twocicle

I'm using a Specialites TA Alize 48T chainring in conjunction with a 10spd Ultegra Di2 6770 front derailleur and Ultegra 10spd chain. The Alize is the "big ring" outer.

A problem I'm having is with the upshifts from the inner to outer, where the chain frequently rides up on the Alize teeth and does not quickly settle into the valleys between the teeth. This "walking the teeth" has caused the chain to derail to the outside.

Looking at the Alize chainring I observed there is no variation in the teeth profiles, other than one single mid-height tooth around the entire circumference. On other outer position chainrings I have on hand (ie: FSA for example) those teeth of tall, medium and low height profile.

Question 1:
I am wondering if filing down the heights of some teeth on the Alize would help the chain to drop into the proper position better and aleviate it from walking along the tops of the teeth?

Question 2:
Also, I looked at my 53T and 52T FSA Super Road chainrings for examples of where to place tooth height variations in relationship to the chainring pins, but it seems a bit random to me. Perhaps one pattern is to reduce the height of the teeth immediately above and after (in the direction of power rotation) a pin location. Is there any guide or wisedom to creating a pattern for teeth profiles?

cny-bikeman

You are proposing the wrong solution. Chainrings used to have all teeth the same height/profile and chains shifted perfectly well. The problem you are having could be from chain or chainwheel wear, excess pedal pressure when shifting or from derailleur adjustment or alignment problems.

fietsbob

As the chain has to climb up to the bigger chainring, why can't you let up on the pedal pressure for a Moment ..so the chain is less strained,
and not have to modify the chainring ?

old guy, not in that big of a hurry ..

2_i 

Walking to the outside should not happen for properly adjusted top limit screw on derailleur. There are different things that can be done to improve shifting. Given the difficulty, I presume that are no ramps or pins there. The first is to try and rotate the outer ring on the crankset relative to the inner. The chain should have at least 2 positions where it can easily climb onto the outer ring, settling right away properly onto the teeth of the outer, while progressing from the teeth of the inner. Eventually around those positions you can work with the teeth profiling.

twocicle

Appreciate the efforts to lend hand, but unfortunately I believe most of the above answers are based on incorrect assumptions.

• First of all, this is occurring with a new chainring and chain.
• Second, I can reproduce the issue on a workstand - little to no pressure.
• Third, the chain climbs to the outer ring without any problem. Actually very happy with that performance. Shift pins do exist and seem to pick up the chain w/o problem.
• Fourth, this is not an issue with the FD adjustment. I have studied the FD action and confirmed it is not over shifting beyond expected, plus the limit screws are set to minimize the range.

I'm not sure if rotating the outer ring would achieve anything because (if I'm correct) the relative position to the inner ring teeth would stay the same due to the symmetrical 5-arm 130bcd arm spacing. There is one thing to this configuration out of the norm... using a 30T granny to 48T big ring. Yeah I know that 18T diff exceeds the normal Di2 FD rating, but I do not believe it is the issue here. Perhaps it is caused by the chain run distance being slightly off between these two rings.

The chain gets up onto the tops of the Alize teeth and in some cases that will cause an overshift, but not always. The main problem to solve is to figure out how to get the chain to drop into the teeth instead of riding on top of the tips.

Chris_W

Take a look at the original ring that came on that crank. How do the teeth line up laterally compared to the crank's mounting tabs? When replacing the stock rings on some cranks with rings from other brands, I've ended up with rings that were too close together or too far apart, which can cause shifting problems. I try to solve this by adding spacers between the tabs and the ring or dremmeling away some of the material on the ring where it mounts.

TA rings are great for giving you more gearing options, but unfortunately they'll never impress you with their shifting quality, longevity, or stiffness. As well as having many TA rings, I have one pair of rings by Stronglight (33/48, 110 BCD) with their special CT2 finish, which they say should improve durability. I've been very happy with those in terms of shifting and durability so far, so you could give those a try if you don't have any luck with the TAs.

Al1943

I had the same problem with my wife's TA chainrings. The chain would "skate" on the tips of the middle chainring teeth instead of dropping over. I cured the problem by filing the inside edge of the teeth with a flat file. The aluminum these rings are made of is soft and files down quickly. It won't take a lot of work to thin them, and the strength of the teeth where the chain runs will not be compromised.
I also moved the chainline outward 1.5 mm with a bottom bracket spacer, but I think filing the teeth was the primary cure.

Ludkeh

You seem to discount the large jump from 30T granny to 48T big ring as being a problem. But, I think that's your problem. The 18 tooth jump is asking a lot from your setup.

I tried to do something similar with my triple, going from a 26T small ring to a 42T middle ring. Could not get it to work no matter what I did. Replaced the 42T middle with a TA 38T middle, problem solved. Shifts became smooth as silk!

twocicle

Barring any real tech info regarding the engineering behind stepping the height of ring teeth, I gave it a shot using as examples the FSA Super Road 53T and 52T, and a middle 39T from a triple. All these had pins and varing tooth heights. So with a fine file in hand I set out to profile (file down the height of) some teeth in the vicinity of where the pins are located.

Result... seems to be working. Shifting up from the granny is a touch less quick, but otherwise the chain is no longer "skating" on the teeth. Will need more time & riding to see how this goes. If this is indeed the solution, then it baffles me why TA does not review the tooth profiling of some other manufacturers.

Al1943

I didn't shorten any teeth, I just filed the sides of the tips.

twocicle

Nuts. On a ride last night had another case of chain overboard (chain tip-toed over the tips of the teeth and dropped beyond the big ring). I'll do a bit more filing (do the sides per Al1943) but otherwise shopping for a different ring that is profiled more to my liking - and hope that works ok.

BTW, in spite of having the Di2 FD outward movement limited to just enough to not have chain rub, the chain drop seems to occur only when shifting to the big ring while the inner cogs of the cassette are in use. Perhaps a factor of this problem is chain angle of attack when moving to the big ring. Problem does not appear to occur when using the outer half of the cassette.

MileHighMark

I've used 46 and 48t Alize/Zephyr outer chainrings with various shifters, and in all cases, the less-optimal 50t stock rings shifted noticeably better. Is the 48t outer ring a must-have?

Al1943

The high limit screw should keep the chain from dropping to the outside of the big ring if the front derailleur is low enough. It should be as low as possible without actually binding against the chain.
It would be best to make the shift up to the big ring with the chain running on the outside hailf of the cassette.

twocicle

Yes, limits screws are set to the minimum - just far enough to eliminate rubbing the chain. That is not stopping the chain from catching on top of the ring teeth and walking off the outside. This "over shifting" seems to be only occurring when the chain angle is greatest (inner half on cassette, then shifting from granny to big ring) and when the chain is skating on top of the teeth. When the chain drops properly into the teeth valleys there is no problem.

I've ordered some Stronglight rings from xxcycles.com to try out next. Had thought of grabbing a FSA Pro Road 48T outer ring which is supposed to be ramped and pinned (unlike the TA which is only pinned), but I opted to get a selection of Stronglight rings in a few sizes... 50,48T - CT2 version of big ring and 30,32,34T grannies.

Bingo Blingo

I had a similar skating issue with Stronglight rings that was caused by my not matching the rings up properly. Make sure that the indicator for alignment (if there is one) lines up for both rings. If there is no indicator (most likely), match the branding imprints so that each ring visually repeats its paired ring.

twocicle

^^^ yes, agree. I do have the rings installed accordingly.

This post was provided by an aquaintance in another thread...
^^^ seems to concur with what I am finding using the Di2 FD. In the scenario I described (inner rear cogs and upshifting to big ring), the FD appears to move outward more aggressively than when using the outer rear cogs (with Di2, the FD trim is automatically tied to the RD cog position). I have the FD limit screws set to the minimum needed to prevent chain rub, so going with other rings.

twocicle

EGAD! I just realized the problem only started after installing a new Ultegra chain. I had ridden for over a month with these TA rings and a used chain before the issue began with the new chain.

Since new chains are usually laterally stiffer used chains, this explains why the shifting issue occurs when upshifting while cross-chained... the new chain maintains its lateral angle of attack and does not settle into the TA teeth properly.

I guess one could say a workaround is to use an older chain, but that would be just silly

Maybe I'll try filing some TA teeth on an angle as I see on some FSA and Shimano rings I have on hand. Makes sense that the tooth angle/ramps would help mesh with a chain when its angle of attack is greater.

2_i 

This is a potential waste of rings and teeth and no, it is generally not about aligning brand names or signs. The latter would have been true if the rings were intended to work as a particular set. You need to rotate the rings using the fact that the number of teeth is usually not a multiple of arms. You rotate one ring relative to other, with up to 4-5 orientations to choose depending on arm number, to find one where the chain has good positions to climb from the smaller onto the larger ring. If it cannot climb it will skate as it does, and it will do so with any other misoriented ring no matter what the brand or teeth shape.

twocicle

I did some more "tooth profiling/filing" and testing last night on the workstand. The chain seems to climb ok, but when shifted at the sharp angles noted above (ie: when cross-chained in the 1st-4th rear cogs), the chain seems to snag oddly on the big ring for the first link or two, then the rest of the links are hung up on the tops of the teeth. It is not really "skating" but hanging up on the teeth. The further inward the cog in use, the worse/more frequent the snag happens. When shifting while using the outer half of the cassette cogs the chain angle is less and the problem never occurs.

Al1943

Is it possible that the FD is too low?

twocicle

Eureka! Chainring Choice or Shifting Ramps? | Off The Beaten Path

Exactly the issues I was describing and how they addressed the problem (though a bit vague).

Plus, because my setup has a slightly wider spacing between the granny & big ring (so the chain clears the 130bcd chainring bolts) I believe that exacerbates the issue by causing even more diagonal skew than in the "bad shift" pictured below.

I'll need to dig deeper into how they profiled their teeth to solve the issue...

The issue I am seeing is this (image from link above):

hueyhoolihan

i think the, as stated previously, that the new chain may be the root cause.
or maybe just the straw that broke the camel's back.

look closely from behind at a chain and examine how and where it flexes, laterally. if on the big, big combo, the chain will lie on the big cog, then bend to the right immediate after leaving the last tooth. it will then remain straight until it reaches the first tooth of the chainring, then it will have to bend back left again so as to lie in line with the rest of the teeth on the chainring. as the distance between the rear cog and the chainring decreases, this becomes more and more difficult. eventually it will refuse to go over the teeth on the chainring. there are many factors involved, including chainstay length, size of rear cog, size of chainring and flexibility of the chain. probably chainring tooth profile too.

anyway, you may have a "perfect storm" of short chainstays, too big cog, minimally beveled and too big chainring, stiff chain and poor chainline. i might try putting on the old chain just to see if it improves things a bit and go from there.

twocicle

^^^ thought of that too, but had tossed the old chain when replaced with new one.

TA CNCs their teeth and leaves them with very sharply square shoulders and a side bevel is only on the tips. These do not come with any diagonal bevel to assist the chain to drop in between the teeth. I'm pretty sure it's destined for the dumpster once the new rings arrive, so meanwhile experimenting with a flat file and a round file to see if any improvement can be made over the stock shape.

I would really like to find some < 52T 130bcd rings that have a similar inner shape and tooth as the FSA Super Road rings do. Not that I love FSA rings, but I think the slope those have would help. I'll probably do some testing with a Super Road 52/34 combo (similar 18t gap to my current 48/30) to see how that performs.

fietsbob

I'd have to watch it shifting on the repair stand in front of me to really see whats going on..

mostly a showcase of Jan's products he got someone in TW to manufacture..

11 speed chain wont dropeasily over an 8 speed tooth , though the pitch is still 0.5", the thickness is different ..




I've been OK with my friction shifted stuff for decades .

the newer Campag race triple or the old Campag Olympus Mountain crank shift fine, 52/42/26t

since I dont expect it to be as instant as a light switch ..

Al1943

From the picture it appears the derailleur is not low enough and needs to be adjusted farther out (shorten the cable).
It also appears that the derailleur is already adjusted as low as it can go without cutting the slot lower or replacing the hanger with a custom built hanger designed to lower the FD.