RobbieTunes

I first noticed this with the DA7700 GS rear derailleur.

No doubt there’s math involved. A lot of pros are using larger aftermarket pulleys on the RDs, and the Wolftooth Roadlink is now being copied by Wheels Mfg and of course, the Chinese fabricators.

I like the Roadlink for the extra clearance.

I’m wondering if the larger pulleys increase chain wrap, and would the combination of both sort of equal a longer cage RD?

At my level, performance is not an issue, but it’s interesting.

nomadmax 

Bunkie

I'm not an engineer but I have stayed in a Holiday Inn I would say the chief reason larger pulleys are touted as "better" is that they reduce the angle/bend of the chain when it goes around the pulleys, thereby reducing friction. I'm sure they have some kind of fancy bearing in them that further reduces friction but I don't know that to be a fact. As far as increasing chain wrap, it would be the same as it was with smaller pulleys; only a greater distance between the pulleys (center to center) would do that.

top506

I have used larger pulleys from low end Shimano RDs (think Tourney) to increase chain wrap a link or two. Let me run a 11-32 cluster with a 53-39 front.

Top

Lemond1985

I have a couple Altus RD's that use the large pulleys. I suspect that the design allows the quick snappy shifts you get with a short cage (vs long cage) RD, but you get the extra chain wrap of a long cage RD. Brilliant idea, and I think it will catch on gradually, as more people run 32 and 34 tooth cassettes on their road bikes.

The Altus RD's are pretty heavy though, so I hope we start seeing the large pulleys on some higher end, lighter RD's.

hokiefyd 

I'm not quite sure what the disadvantages of these larger pulleys are. I have a number of lower end derailleurs that have 13- and 15-tooth pulleys and I also have a number of more mid-tier derailleurs that have 9- and 11-tooth pulleys. I have swapped them before and there don't seem to be any negative consequences. Indeed, the drivetrain just seems to operate "smoother" with the larger pulleys (and perhaps even quieter).

Shimano put a 13- and 15-tooth combo on their lower end RD-M310 (Altus) many years ago. Grant Peterson loves it (or at least claims to). Chain wrap is increased in a smaller package (this makes sense -- a sinuous path is longer than a straight path). This pulley arrangement (large top pulley, even larger bottom pulley) is what many of the current race teams use. I find it really interesting that Shimano tried this on an Altus derailleur and...then...nothing. It hasn't escalated through the design chain, and it doesn't even exist anymore in the current Altus lineup (the current Altus RD-M2000 uses conventional 11-tooth pulleys).

My guess is it really does make little-to-no real world difference. If one design approach was clearly superior, then wouldn't all derailleurs be using it...or at least the more expensive ones? Why'd Shimano seemingly try this on an entry level derailleur and then apparently abandon it?

jamesdak 

Correct me if I'm wrong but with a larger pulley the distance the chain has to pass around each pulley is longer. So it seems it should increase chain wrap by a link or two for sure.

nomadmax 

The larger pulleys do need more chain length for the same gearing set up as the smaller pulleys, BUT they do not increase chain wrap/total capacity. Only a greater pulley distance center to center (ie a longer cage) will increase chain wrap/total capacity. The additional chain needed for the larger pulleys is used up going around the larger pulleys, ie the chain is functionally the same length as the smaller pulley set up. In fact, a larger pulley may decrease total cog capacity because the larger top pulley will come in contact with the cogs sooner. I would think that a Road Link could mitigate that but it won't increase capacity either, it just places the top pulley further from the cogs for interference sake.

My guess is, the selling point is that the chain doesn't travel as torturous a route as it does with smaller pulleys, ie doesn't bend as much going around them so it provides less friction loss.

Lemond1985

They will bring it back at some point.

I remember assembling some Adamas Ax dept store bikes at one of my first jobs, back in '81 or so. They used octalink BB's (which were abandoned, then came back years later in the 9 speed groups) as well as brake cables under the bar tape, which was also abandoned and then re-adopted on all groupsets 3-4 years later.


That groupset also contained a lot of stuff that was not re-adopted though, like rubber padded turkey levers, and those weird aero brake calipers that kind of resembled Campy Deltas.

bulgie 

Not quite true. Think about the limit at the Big/Big combo, a gear we try not to use much, but it should shift into without breaking anything. When you're straining the wrap capacity of the mech, in big-big the chain is almost straight, not wrapping around the pulleys much if any. So that's why big pulleys increase total chain wrap, because you're comparing the wrapped state in small-small to an almost straight chain.

Another reason the big pulleys reduce friction loss is that they turn more slowly. Whatever superduper bearings you put in 'em, you can use those same bearings in a small or large pulley, so it's not a question of bearing quality. The bigger radius does make each link go through a smaller angle change though, as you said, so you pick up a tiny friction decrease there too. The total friction change from both sources is small but it adds up on a long ride or a performance-critical one like time trial.

Downside is pretty much just weight, and maybe looking funny, but otherwise it's all upside, unless I'm missing something.

Mark B in Seattle

nomadmax 

I'm standing fast on it won't increase total capacity.

hokiefyd 

I think this post may be conflating chain wrap with largest sprocket supported.

Chain wrap: larger pulleys will increase chain wrap. Imagine a pulley with a very small radius. Or no pulley at all -- just the pin bolt itself. The chain's path through the cage would essentially be linear. It wraps around the lower pin, travels straight up to the upper pin, and then wraps around that to join the freewheel or cassette. Now add pulleys. A standard 11-tooth pulley would increase the wrap because the line would have a slight S-shape to it. It would no longer be exactly linear. Now change the 11-tooth pulleys to 15-tooth pulleys. The path the chain takes becomes more sinuous -- and longer! The larger the pulleys, up to the point where they conflict with each other in the cage, the more chain wrap capacity is available.

The more horizontal offset you have to the chain line through the derailleur cage, the more chain wrap capacity you have (assuming no other changes to the cage design).

Largest sprocket supported: a larger top pulley would indeed reduce the size of the largest sprocket supported. This is usually able to be overtime (within reason) with a B-screw adjustment.

beicster 

I think the rear derailleurs look pretty funny already these days.

conspiratemus1

The larger pulleys would require the chain to be longer (by the same number of half-links as there are extra teeth on one pulley.). But the amount of chain wrap, the important thing for capacity , is dictated by the straight-line distance between the pulleys, as a previous poster pointed out. It’s the distance back that the lower pulley can move as you shift toward small-small that matters, and this is independent of pulley size.

A derailer that had no pulleys, just pins that chain rode around on, would wrap chain in exactly the same manner that real derailers do. The pulleys are just there to provide a low-friction riding surface for the chain. Now, there may well be mechanical benefits from larger pulleys, sure, but chain wrap isn’t one of them.

nomadmax 

What he said.

jamesdak 

Or is it what post #11 said?

hokiefyd 

This is one way to influence chain wrap. A long cage using 11-tooth pulleys will have more chain wrap capacity than a short cage using 11-tooth pulleys. Another way to influence chain wrap is to re-size the pulleys. A long cage using 15-tooth pulleys will have more chain wrap capacity than a long cage using 11-tooth pulleys.



Precisely! This demonstrates the fact that larger pulleys are a way to increase chain wrap capacity. Why does the chain need to be made longer with larger derailleur pulleys? You didn't change the cassette or freewheel. You didn't change the chain rings. You didn't lengthen the chain stays. No, what you did is increase the chain wrap of the derailleur. It consumes more chain, which effectively "shortens the chain" (requiring a longer one).

natterberry 

How fast do I go with aftermarket 13T aluminum pulleys?

hokiefyd 

I made a little model using a derailleur cage and some pulleys I have laying around. I traced an outline of the cage and entering and exiting chain line, and marked a "start" and "end" point from which to measure consistently. I think we all agree that a derailleur cage's capacity to consume links of chain (regardless of its pulley size or cage length), and then give it back to the drivetrain, is the means by which it accomodates chain wrap. This model uses the SAME cage, but varies pulley size.

Picture 1: 10-tooth pullies. From the starting mark (which starts an outer link), this model consumes 7 complete outer links, and starts into the 8th outer link at the end mark.



Picture 2: 13-tooth pulleys. This model consumes about 8.5 outer links.



Picture 3: one 13-tooth pulley and one 15-tooth pulley. This model consumes 9 outer links and is starting on its 10th outer link at the end mark.



Changing pulley size is one way to consume extra links of chain in the derailleur. The more links of chain a derailleur consumes, the more chain wrap capacity it has.

hokiefyd 

To follow up, the 10/10 model physically consumes about 7 inches of chain (that's 7 outer links and 7 inner links...almost starting on the 8th outer link). The 13/15 model physically consumes about 9 inches of chain.

Say, for example, a shift from the small ring to the large ring requires an additional 1.5 inches of chain (just to make up a number). That's 21% of the 10/10 model's capacity, and it would have to swing forward by some angular amount. However, 1.5 inches is only 16% of the 13/15 model's capacity. Therefore, it wouldn't have to swing as far forward as the 10/10 model would for the same shift. It's giving a smaller percentage of its total chain storage back to the drivetrain.

So while a shorter cage derailleur can't move its lower pulley forward and rearward as far as a longer cage derailleur does, with larger pulleys, it doesn't have to. Its angular movements are smaller for the same chain wrap requirements of the drivetrain during a shift. This is how a shorter cage derailleur can have a similar chain wrap capacity as a longer cage derailleur -- larger pulley size.

conspiratemus1

The chain is longer with big pulleys, just as it is with big chainrings and big sprockets, but it doesn’t wrap more. Wrapping implies a shortening of the effective length of the chain, a change in the length as the lower pulley moves back (and as the body rotates on the upper pivot bolt). Big pulleys don’t affect the way the chain length changes during operation of the derailer, and so they don’t affect capacity.

Going to a 60-tooth front and a 32 rear requires a longer chain, but you don’t get more wrap capacity just because the chain is longer. You have to increase the cage length, or get more pivot room at the B-bolt. Now if you could somehow change the diameter of the pulleys as you shifted, then, yes, you would increase wrap, because as the pulleys expanded, they would take up slack in the chain at the wrap limit. But you can’t, so they don’t.

This is not the equivalent of arguing about how many angels can dance on the head of a pin. It’s testable. Buy a couple of big pulleys, stick them in your der. cage, add the requisite link or 2 to the chain, and measure to see if you get more wrap capacity. And report what you find.

Edit: Your post with the thoughtful and thought-provoking mock-up appeared while I was typing my post and so I’m seeing it only now. I’ll study it carefully. Promise.

jamesdak 

Yep, it's that time folks.

conspiratemus1

The arrival of Popcorn Man means it’s time to go do something else. I’m taking the hint.
Thx.

hokiefyd 

I think this is a good techincal discussion over what's really happening in a bike drivetrain. Nobody's been insulting or condescending. I really do think that changing pulley sizes and cage length are both ways that derailleur engineers can increase or decrease chain storage in the derailleur, and I think my pictorial model has demonstrated that. But, if I've misinterpreted something, I hope we can keep the discussion going.

I'm here to learn!

RobbieTunes

Hold it, there. I've not insulted anyone yet, real or imagined.

But, I am watching the TdF. The panel addressed the larger pulleys and said "half a watt."
Now, those Ceramic Speed pulleys are selling for $600.

Other versions are as low as $60. Or the price of a nice used Ultegra or Centaur long-cage RD.

There are also the carbon and Ti custom cages coupled with the oversize ceramic-bearing, double-stitched for extra cushioning, kung-fu grip pulleys, and from what I understand, the bike freakin' rides itself.

I'm actually thinking about selling one bike and then putting all that money into aftermarket gadgets on another bike, just to see.

I did pick up on the long-cage = smoother, short cage=quicker opinion.
I kind of agree, but I'm not sure if I'm expecting it and being confirmed or it's accidentally getting my attention and I just rationalize it.
I just know that few things shift quicker than a 1-cog jump on a Dura Ace short cage 11-sp setup.
It's just more fun with a 9-sp because it's shiny AND old and there's something to that cable routing, I swear it.

I'll experiment with an SRAM Red short-cage 10 vs an SRAM X9 clutched medium-cage.
My guess is the Red with a Roadlink will be quicker. And a lot lighter.

droppedandlost 

Do you size the chain differently with larger pulleys? Big-big +2 seems RD independent. Seems like a bit more wrap to me.