Chain Length Issue
 

Took out my cross bike for a shakedown yesterday. When I am in the small chain ring and standing and really stomping up hills, the chain was jumping around on the cassette seeming to shift on its own. Otherwise, when I am sitting or riding a flat the system shifts flawlessly. If I am in my big ring up front, and standing on a climb, the chain does not skip around on the cassette. My theory is that the chain might be too long. When I put the chain on the big ring in front and the smallest cog in back, the pulleys are not perpendicular to the ground, but instead at about a 65% angle. If I shorten the chain a few links, making the pulley cage perpendicular to the ground, might that solve the problem?

the way i check a chains length is disconnect the chain, take it out of the rear derailleurs pulleys. wrap it around the large front chainring and the large rear cog. pull the chain tight and overlap by at least 2 full links. that should be the length of your chain.

What happens when you shift into the big front/big rear combination? The derailleur cage should be pulled almost horizontal yet still have just a skosh of slack.

A chain that's too short can cause a bunch of damage to your bike in less time than it took me to type this sentence.

As long as you don't have slack in the chain, it should be short enough (for whatever gear you're in). Obviously, you're most likely to get slack in the small ring-small cog combination. Since the chain is jumping around when you are in the small ring, that's where you should be checking for slack. If you have slack, you can shorten the chain, as long as it remains long enough to shift into the large ring-large cog combination. (Some people will shorten it beyond this and just avoid the large-large combination, but if you forget and try to go there, it can really mess things up. I wouldn't do it.) If the chain is already as short as you can go safely, you might consider running a derailleur with a longer cage, if you don't already have one.

It's possible, I think, depending on the nature of your frame and your size and weight, that if you're really stomping on the hills, your frame might be flexing enough to cause the ghost shifting. Chain length might not be the problem.

The system works fine when I am in the large front chain ring, even on the largest rear cog. I don't think the ghost shifting is due to my frame (Cannondale CaadX) as I do not experience chain skip issues when I am standing and stomping up hills on the large front chain ring. One more point to mention, this bike was brand new in October 2011, so I've only used it for part of one cross season, so I have eliminated any kind of wear issue from the equation. I am wondering whether the bike shop installed the chain too long when I purchased it. I did notice the same issues last year, and want to resolve it before this season starts. If the chain were too long, would it be common to only experience chain skip when I am standing on a climb, or would the problem reveal itelf in other situations as well?

One other possibilty. I did have several crashes during the season. If the derailleur hanger was bent ever so slightly, would it be possible that it would cause the chain to skip under load and only in the small front chain ring?

Based on what you've said, I agree that it's unlikely to be frame flex. I'd recommend checking chain length first, as sloar (post #2) suggests, and then going from there. My own experience has been that I am most likely to get chain skip if the chain is too long and there's slack in it, and if I'm applying a lot of pressure to the pedals somewhat unevenly (mashing as opposed to spinning). So, yes, under certain circumstances but not others -- at least for me. I suppose a bent derailleur hanger could be the problem, too. I'm not sure why it would cause difficulties only when you're in the small ring, but perhaps it could have something to do with affecting the chainline to the small ring as compared to the big ring. In any event, as I say, I'd start with chain length.

Also inspect your chain rings and chain wheels. A worn tooth or teeth can cause the problem.

As with most mechanical issues on a bicycle, one starts with a knowledge of what needs to happen for the problem to occur. In this case, "ghost shifting" happens only in the small chainwheel would be due to one of three circumstances:

1. The derailleur position or too long a chain causes chain/cassette interference
2. The derailleur moves from it's position relative to the cassette.
3. The derailleur cage is not properly aligned to the cassette (parallel in both planes).

If the problem occurs in other than the largest rear cog #1 is unlikely, but confirming correct B tension adjustment and chain length should be done anyway.

Given the fact that the problem does not occur in the higher torque gears on the large chainwheel we can eliminate #2 without further consideration of the frame involved, shift lever, etc.

Therefore the derailleur must not be aligned to the cassette. But why then does it not occur in the large chainwheel? There are two planes in which the derailleur can be out of alignment. Many people are familiar with the one that involves the bottom of the cage being closer to the cassette than the top, usually occurring from a fall on the right side of the bike bending in the hanger.

The other plane, though is whether the pulley when viewed from the top is not parallel to the vertical front-to-back centerline of the bike. That misalignment most often occurs with a foreign object or accident interfering with the chain or derailleur and causing severe stress. Both of course can occur at once.

The key to the difference in shifting between operation on the large and smallest chainwheel is the rotation of the pulley. If the derailleur hanger is merely bent inward the relative position of the top pulley under a cassette cog does not change (or at least very little). However, if the pulley assembly/derailleur or the hanger is misaligned in the other plane (twisted) the top pulley position will change significantly.

One can easily determine the latter problem by placing the derailleur on the small chainwheel and a middle or smaller rear cog. Pull the derailleur backward to rotate the cage and observe whether the cage orientation changes. You can also check be adjusting the tension to see if that eliminates the problem.

The fix is as follows:
First, align the derailleur hanger - with a tool, not by eye. Then check the derailleur. If you can't see a problem rotating the pulley, test ride. If the pulley seems to be twisting, or if the test ride fails, replace the derailleur,.

p.s. Also, just noticed the "shakedown" reference. Exactly what does that mean - new bike, first hard ride, just installed new components, etc? We are also missing one critical piece of info that one should always include. When did the problem start and were there any other changes or incidents that occurred at or just before that time?

Thanks for the detailed reply cny-bikeman. I cannot pinpoint the precise time when the bike started behaving in this manner, but I am fairly certain that it began after I had a crash in a race in October 2011. The problem persisted through the end of the season, and frankly I forgot about it over the winter. I took my bike out for its first hard ride of the season yesterday. I have not changed anything in the way of components or made any adjustments since the end of last season. The reason why this problem has just surfaced now is because this was the first time I took the bike out for a hard ride. I did take the bike out last week for some easy spinning (including hills) and noticed no problems.

Why wouldn't the twisted rear derailleur affetc the shifting when it is on the large front chain ring? It seems that you would have the same degree of twist whether the chain is on the large or small chaniring and so the alignment would be imporper in either case.

The timing certainly fits with my deduction. No, the amount of misalignment can indeed change with the rotation of the pulleys. To illiustrate that, place your hand in front of your face, edge on and vertical (fingers pointing straight up, thumb toward you) and sight past your fingertips to a fixed vertical item (door frame for example. Now rotate your hand forward so that your fingers point forward at a 45 degree angle. Notice your fingertips are still in line with the target.

Now go back to the original position but turn your hand slightly to one side like a rudder, so that you can now partially see your palm. Again make sure you are sighting over your fingertips to that vertical plane. Now rotate your hand forwrd in the vertical plane the same as before to 45 degrees. You will see that your fingertips are now to the left of the target.

Your hand is the pulley assembly, the target is a cassette cog. When your hand is upright it is on the large chainwheel, when it's pointed forward it's on the small one.

Chain length doesn't affect the chain's traction on the sprockets either way. If you think about it for a second you'll see that. The chain on top is tight reflecting the bike driving force, but the bottom loop is basically slack, having only the tension from the RD idler cage spring. If the sprocket didn't keep the chain engaged, it would pull the bottom loop around all the time regardless of chain length.

So you need to look elsewhere for your answer. I suspect that the RD tab may not be aligned perfectly so the trim changes slightly as the RD pulloy swings it's arc. of if this is a Shimano or Campy derailleur (doesn't apply to Sram) the upper pulley rises and falls with RD cage rotation. That means the B-screw adjustment needs to be checked in every gear combination. If you didn't do this there's a decent chance that when on the smaller front sprocket, the idler cage rotates, bringing the pulley high and into contact or near contact with the sprocket. On a hunch, without seeing your bike, bring the B-screw in to lower the RD a bit and see if that magically clears the problem.

Your statement is true, but in my post at least I knew the chain's traction was not the issue, but rather the secondary effect that the wrong chain length might have on interference between the cogs and derailleur cage. It certainly can't hurt to confirm correct chain length. Several of us have indicated the need to check the b-tension screw - we don't know at this point if the OP has tried that or anything else.

The coincidental placement right right after your post, may have misled you into thinking I was referencing. I wasn't, and in fact had I read your detailed post, I might have not posted at all. But many people don't see it import of a floating (vertically) jockey wheel as used by Shimano and Campagnolo. It's one thing to set the B-screw, off the largest sprocket, as Shimano suggests, but many people forget the other requirement to re-confirm the adjustment using the smaller chainring.

While Shimano suggests a chain length such that the idler cage is vertical ia a middle gear, this isn't critical, as long as the chain is between the maximum (Campy small/small method) and minimum (Sram big/big +1" method) length. Regardless of the method, the B-screw can be set so the RD has decent clearance in all combinations.

Your well explained point about hanger twist is also useful. I touched on this quickly, but you fleshed it in greater detail.

Actually, FB I'm glad you did post because I did not think of or speak to the issue of checking the B screw in different combo's.

Thanks all for the responses. I tinkered with the bike last night. The chain does not appear to be too long. It was not sagging at all when I had it in the small front chain ring and small rear cog. I also pulled back the rear derailleur cage as suggested by cny-bikeman and there is no observable twist to the derailleur cage. I have not yet checked the derailleur hanger, but my assumption is that if it were a bent derailleur hanger, it would affect the shifting in the large front chain ring as well as the small. I also checked the B-screw according to the Park Tool webiste instructions. (This is a 10 speed Shimano 105 derailleur). Frankly, when I loosen or tighten the B-screw I dont see a whole lot of movement in the derailleur in relation to the rear cog (at least with respect to the largest rear cog). I haven't checked the B-screw adjustment to the other rear cogs.

A question on measuring chain length: In Leonard Zinn's book on raod bike maintenance, he says that when the chain is the correct length, the jockey wheels should be in a straight line perpendicular to the floor. Is this a proper way of measuring chain length? If so, my jockey wheels are not perpendicular, but angled at about 75 degrees.

http://www.parktool.com/blog/repair-...-length-sizing

I prefer Largest Cog and Largest Chainring Method

Please don't ask for advice if you are going to dismiss it. Several of us have explained that it is not the chain length per se, and sagging in the small-small combo is irrelevent in any case. You may or may not be able to detect the twist, as you are not accustomed to what to look for, and your "assumuption" about the effect of hanger twist is flat out incorrect, as I took pains to explain with an example. As it happens I had exactly that problem on my own bicycle, and the only thing that fixed it was hanger alignment - although that is hardly the first time I've seen the issue.

I don't generally trot out my "credentials" but rather let my posts stand on their own. This is an exception. I was a mechanic, service manager and business owner in the bicycle industry for over 20 years. I was considered the best person in my city for frame and fork alignment, and did more fitting work than anyone in the area, short of a framebuilder about 30 miles away. I designed and taught a 60 hour bicycle mechanic course, raced, commuted and toured, including a 3 month 10,000 mile trip in 1976. I have solved many problems that have stumped other mechanics in shops where I have worked, and some on here as well. I gave you the procedure to follow to solve the problem, and what I deduced was the cause (accident/impact) was most likely spot-on. And yet your "assumption" trumps all of that.

I do not come to conclusions and recommendations I post from my opinion or some mental exercise imagining bizarre reasons for problems, or "what happend to me one time," but rather from all of that experience coupled with a logical approach to problem solving. The same holds for FBinNY, mrrabbit, hillrider and others. If you want to discard all that then don't ask the question in the first place.

At this point you are not going to solve this online or on your own. You need to take the bike to a competent shop and have it checked by the head mechanic. You probably won't listen to this but it would probably help to advise the shop that you have been told that the most likely problem is twisting of the hanger or rear derailleur/pulley. Unlike you I would not be bothered if I were told I was wrong and that there was something I did not consider or overlooked.

Im not sure where you're coming from with this. I wasn't rejecting your analysis, just asking for clarification so I could better understand it. In your earlier response, you said that a twisted derailleur could cause the bike to shift differently in the small vs. large chain ring. I never questioned that. If you carefully read my response, I said that it was my assumption that a bent derailleur hanger (not twisted derailleur) would cause it to shift the same in both chain rings. If you meant your analysis applied to a bent hanger as well, you should have said so. Don't worry, I won't dare question your supreme knowledge again on this forum. This is not the first time you've gone after other posters with the "don't ask for advice if you're going to dismiss it" line. I don't give a crap how long you've been in the business--it doesn't give you the right to speak to people as if they were a child.

To the OP,

Stop focusing on the chain length. As cny-bikeman said, you have the benefit of advice from people with something like 75-100 years of experience between them, telling you that the most likely issue is a bent or twisted hanger or a high jockey pulley (B-screw adjustment).

There are 3 ways to measure a chain, all equally valid, except that everybody has a favorite. Campagnolo suggests using the longest chain possible (no sag on small/small) Sram recommends the shortest (big/big +1") and Shimano is like the baby bear taking a middle ground position within their vertical cage method.

Note that if there's no extra RD take up capacity, all three methods yield the same answer, the minimum is the maximum and the middle option all at the same time.

Go back to the bicycle, shift to a problematic combination and look at the area where the chain leaves the upper pulley and meets the sprocket. There should be (must be) roughly 1" of free chain free between the two. If the pulley and sprocket are touching through the chain or early so you'll experience the type of problem you're having.

You correct that with the B-screw, but don't just turn it in, instead pull the RD back against the spring a bit to take the weight off, adjust the screw then let the RD swing forward. If adjusting the B-screw doesn't change anything, then possibly the stop cam is damaged or mismounted.

A bent hanger may also cause problems because the plane the RD cage rotates in isn't parallel to the plane of the bike. That means that as the jockey wheel will move inward (or out) as the cage rotates causing a change in trim.

From your description, I lean more to the high pulley (b-screw) theory, but take advantage of the advice you've been given, and check both the hanger and the pulley height. If both are correct, then something elso might be happening, and you should let a mechanic look and see if with the benefit of hands on he can solve it for you.

It was not just my advice, but FBinNY's as well that chain length was not the cause and that twisting likely was, and apparently you missed this part of my post:

"However, if the pulley assembly/derailleur or the hanger is misaligned in the other plane (twisted) the top pulley position will change significantly."

I acknowledge I came across heavy handed, and I apologize. Probably did not help that I just left a thread where the poster rejected all but the responses that massaged his totally unrealistic expectations as hostile or useless, but I consider expecting someone to acknowledge the responsibility inherent in asking for help as well as for giving it to be an adult concept. I don't expect anyone to slavishly follow my advice or diagnosis - or anyone's.

cny- your drinking too much coffee. Chill out. We're here to lead horses to the water, but it's strictly up to them whether they want to drink or not.

FB,

I know - not too much coffee, too much stress spilling over into something that's supposed to be fun. Thank s as always for your voice of reason. Signing off this one.

--Don