Don in Austin

Seems totally reasonable that cross-chaining is bad, wears the chain and sprockets, and wastes power. BUT, this sceptic wants to know if anyone has ever done scientific testing to determine to what extent. Wear factor...cassettes and chain rings don't usually get replaced because the sides of the teeth are worn as far as I know, and it seems that is the kind of wear that would be increased. Would offset sprockets really make a difference in what wears out a chain? Is increased friction from cross-chaining actually significant or only a theoretical problem of miniscule importance?

My standard for keeping the derailleurs tweaked on my five bikes is for ALL gear combos to work. One of the bikes does not quite meet this standard due to a short chain length and the front derailleur rubbing in the extreme cross-chain combos. The other four it seems I could ride all day cross-chained.

This is not to say that I am arguing in favor of cross-chaining -- I do attempt to avoid it.

I am just interested in objective evidence as to just how bad it is. "Everybody knows" I would not consider objective evidence, nor the inability to set up a bike to shift reliably in all combinations. I have done it with 4 out of 5 and I am sure some can be and some can not. Obviously, I am asking how bad it is for those bikes that CAN be set up to shift reliably in all combinations.

I am the kind of person who questions everything and likes more than "everybody knows" to truly convince me of something.

Don in Austin

DannoXYZ 

It's not the sides of the plates you need to worry about. It's the pins and rollers on the inside. Unless they're perfectly lined up, the contact-area is decreased significantly while pressure at the contact increases.

Hmm, I haven't seen any actual testing, but my in-the-field experience of 10-years in a shop seems to indicate wear-rates 200-400% faster. But that is combined with the user's methodology as well. Cross-chainers tend to do lower-quality and lower-rates of chain-maintenance as well.

Chester Kyle came up with a mathematical model a couple decades that show more frictional-losses from using smaller gears due to the higher rotation of each link around the gear (due to tighter wrap). Using larger gears had less friction. This is with a perfectly straight chainline too. So I would bet that bending the chain 4x per revolution on the extreme ends would significantly increase friction and wear.

hobkirk

Hmm right back at you, Danno. Thoughtful answer. I will need to think about this and examine the geometry. Your logic seems reasonable, although I am not sure where "bending the chain 4x per revolution" comes from - wouldn't extreme cross-chaining would be 34/11 or 53/28? And is the wear most evident on the chain, gears, or rings? Or all? And if gears, on the smallest and/or largest?

The more I know the more I know I don't know - Sheesh!

Chris_W

I did read one study showing that using smaller cogs on the back is less efficient, with the explanation along the lines of Danno's of the amount that the chain has to bend. So, a 39-16 gear is more efficient than a 30-12 gear (despite having similar ratios), even when cross-chaining is not a factor. The experiment showed this using an infra-red camera to measure the heat build-up in each combination caused by the friction involved. This is also the reason why a few pro' racers have started to use modified rear derailleurs with oversized derailleur pulleys - see Contador's 2010 TdF bike here.

abdon

No comment on efficiency but as far as wear goes, my experience is that the chain carries the brunt of the wear, and that you don't really start hurting your other components until your chain is nice and stretched.

I don't understand what the fuss is all about. Chains are consumable items; you are supposed to replace them. I guess if I were to stop cross chaining at the same time I'm off-the-saddle stomping and shifting that my chain would last longer, but that would take the fun out of things I do test and change my chain on a regular basis.

De only way to not wear your chain is by not riding your bike If you do ride your bike, I guess it is a matter of priorities; chain wear vs. freedom to stomp your bike in any way you like. Considering how cheap chains are this is not such as expensive habit.

scirocco

The derailleur pulleys are not subjected to the pedalling force, only the derailleur spring tensioning force, unlike the main cassette cogs.

I'm not sure exactly what engineering analysis the recent trend of larger derailleur pulleys is based on, but if it's the generation of enough heat that you can measure it with an infra-red cam, I'll be mighty surprised.

tcs

"Cross-chain gears make little difference." - Dr. Chester Kyle and Frank Berto, "The mechanical efficiency of bicycle derailleur and hub-gear transmissions", Human Power (Technical Journal of the IHPVA), #52 Summer 2001.

Don in Austin

I thought that was possible. I wonder if they refer to efficiency loss or wear or both.

Don in Austin

DaveSSS 

Cogs can be ruined by a chain that has almost no elongation. I've proven that a couple of times with Campy chains that typically show very little elongation, even after 6,000 miles.

Often overlooked is chain tension. The smaller the chain ring, the higher the tension.

I rode in the mountains with a 28T little ring and wore out one steel cog in only 6,000 miles. I also had a bike with four Ti cogs and wore-out both the 19and 21T in only 4,000 miles.

Cross chaining isn't real smart, but sure you can do it - all day long if you choose. With a Campy chain, the result will be extreme side wear, more than elongation. The chain that I used for 6,000 miles only had .2% elongation, but the side clearance was about twice as much as a new chain and I don't use the extreme chain angles.

DannoXYZ 

Imagine that looking down over the chain & rings from above. All of the chainrings and cogs are in parallel planes. When the chainring and cog are in the exact same plane, then there's no cross-chainring.

In the extreme cases of cross-chainring, imagine 4 points where the chain meets and releases from a cog & chainring. This is where it bends. The parts on the cog/chainring themselves will be perfectly straight while the parts in between will have to go through a lateral bend in order to reach the out-of-plane next part.



The wear on the rings won't be significant relative to the wear that occurs from shifting. The chain is bent even more severely then and there are spots where the chain-plates meets the teeth at an angle. This cuts the corners of the teeth and removes some material. You can really see this occur when you install a brand-new chainring. SuperGlide chainrings have this wear "pre-cut" into the rings at certain spots to help improve shifting.

fietsbob

Donations for me re entering college to study Mechanical engineering greatly encouraged, Ill get right on it..

tcs

In their paper on efficiency they were refering to efficiency.

tcs

cobba

For some strange reason I don't think cross-chaining is a good idea if you want to maximise your drive-trains life. The angles in the following image mightn't be exact but they give you a bit of an idea of what's going on and I don't think that a chain is meant to do what is shown in it.

Cross-chaining.gif

bradtx

Don, My only scientic study is the observation of a chronic cross chainer and the comments by her husband who maintains her bike and is well qualified to do so.

Since she likes to keep the chain on the large chain ring, it becomes knife edged and has about half the life span of a non cross chained ring. The tension pully wear rate is next in line while the rear cogs show the least effect, but they too are short lived. Her road bike shows the effect of cross chaining more quickly than her mountain bike. I don't really know why... smaller big chain ring, longer chainstay or perhaps a combination of both?

Brad

fietsbob

More friction = less efficiency

cny-bikeman

The one thing I can say without testing is that cross chaining on the small chainwheel and the smallest cog(s) will wear out the chain and cog faster than the alternative, which is riding a combo with the larger chainwheel/cog combo that is comparable in gear ratio. In that case the cross chain aspect is only secondary. The main point is that with the small/small combo the same force is being applied to fewer total teeth both front and rear, so the components will wear faster. For that reason I may ride large cogs in the rear cross chain but never the smallest two cross chain.

FBinNY 

The problem with how people think about cross chaining is that too many think digitally, calling outer with outer seven OK and with inner two no good (9s). Likewise in the opposite direction.

That's flawed way to think about it. It isn't a matter of black and white, but shades of gray depending on the degree of mis-alignment.

On a typical 2x9 system the chainrings are aligned on either side of the middle of the cassette. That means that using the outer with the 3rd-5th sprockets is most efficient (from a chain line standpoint) 2nd &6th less so, the 1st & 8th still less and the 9th the least. A similar analysis applies to the inner chainring. Once you plot it you see that the supposedly poor outer and 8th combination has roughly the same deflection as the commonly used correct outer/outer combination.

Cross chaining is slightly worse with triples because the offset is increased slightly, but not earth shakingly worse.

I won't get into the nature and causes of wear because the prior posters have already covered it, except to reiterate that the combination of lower tension, and larger radii make all chain drives more efficient with larger sprockets, though the idlers on the slack side aren't a significant factor either way.

The only real problem with cross chaining is that with today's gated chainrings it increases the likelihood of a chain derailing spontaneously at a shift gate, since the chain is feeding from the side. Shorter chainstays also magnify all problems of poor chain line, since the same deflection results in more angle.