Derailleurs...what do you look for in yours?

tom_goggin

Im doing a university project looking into re-designing the current derailleur/shifter system to combat problems such as weathering/corrosion and the amount of maintenence required to keep shifting crisp. I know the idea other topics have discussed potential re-designs of gear shifting mechanisms, but what I really wan to know is...what do you want from your shifter/derailleur? Any comments are appreciated, thanks in advance!

supersonic

I don't really find corrosion to be a problem except on the ultra cheap ones which is more due to the materials used. They are pretty weather proof too - the cables are the main problems with poor shifting, but there are remedies for that as well ie full length outers or inner sleeves. A quick wash down (if needed) and a squirt of lube on the pivots and bushes and that is it really! Hydraulic systems already exist, as do electronic systems. In the trade we always see the derailer is dumb - it does what the shifter tells it, then the chain and cassette help the shift with the ramps etc.

Total weather proofing seems to be more of a hub gear thing now.

The modern derailer is a very evolved technology, with the likes of Shimano now producing adjustable mechs for different capacities and SRAM with superlight designs.

daniel_owen_uk

You know I was looking at one the other day and something struck me, why does the cable almost always come from the back?

You are pulling from the front, so why have it turn the corner?

supersonic

It needs a loop of outer anyway because the mech cage moves.

Anonymous

SRAM's cable routing is direct, and so are Shimano's newer "shadow" derailleurs.
However, I've noticed that on a lot of bikes where the rear mech cable goes along the chainstay, as opposed to the seatstay, even this direct cabling still gives an odd angle on entry into the derailleur.

saltyman

im running SRAM X-0 rear mech and shifters, never had any issues with them, lightness isnt my priority, reliability is when it comes to gears. yes they are light and expensive but i have had no problems, exactly what every MTB rider wants.

no idea what else i would want from a rear mech come to think of it.

front shifters, i always prefer rapidfire thumb style shifters. used a few different types of gripshift and found it too slow and unreliable for me.

The Northern Monkey

I want bullet proof protection. Something thats not gonna corrode and is light without sacrificing strength...... oh wait, I already have that in the Saint derailleur!

I'd say that slick shifting is as much to do with the shifters as it is to do with the derailleur though.

Anonymous

Ah, saltyman, your first pic hits the nail on the head with what I was trying to explain earlier.
See that funny angle that the sable enters the mech, compared to the angle of the inner section of cable? It's as if they're designed primarily for seatstay mounted cables.
Oh, and this isn;t a slight against SRAM, shimano's Shadow ones are the same.

supersonic

Rear mech hangers: how about looking at materials here? Thinner ti ones that can be bent back more readily? Or maybe a cheaper plastic version.

As the prices they normally are are absurd.

Anonymous

I paid a quid for my old wolfridge hanger, and they had a few of them in stock in the bike shop.
However, I don't think the dropouts for my new one will be anywhere near as cheap, sadly, although they do look as though they'll take a hell of a lot more abuse than the older ones.

(well, they're not dropouts on the new one, technically, as it's a maxle rear end, but you know what I mean)

NatoED

i find road mechs last longer than MYB ones on my XC bike i've got a 10 year old Tiagra that i used Dj and Dh before going all XC. I've not bout a MTB mech for years and neither have most of my mates. Plus they cost a hell of a lot less than MTB mechs

rhyko7

sounds like a good little project- i would be very impressed if you managed to re-design the rear derailleur mechanism.
Things I don’t like or would like on rear mechs

Better cable sealing so no contamination can get in cable
Smaller design so it sticks out less vertically and horizontally maybe not have two jockey wheels?
Obviously direct cabling
More precise shift – one click (good ones already have)
Lightweight
Stable under vibration (shimano aren’t sram are)
Strong (shimano are, sram aren’t)
Maybe an option to jump gears -more than one change_no idea how you could do that. Probably bad idea lol

That’s all I can think of

tom_goggin

cheers for all your comments guys, in particular the last post, they've definately given me some food for thought. keep the suggestions coming!

I've got an XT setup on my Patriot 7+ and have never quite got it working as I'd like, hence I thought of the project. I'm looking at upgrading to a Saint as they seem to get pretty good reviews and aren't as chunky as they used to be.

good idea with the mech hanger Supersonic, i remember forking out loads to get a replacement for mine a few years back! Small replacement parts like those are always way overpriced! I somehow managed to lose the bolt-thru axle for my Manitou Shermans when moving to and from uni a couple of years back, and had to fork out £40 only for it to turn up some months later in a really obvious place! ah!

Does anyone know which was the last MBUK to have a grouptest for shifters and derailleurs? My reading of MBUK has been pretty on/off over the last couple of years!

fourcrossjohn

g box any one?

Anonymous

Rhyko, I can;t see how a rear mech would work with only one jockey wheel.

nicklouse

tom_goggin
if you cant get the XT working right then you wont get the saint set up right.

meesterbond

The thing it, the whole derraileur concept is a pretty crude solution to the problem, but as it's (largely) all we've had for so many years the big companies have refined the back-side off it, so by my reckoning it must be getting towards the point where it's as good as it's ever going to get. Using brute force to change gears really isn't that elegant.

For me, the future is elsewhere, whether that's hub gears or g-box style things. Imagine if one of those had had the money spent on it that, say, the rear mech had...

A complete aside, and not overly relevant to the thread but thought I'd share all the same!

_Ferret_

How about developing a system to work with the upcoming beltdrives?

I would like more tension on the chain that wouldn't affect the shifting performance.
What about moving the position of the whole mech - setting it between the seat stays and drop outs and using a linear movement to change gears.
Or a super short cage version to keep the mech out of harms way?
Or a self cleaning / oiling attachment on the bottom jocky wheel (or in the cage between the jocky wheels)?

How much time have you got???

ride_whenever

whats the name of the continuously variable hub gear? That is the future without a doubt.

rhyko7

yeehaamcgee wrote:

Rhyko, I can;t see how a rear mech would work with only one jockey wheel.

nor me, i was thinking no jockey wheels, i had a vision of a chain tensioner type device-its probably not possible, but its an idea. :roll:

desweller

Electronic servo-operated hub or bottom bracket gearbox. Industry standardised shift signals so any electronic shifter can be used with any electronically actuated gearchange, whether it be hub gear or derailleur.

nicklouse

_Ferret_ wrote:

How about developing a system to work with the upcoming beltdrives?

I would like more tension on the chain that wouldn't affect the shifting performance.
What about moving the position of the whole mech - setting it between the seat stays and drop outs and using a linear movement to change gears.
Or a super short cage version to keep the mech out of harms way?
Or a self cleaning / oiling attachment on the bottom jocky wheel (or in the cage between the jocky wheels)?

How much time have you got???

more tension can effect the bearings in the hub and bb.
mech between the seat stays cant change the rear gears, can the current system is linear.
short mech will limit the number of gears (change in teeth) you can have.
thats what maintainance is for.

_Ferret_

Ideas Nick, Ideas. Not patents.

More tension can affect the bearings in the hubs and BB - only slightly though, it's more likely to affect the freehub. I'm also not expecting such a huge amount of tension that it puts that much stress on the bearings, that would be a bit OTT.

Why couldn't a mech placed near the cassette but between the seat stays and drop outs not change the rear gears? I admit it does depend on the frame design a lot but a linear actuator with a sprung swingarm for the chain is possible.

Finally - I don't like maintainance. Find me a derailier that cleans my chain while I ride and I'll be a little happier.

Daz555

Ultimately I think what mountain biking needs is the removal of the derailleur completely.

We need a mainstream solution that retains the pedaling efficiency of the classic derailleur and is not too weighty.

As for stuff to improve current derailleur - for me it has to be around my two biggest gripes with derailleurs, 1.Cable tension/stretch and, 2.How close it is to the ground, and therefore prone to being wrecked.

Hydraulics would sort the cable issue of course.

llamafarmer

meesterbond wrote:

The thing it, the whole derraileur concept is a pretty crude solution to the problem, but as it's (largely) all we've had for so many years the big companies have refined the back-side off it, so by my reckoning it must be getting towards the point where it's as good as it's ever going to get.

This - A bit like the internal combustion engine, it's a fairly crude idea that has been so highly refined it now works very well at a very low cost.

If it's a final year project you're never going to have the time and resources to improve on what SRAM and Shimano have been refining for decades.

To be brutally honest, you need to be careful that you've got a proper workable problem at the heart of your project, rather than something contrived because you love bikes. I'm not trying to piss on your campfire, but in my experience a well chosen project aim can make or break your final project. Have a good chat with your project supervisor and good luck!

What course are you doing btw?

MrChuck

meesterbond wrote:

The thing it, the whole derraileur concept is a pretty crude solution to the problem, but as it's (largely) all we've had for so many years the big companies have refined the back-side off it, so by my reckoning it must be getting towards the point where it's as good as it's ever going to get. Using brute force to change gears really isn't that elegant.

For me, the future is elsewhere, whether that's hub gears or g-box style things. Imagine if one of those had had the money spent on it that, say, the rear mech had...

A complete aside, and not overly relevant to the thread but thought I'd share all the same!

What I ws going to say really. It may seem a bit short-sighted and Luddite to say that things are as good as they're going to get and there's no need to improve them, but in the case of derailluers I think there's only room left for tweaking really, or maybe changes in materials. However Shimano and the rest will be well ahead of you, and there are probably market/economic reasons as well as techincal reasons why some tweaks you might think of haven't appeared yet.

The interesting stuff is going to be what replaces derailleurs, or maybe something to make them work with belt drives. Bear in mind that alternatives already exist (Rohloffs, NuVinci etc.) and then ask yourself why it is that they're not more common than they are now.

HTH and good luck!

joshtp

OK, heres an idea i had a few months back, its a bit wack but it could work, kind of like a bolt on hub gear, to go where the cassete is normally, so you dont need a whole new hub. it would have an infinate gear slope so any indexing would be in the shifter, or you could have a friction style shifter to give absoulutely any gear. i even did some very rough tech drawings, and asked a few boffins i know, they all reconed it would work, if you want more indepth tech details PM me, im not going to let my design be used by anyone, if it ever gets made i want credit, profit, and one for my bike.
as for what i look for in a mech/shifter..... well, damage proofing, toughness, lightness, extreemely reliable, clean, instant, totaly smooth shifts, with no noticable trasition stage. acompanied by a light click.and an easy to push/pull shifter. very smooth.

brownix

lol How about a Hydraulic shifter, would be superb

scale20

Theres a lot of good replies here and to be honest I think you will have a job on your hands with your project but best of luck to you.

Over the years the likes of Shimano and SRAM have no doubt thrown millions of pounds into mech design and the fundamentals have not really changed much they are just gettin lighter, more responsive and nicer looking.

I run Shimano on my Gary fisher and SRAM on my Pace, If I had the funds I would replace the shimano on my fisher with SRAM, I reckon they have hit it on the head when it comes to smoothness and precice shifting.

I struggle to come up with anything I would like to see on a rear mech, I think the X0 I have does the job a treat. I find the SRAM's so much easier to work on and set up.

meesterbond

I wonder how long it will be before Shimano apply their road Dura-Ace Di2 technology to MTBs... forget about cables and hydraulics completely.

Whilst I find the whole idea largely unnecessary, the idea of a system which auto-trims for perfect shifting all the time is quite appealing!

Anonymous

What's all the fuss about belt dives? What problems do you guys think that is going to solve?

As for Joshtp's

totaly smooth shifts, with no noticable trasition stage

That, has more to do with cassette design. The chain will always have to travel about half a revolution of the cassette to fully change gear. No redesign of rear mech could eliminate that.