), which on being replaced restored full functionality.  Secondly, folded up within a slot on the inside of the chassis, a complete schematic, including part numbers and photos of oscilloscope traces at all the important test points.  Anyone with sufficient electronics knowledge could have diagnosed and repaired a fault, or indeed built their own copy.

Modern electronics is the equivalent of glued together plastic parts.  The ways that they're commonly engineered to maximise profit margins on consumer products are no more inherent to electronics than they are to plastic.

Re: Are all front mechs created equal?
« Reply #43 on: 04 September, 2017, 10:06:12 pm »
The problem with electronics in cars isn't so much that it's difficult to repair, as that it's all proprietary vendor-specific secret sauce.  Diagnostics are limited and prohibitively expensive - you take it to the dealer and they plug in their test computer that says there's a problem with the foo sensor, so you replace that and then it starts moaning about the bar valve servo, so you replace that and it doesn't fix the problem and then the next logical thing to try is a new ECU but that's more than the car is worth so you scrap it.

Same principle applies to most e-bikes, and indeed Di2.  It's just that they're an order of magnitude less complicated than running a modern internal combustion engine, so I expect the mean-time-to-uneconomical-repair to be somewhat better.  I expect that batteries dying of old age will be the usual failure mode, with newer better systems looking much more attractive than an expensive replacement battery.

My parents had a B&W telly from prehistoric times (or at least the early 1970s).  It fell off a shelf in a caravan once when I was small, failed to work, and gathered dust in the loft for years.  As a curious teenager I dug it out, opened it up and discovered two things:  Firstly, a broken inlet fuse (looked to be the result of mechanical trauma rather than overload), which on being replaced restored full functionality.  Secondly, folded up within a slot on the inside of the chassis, a complete schematic, including part numbers and photos of oscilloscope traces at all the important test points.

Modern electronics is the equivalent of glued together plastic parts.  The ways that they're commonly engineered to maximise profit margins on consumer products are no more inherent to electronics than they are to plastic.

The problems arise when the dealer doesn't want to replace the bit in the way the maker proposed and doesn't have the skills to do otherwise, as happened to a new (under warranty) VW Touareg, which was immobilised for over 8 months because the dealer (or VW) didn't want to replace a complete module and couldn't work out which bit of the module was not working. I never heard the final resolution but the owner was most unchuffed.

Yes the problem with electronics is so often the cost-cutting approach. I think that when it's a card in a piece of industrial production machine and replacement cost plus downtime is a more imposing figure then the quality improves, as if by magic.

Re: Are all front mechs created equal?
« Reply #44 on: 04 September, 2017, 10:12:12 pm »
....Same principle applies to most e-bikes, and indeed Di2.  It's just that they're an order of magnitude less complicated than running a modern internal combustion engine, so I expect the mean-time-to-uneconomical-repair to be somewhat better.....

but they also are not sitting on a suspended platform, they can see more weather/road salt, and are not positioned where they will occasionally (or regularly) warm up and dry out in service. In other words bicycle electronics can have a much harder time of it.

Just having properly reliable wiring seems beyond most car makers, and well beyond anything that is found on a bicycle.... ::-)

Why add another layer of complexity in order to have an unfixable, unreliable electronic control of limited utility?  The bike already has the most sophisticated control system available, the human brain....

cheers

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #45 on: 04 September, 2017, 10:18:23 pm »
Why add another layer of complexity in order to have an unfixable, unreliable electronic control of limited utility?  The bike already has the most sophisticated control system available, the human brain....

Because sometimes - like with the gears in a car - the human brain, human hands or human legs aren't quite up to the job.  Or the human simply can't be arsed.  Or because in a given niche electronics do work substantially better than the mechanical alternative (I'm thinking hand disabilities, tandem gear cables, or a second set of shift buttons on tri-bars).  Or maybe they will actually turn out to be more reliable.  Most of the Di2 horror stories I've seen are user error (ie. forgetting to charge the batteries), which isn't the fault of the technology, or due to exposure to the kind of weather conditions that cause gear cables to stick.

None of which is going to make me ditch the simple, (mostly) reliable and easily serviced at the roadside bar-end shifters on my touring bike.  But other bikes have other uses.

Re: Are all front mechs created equal?
« Reply #46 on: 05 September, 2017, 10:37:20 am »
Because sometimes - like with the gears in a car - the human brain, human hands or human legs aren't quite up to the job.  Or the human simply can't be arsed.  Or because in a given niche electronics do work substantially better than the mechanical alternative (I'm thinking hand disabilities, tandem gear cables, or a second set of shift buttons on tri-bars).  Or maybe they will actually turn out to be more reliable.

.....other bikes have other uses.

There is a small benefit in certain conditions to some of the above. Mostly not though.  You forgot to add 'keeping up with the Joneses/Pros' to you list of reasons why folk might buy Di2. I think this is the main one.

  It is my contention that the vast majority of folk who buy Di2 do not get any substantial benefit from using the system and are effectively not admitting to themselves that they are in fact touring rather than racing.

BTW when I asked the chap with the (seemingly unreliable) bike why he'd chosen Di2 he told me (and I paraphrase) that he'd been told that his (healthy, normal)  fingers might get tired otherwise.

 I had to stop myself from looking too baffled or  laughing out loud.

cheers

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #47 on: 05 September, 2017, 03:01:19 pm »
Because sometimes - like with the gears in a car - the human brain, human hands or human legs aren't quite up to the job.  Or the human simply can't be arsed.  Or because in a given niche electronics do work substantially better than the mechanical alternative (I'm thinking hand disabilities, tandem gear cables, or a second set of shift buttons on tri-bars).  Or maybe they will actually turn out to be more reliable.

.....other bikes have other uses.

There is a small benefit in certain conditions to some of the above. Mostly not though.  You forgot to add 'keeping up with the Joneses/Pros' to you list of reasons why folk might buy Di2. I think this is the main one.

  It is my contention that the vast majority of folk who buy Di2 do not get any substantial benefit from using the system and are effectively not admitting to themselves that they are in fact touring rather than racing.

Sure.  And the majority of iPhones are bought for willy-waving.

That's a good thing.  It means iPhones are a cheap, reliable mass-market product for the blind people who need them.

Same goes for Di2.  The wannabe-racers make it a viable option for those who stand to benefit.  If it wasn't for them, mainstream bicycle technology would be stuck in the 1930s.


Quote
BTW when I asked the chap with the (seemingly unreliable) bike why he'd chosen Di2 he told me (and I paraphrase) that he'd been told that his (healthy, normal)  fingers might get tired otherwise.

I had to stop myself from looking too baffled or  laughing out loud.

I believe at least one audaxer of this parish invested in a bike with Di2 to avoid neuropathy on LEL, after their experience last time round.  Seems like a good reason to me.  Hands are far more precious than bike rides.

Re: Are all front mechs created equal?
« Reply #48 on: 05 September, 2017, 05:03:39 pm »
I believe at least one audaxer of this parish invested in a bike with Di2 to avoid neuropathy on LEL, after their experience last time round.  Seems like a good reason to me.  Hands are far more precious than bike rides.

Fundamental health problems aside, they might be better off investing in a decent bike fit rather than the world's most expensive shifters/gears. If they were struggling to change gear it was most likely a symptom of a sub-optimal riding position rather than anything else; different shifters is a sticking plaster fix in all probability. It is not as if they are the only type of shifter that are easy to use, either....

cheers

Kim

  • Timelord
    • Fediverse
Re: Are all front mechs created equal?
« Reply #49 on: 05 September, 2017, 05:04:10 pm »
AIUI that's exactly what they did.


Anyway, you've made it abundantly clear that you hate Di2.  Doesn't mean the rest of us have to.