2X chain rings > 16 teeth apart?
01-28-21, 03:22 PM
#1
Curmudgeon
Thread Starter
Join Date: Nov 2018
Location: Triangle NC
Posts: 336
Bikes: Specialized Diverge Comp
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 157 Post(s)
Liked 32 Times
in
28 Posts
2X chain rings > 16 teeth apart?
I'm thinking of replacing my Shimano Ultegra 52/36 with Shimano Ultegra 50/34. 16 tooth "spread" seems to always be the mechanical max. Or could I replace only the small ring and go 52/34?
What is the reason for 16 tooth max spread?
Derailleur length?
What is the reason for 16 tooth max spread?
Derailleur length?
01-28-21, 03:51 PM
#2
Senior Member
Join Date: Aug 2009
Posts: 2,394
Mentioned: 14 Post(s)
Tagged: 0 Thread(s)
Quoted: 1562 Post(s)
Liked 1,734 Times
in
974 Posts
Yes. Rear Derailleurs are typically spec'ed for max front chain ring difference. The other issue is that when there is too much of a difference between chain ring sizes, the chain can rub on the large chain ring when you are riding on small-small. My son's bike does this. Granted his bike has 650c wheels and a shorter chain stay than the typical 700c wheeled bikes that these groupsets are designed for.
Likes For icemilkcoffee:
01-28-21, 04:20 PM
#3
Really Old Senior Member
If you look at the recommended cassettes for that Gruppo and take the worst case scenario RE: chain wrap, the 16T (or other) is about all the wrap capacity you have left.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
Likes For Bill Kapaun:
01-28-21, 04:26 PM
#4
Senior Member
You can always try it, but the shifting ramps won't line up properly and the compensating shift at the cassette might go from two to three sprockets. My guess is it will shift poorly.
Last edited by DaveSSS; 01-29-21 at 10:20 AM.
Likes For DaveSSS:
01-28-21, 04:29 PM
#5
Senior Member
Join Date: Jul 2002
Location: Sacramento, California, USA
Posts: 40,865
Bikes: Specialized Tarmac, Canyon Exceed, Specialized Transition, Ellsworth Roots, Ridley Excalibur
Mentioned: 68 Post(s)
Tagged: 0 Thread(s)
Quoted: 2952 Post(s)
Liked 3,106 Times
in
1,417 Posts
If you look at the recommended cassettes for that Gruppo and take the worst case scenario RE: chain wrap, the 16T (or other) is about all the wrap capacity you have left.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
Likes For caloso:
01-28-21, 04:43 PM
#6
Senior Member
The wider the shift, the harder it is to guarantee consistent smooth mechanical operation. Even when it's smooth, a super-large gear ratio change can be annoying: the bigger the change, the more rear shifts you need to make to compensate for it.
Also, modern front derailleurs use fancy shaping on their cages to optimize shift quality, and this shaping makes certain assumptions about what the shifts look like. This is why front derailleurs for triples look different from front derailleurs for doubles. Most modern 2x road cranksets have tooth changes in the mid-teens between big and small rings, so the front derailleurs are optimized around this.
A big difference between small and big ring can also mean that the big ring is "in the way" of the chain if you're on the small ring and trying to use the smaller cogs. (How severe this is depends on the exact chainring sizes and the chainline).
You also need to keep sprocket matching in mind. The pins and ramps on modern cogs and chainrings are designed to be paired with specific other cogs and chainrings in order to work properly. When you mix-and-match rings that aren't designed to go with each other, sometimes it works fine, but sometimes the shifting is far worse than on a properly-matched setup.
This being said, there's nothing inherently impossible about a bigger jump. Quite a few mid-century French bikes had ~20-tooth jumps between the big and small ring, for instance. They made them work.
If you want a wide-range setup for a modern drivetrain, look into Wickwerks. They make a 53-34 pair, and it's reputed to shift quite well.
Also, modern front derailleurs use fancy shaping on their cages to optimize shift quality, and this shaping makes certain assumptions about what the shifts look like. This is why front derailleurs for triples look different from front derailleurs for doubles. Most modern 2x road cranksets have tooth changes in the mid-teens between big and small rings, so the front derailleurs are optimized around this.
A big difference between small and big ring can also mean that the big ring is "in the way" of the chain if you're on the small ring and trying to use the smaller cogs. (How severe this is depends on the exact chainring sizes and the chainline).
You also need to keep sprocket matching in mind. The pins and ramps on modern cogs and chainrings are designed to be paired with specific other cogs and chainrings in order to work properly. When you mix-and-match rings that aren't designed to go with each other, sometimes it works fine, but sometimes the shifting is far worse than on a properly-matched setup.
This being said, there's nothing inherently impossible about a bigger jump. Quite a few mid-century French bikes had ~20-tooth jumps between the big and small ring, for instance. They made them work.
If you want a wide-range setup for a modern drivetrain, look into Wickwerks. They make a 53-34 pair, and it's reputed to shift quite well.
Likes For HTupolev:
01-31-21, 11:48 AM
#7
Senior Member
Join Date: Oct 2017
Location: Chapel Hill NC
Posts: 1,683
Bikes: 2000 Litespeed Vortex Chorus 10, 1995 DeBernardi Cromor S/S
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 645 Post(s)
Liked 797 Times
in
446 Posts
ive been using TA big rings (50t and 51t) with Campag 39t smalls for years. They’re pinned and ramped, but they’re not “matched” with any specific small ring, nor are they marked for any specific orientation wrt the crank or the small ring. Front shifting has always been completely reliable
Likes For Litespud:
01-31-21, 02:09 PM
#8
Senior Member
The incoming chain. In order to do their job optimally, pin-and-ramp shift gates rely on the chain coming from a certain angle, and also on which part of the link contacts which part of the shift gate. The teeth on a chainring usually aren't aligned to each bolt hole in exactly the same way, so not only does the size of the small ring matter, its orientation on the crank spider can affect engagement with shift gates as well. This is why both chainrings in matched pairs often have a "timing" indicator mark to show which orientation to install them in.
Where this gets a little complicated is that there are two types of links in a chain, and thus two ways for a chain to sit on a chainring. Whether a pin passes by an inner or an other link matters. Rene Herse even argues that the interaction between shift gate design and chain alignment can affect which chainring combos are optimal to design shift gates around, and uses this to explain why they offer a 48-33 instead of a 48-32 when their other combinations use a 16-tooth jump. (I'm not sure that Heine's explanation is entirely complete, but I did find it interesting that when Shimano's GRX announcement appeared a few weeks later, Shimano had also avoided 48-32 and offered a 48-31 instead.)
Sometimes things "just work", especially when you're talking about fairly narrow shifts. Like, racing cranks from the 1970s with unramped chainrings tended to shift great, but they were also usually 52-42 or thereabouts.
It's not that every setup needs shift gates in a particular placement and orientation. It's that by having shift gates in a particular placement and orientation, you can guarantee great shifts, and you can do it even for fairly wide shifts that normally create big issues for front shift quality.
//==========================
Actually, I'd go as far as to say that mismatched pins and ramps can sometimes make things a bit worse.
For instance, I recently replaced the unramped 38T middle ring on my gravel bike with a ramped 38T chainring from TA. Previously, if I wasn't aggressive enough with an upshift from the small ring, the chain would just make noise rubbing on the side of the middle ring. But now, the pins can start catching-and-releasing the chain, causing the drivetrain to repeatedly jolt. Obviously this is avoidable with good technique, but it is a more annoying failure mode than the old chainring had.
Where this gets a little complicated is that there are two types of links in a chain, and thus two ways for a chain to sit on a chainring. Whether a pin passes by an inner or an other link matters. Rene Herse even argues that the interaction between shift gate design and chain alignment can affect which chainring combos are optimal to design shift gates around, and uses this to explain why they offer a 48-33 instead of a 48-32 when their other combinations use a 16-tooth jump. (I'm not sure that Heine's explanation is entirely complete, but I did find it interesting that when Shimano's GRX announcement appeared a few weeks later, Shimano had also avoided 48-32 and offered a 48-31 instead.)
ive been using TA big rings (50t and 51t) with Campag 39t smalls for years. They’re pinned and ramped, but they’re not “matched” with any specific small ring, nor are they marked for any specific orientation wrt the crank or the small ring. Front shifting has always been completely reliable
Sometimes things "just work", especially when you're talking about fairly narrow shifts. Like, racing cranks from the 1970s with unramped chainrings tended to shift great, but they were also usually 52-42 or thereabouts.
It's not that every setup needs shift gates in a particular placement and orientation. It's that by having shift gates in a particular placement and orientation, you can guarantee great shifts, and you can do it even for fairly wide shifts that normally create big issues for front shift quality.
//==========================
Actually, I'd go as far as to say that mismatched pins and ramps can sometimes make things a bit worse.
For instance, I recently replaced the unramped 38T middle ring on my gravel bike with a ramped 38T chainring from TA. Previously, if I wasn't aggressive enough with an upshift from the small ring, the chain would just make noise rubbing on the side of the middle ring. But now, the pins can start catching-and-releasing the chain, causing the drivetrain to repeatedly jolt. Obviously this is avoidable with good technique, but it is a more annoying failure mode than the old chainring had.
01-31-21, 05:19 PM
#9
Senior Member
Join Date: Oct 2017
Location: Chapel Hill NC
Posts: 1,683
Bikes: 2000 Litespeed Vortex Chorus 10, 1995 DeBernardi Cromor S/S
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 645 Post(s)
Liked 797 Times
in
446 Posts
The incoming chain. In order to do their job optimally, pin-and-ramp shift gates rely on the chain coming from a certain angle, and also on which part of the link contacts which part of the shift gate. The teeth on a chainring usually aren't aligned to each bolt hole in exactly the same way, so not only does the size of the small ring matter, its orientation on the crank spider can affect engagement with shift gates as well. This is why both chainrings in matched pairs often have a "timing" indicator mark to show which orientation to install them in.
Where this gets a little complicated is that there are two types of links in a chain, and thus two ways for a chain to sit on a chainring. Whether a pin passes by an inner or an other link matters. Rene Herse even argues that the interaction between shift gate design and chain alignment can affect which chainring combos are optimal to design shift gates around, and uses this to explain why they offer a 48-33 instead of a 48-32 when their other combinations use a 16-tooth jump. (I'm not sure that Heine's explanation is entirely complete, but I did find it interesting that when Shimano's GRX announcement appeared a few weeks later, Shimano had also avoided 48-32 and offered a 48-31 instead.)
Sometimes things "just work", especially when you're talking about fairly narrow shifts. Like, racing cranks from the 1970s with unramped chainrings tended to shift great, but they were also usually 52-42 or thereabouts.
It's not that every setup needs shift gates in a particular placement and orientation. It's that by having shift gates in a particular placement and orientation, you can guarantee great shifts, and you can do it even for fairly wide shifts that normally create big issues for front shift quality.
//==========================
Actually, I'd go as far as to say that mismatched pins and ramps can sometimes make things a bit worse.
For instance, I recently replaced the unramped 38T middle ring on my gravel bike with a ramped 38T chainring from TA. Previously, if I wasn't aggressive enough with an upshift from the small ring, the chain would just make noise rubbing on the side of the middle ring. But now, the pins can start catching-and-releasing the chain, causing the drivetrain to repeatedly jolt. Obviously this is avoidable with good technique, but it is a more annoying failure mode than the old chainring had.
Where this gets a little complicated is that there are two types of links in a chain, and thus two ways for a chain to sit on a chainring. Whether a pin passes by an inner or an other link matters. Rene Herse even argues that the interaction between shift gate design and chain alignment can affect which chainring combos are optimal to design shift gates around, and uses this to explain why they offer a 48-33 instead of a 48-32 when their other combinations use a 16-tooth jump. (I'm not sure that Heine's explanation is entirely complete, but I did find it interesting that when Shimano's GRX announcement appeared a few weeks later, Shimano had also avoided 48-32 and offered a 48-31 instead.)
Sometimes things "just work", especially when you're talking about fairly narrow shifts. Like, racing cranks from the 1970s with unramped chainrings tended to shift great, but they were also usually 52-42 or thereabouts.
It's not that every setup needs shift gates in a particular placement and orientation. It's that by having shift gates in a particular placement and orientation, you can guarantee great shifts, and you can do it even for fairly wide shifts that normally create big issues for front shift quality.
//==========================
Actually, I'd go as far as to say that mismatched pins and ramps can sometimes make things a bit worse.
For instance, I recently replaced the unramped 38T middle ring on my gravel bike with a ramped 38T chainring from TA. Previously, if I wasn't aggressive enough with an upshift from the small ring, the chain would just make noise rubbing on the side of the middle ring. But now, the pins can start catching-and-releasing the chain, causing the drivetrain to repeatedly jolt. Obviously this is avoidable with good technique, but it is a more annoying failure mode than the old chainring had.
Likes For Litespud:
02-01-21, 03:17 PM
#10
Senior Member
Join Date: Jul 2006
Location: San Jose (Willow Glen) Ca
Posts: 9,846
Bikes: Kirk Custom JK Special, '84 Team Miyata,(dura ace old school) 80?? SR Semi-Pro 600 Arabesque
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 2338 Post(s)
Liked 2,822 Times
in
1,541 Posts
If you look at the recommended cassettes for that Gruppo and take the worst case scenario RE: chain wrap, the 16T (or other) is about all the wrap capacity you have left.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
Add the fact that a 16T jump is getting pretty large. Any more than that starts to result in much crappier shifting.
__________________
Life is too short not to ride the best bike you have, as much as you can
(looking for Torpado Super light frame/fork or for Raleigh International frame fork 58cm)
Life is too short not to ride the best bike you have, as much as you can
(looking for Torpado Super light frame/fork or for Raleigh International frame fork 58cm)
Likes For squirtdad:
02-01-21, 03:45 PM
#11
Really Old Senior Member
not 100% current shimano rear derailers have a number of specs for ranges. Mininum read cog, Maximum Rear cog, Chain wrap and maximum front ring difference. It is entirely possible to be within the chain warp spec but out of front ring difference. Shimano seems pretty clear about no more that 16 front difference in that they don't sell current cransets with greater than 16 t difference.
Likes For Bill Kapaun:
02-01-21, 04:04 PM
#12
Senior Member
Join Date: Jul 2006
Location: San Jose (Willow Glen) Ca
Posts: 9,846
Bikes: Kirk Custom JK Special, '84 Team Miyata,(dura ace old school) 80?? SR Semi-Pro 600 Arabesque
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 2338 Post(s)
Liked 2,822 Times
in
1,541 Posts
using current ultegra medium cage..... speed at 39 chaing wrap, 16 t max front and 11/34 casset (rear min/rear max) you get 39 teeth so zero left.....
my point was the 16 t and the chain warp are independent specs. you could do an 11/28 and a 52/34 and be under wrap at 35 but out of speck at 18 up front a
my point was the 16 t and the chain warp are independent specs. you could do an 11/28 and a 52/34 and be under wrap at 35 but out of speck at 18 up front a
__________________
Life is too short not to ride the best bike you have, as much as you can
(looking for Torpado Super light frame/fork or for Raleigh International frame fork 58cm)
Life is too short not to ride the best bike you have, as much as you can
(looking for Torpado Super light frame/fork or for Raleigh International frame fork 58cm)
Likes For squirtdad:
02-01-21, 04:12 PM
#13
Really Old Senior Member
using current ultegra medium cage..... speed at 39 chaing wrap, 16 t max front and 11/34 casset (rear min/rear max) you get 39 teeth so zero left.....
my point was the 16 t and the chain warp are independent specs. you could do an 11/28 and a 52/34 and be under wrap at 35 but out of speck at 18 up front a
my point was the 16 t and the chain warp are independent specs. you could do an 11/28 and a 52/34 and be under wrap at 35 but out of speck at 18 up front a
I know you'll find a few exceptions (with these RDER's designed for 1X systems with extremely large cogs etc. who knows?) you'll want to use, but my "method" simply matches up too well, too often to ignore.
This is just a waste of time, so I'm done.
Likes For Bill Kapaun:
02-02-21, 09:18 AM
#14
Senior Member
To be more accurate, every RD must have 2-3T of additional wrap to handle any possible chain stay length. If you apply the rigorous formula on the park tool website, you'll find that the needed length for a bike might come out to 53.25 inches, so a 54 inch chain is needed and the rest is excess that the RD cage must still tension in the little/little combo. Some other chain stay length might require 53.90 inches, so a 54 inch chains still works. My current setup could use a 54 inch chain, but it's awfully tight in the big/big so I use a 55 inch chain, that's a little too loose in the little/little that I never use.
Manufacturers often don't list a maximum wrap capacity, but they do list a specific model to be used with specific cassettes. Sram claims that my force axs RD won't work with a 10-36 cassette, but it does. I also use a 16T difference at the crank, that they also don't support. In total, I have 6T more wrap than recommended.
Manufacturers often don't list a maximum wrap capacity, but they do list a specific model to be used with specific cassettes. Sram claims that my force axs RD won't work with a 10-36 cassette, but it does. I also use a 16T difference at the crank, that they also don't support. In total, I have 6T more wrap than recommended.
Likes For DaveSSS:
