Topic: chain and sprocket change: how much truth there is in conventional wisdom?

[Raul CB750K1]

Going further in the "conventional wisdom" issue, I found another topic that seemed very much it when it comes to chaind and sprocket renewal. I decided to change the original chain from my CB750, but both sprockets looked good to me. Conventional wisdom say "change the whole set". Why? What if the new chain is slightly shorter than the "stretched" old one? It will make its way to fit the sprockets, one way or the other, won't it? That the new chain will last shorter? So what? It still is money saved comparing to buying new sprockets too.

Conventional wisdom say that the chain stretches. I read here, and I believe it as it makes more sense, that it is not that the metal stretches but that the pins and rollers wear. That would also explain the side-to-side play of the chain. So what if when you pull the chain you can see the space of the sprocket teeth? What if the sprockets show some "sharp teeth"? As long as it is not worn down, the chain and sprockets are there for transmitting power. If the chain is tight so it can't jump out of the sprocket, what's the point of renewing parts that still are in working order? Sure thing, you can err on the safe side, but you could also change your tyres way before the threadwear indicator to err on the safe side. I guess the point is to find the perfect balance between usefulness and cost-effectiveness.

[Deltarider]

Raúl, you are right. IMO it's only the pitch that counts.

Having done over 100.000km with a CB500, this is my experience: to begin with: on my bike a set of sprockets easily serves two standard (non O-ring) chains. Then after examining the sprockets, I can simply turn them around and they serve another two chains. There's no reason to be afraid they'll break down. The CB500 simply doesn't produce that amount of hp.

[bryanj]

Raul, for too many years to think about as long as the chain is replaced before it gets totaly junked, and yes you are right it does NOT stretch the bushings wear, you CAN get three chains onto one set of sprockets BUT if you use the same make/grade of chain you will find each subsequent chain lasts less time that the previous one.

As soon as metalic parts are run together wear takes place and whilst wear in the sprocket teeth may not be visible to the naked eye it is there and will wear the new chain,
Sprocket wear usually takes place at the root of the tooth on the side that is pulling the chain (gearbox) or being pulled by the chain (wheel) so what we used to do if it was a "good" customer was this:-

First chain replacement turn sprockets over so that its the other side of the tooth that is pulling/being pulled

Second replacement warn owner life is limited and will need everthing is less time than he thinks.

Problem or antsi customers do exactly what they ask for and write in big letters on reciept "INCOMPLETE JOB AT CUSTOMERS REQUEST--NO GUARANTEE WHATSOVER ON PARTS OR LABOUR"

[Gordon]

Well, at least with sprockets you can actually see if they're worn out or not.  Something you can't do with oil.  I'm definitely not a religious "change the sprockets with the chain every time" kind of guy.  I change them when they look worn out, which to me tends to be after the second chain.  One thing that helps is to either have a set of new sprockets to compare the old ones to, or at least look at a picture of a new sprocket on the computer, because it's easy to forget just how fat the teeth are on a brand new sprocket when you've gotten used to looking at a worn one. 

[Gordon]

I think we might need a "Conventional Wisdom" sub-forum just for Raúl.

In all seriousness, though, it might not be a bad idea.  There's probably dozens and dozens of possible topics like this that could be discussed and I'm sure they'd all be really helpful to anyone looking for information about them.  In fact it's difficult to think of many tech/maintenance things we do to our bikes that doesn't involve some amount of conventional wisdom. 

[eldar]

Along this line, tire age.
How old can a tire be before it should be junked?
I suppose this is a hard one to answer though as different compounds age differently.

[scondon]

   I'm often swapping front sprocket to change final drive ratio without swapping chain or rear sprocket. Have also changed chain without touching sprockets and visa versa. Pretty much just replace what's worn unless it's a new to me bike. I like to start with all new stuff so I have a better handle on things.

   These components do transmit power to the rear wheel as mentioned, Raul. A chain that is adjusted far from its recommended play will change how this power is delivered whether components are new or not. Wear is like an uneven adjustment the way I see it. Replacement of parts depend on what I want for the bike. Never had a chain break(I don't include the master link that popped out on me) or teeth go missing from a sprocket so I imagine that running parts until the teeth round off could be OK and that would certainly save some $$

Along this line, tire age.
How old can a tire be before it should be junked?
I suppose this is a hard one to answer though as different compounds age differently.

   Please wait until all passengers have boarded and the train has left the station

[Tower]

If we examine how the chain and sprockets interact we might get some insight into maintenance, rotation and replacement approaches.

Since we know that a given chain link, will cycle over the same sprocket teeth in a predictable interval, depending on the product of sprocket size and chain length, then we can assume that chain and sprocket wear will follow this pattern.

To improve tooth  (and chain) longevity, wouldn't it be sufficient to simply jump a tooth every so often.  In this way, the wear is spread more evenly over the length of the chain, thereby tremendously increasing a chain's lifespan.  Ultimately, if chain lifespan is extended to its absolute peak by this rotation method (say twice as long wearing as without rotation), then when it finally comes to chain replacement, isn't it more likely that the sprockets would also need to be replaced as well?

If you agree with that logic, then it might be true that chains and sprockets should always be replaced together.  What we haven't been told is that chain rotation is needed to make that worth while. - just a theory

[Deltarider]

Interesting theory, Tower,
Could explain why Honda changed the 100 links chain on the original CB500 (sprockets: front 17T, rear 34T) for 98 links chains on later models.

[scondon]

 As I understand it, tooth count also plays a pivotal role. 17 is a prime, but 34t is 17x2. Does anyone know the equation that one can plug tooth count and link count to determine how many revolutions it takes for the same link to contact the same tooth?

[eldar]

I'm sorry, I guess when I get THAT bored, I take a nap!

[Deltarider]

I'm sorry, I guess when I get THAT bored, I take a nap!

OK, but then count teeth, not sheep!

[scondon]

I'm sorry, I guess when I get THAT bored, I take a nap!

 

[Tower]

This should work:
Chain Rotations (CR) = Sprocket teeth /Chain teeth * x , where x is the number required to obtain the smallest whole number of chain rotations.

e.g.1 CR = 17 / 110 *x , Therefore when x = 110, CR = 17 (smallest whole number)
e.g.2  CR = 34/110 *x, Therefore when x = 55, CR = 17

In other words the same link hits the same tooth every 110 turns of the sprocket and 17 turns of the chain in example 1, and for every 55 turns of the sprocket and 17 turns of the chain in example 2.

Edit: maybe a third example would help to show its not always as straightforward as the examples above imply.
e.g.3 CR = 35/120 *x, Therefore when x = 24, CR = 7.  i.e. the same tooth hits the same link every 24 turns of a 35th sprocket running for 7 turns of a 120link chain.

[scondon]

 Awesome, thanks Tower

[Tower]

I guess this implies that if running 17/48 sprockets with a 50 link chain (like what's on my CB750), then one apt to get much more mileage out of the chain by switching to a 17/47 or 17/49 sprocket combo.

The adventurous could try removing a link when the chain is almost worn out, instead of replacing it.  But, you should know that an odd number of links is non-standard approach (at least by conventional wisdom)

[scondon]

I guess this implies that if running 17/48 sprockets with a 50 link chain (like what's on my CB750), then one apt to get much more mileage out of the chain by switching to a 17/47 or 17/49 sprocket combo.

  My god man, how ever did you muscle a 50 link chain onto your bike

      I've got 105 links on the same sprocket combo as you and considered getting a 47 tooth rear(couldn't find a 49T) just to test this out, even though it was one tooth in the wrong direction of what I prefer for a FDR.

    I've been getting such great mileage out of this current set-up(17/48 105L), despite my horrid abuse and neglect of parts involved, that I have postponed any further research for the moment

[Tower]

Hehehe.  My chain is 100 links by your counting method @scondon.   
(BTW you can't have a chain with an odd number of links if you count that way - can only add or subtract in multiples of 2 links)    Tricky, no?  You're the only one that caught on - or maybe paid any attention. Hehehe.

[scondon]

Hehehe.  My chain is 100 links by your counting method @scondon.   
(BTW you can't have a chain with an odd number of links if you count that way - can only add or subtract in multiples of 2 links)    Tricky, no?  You're the only one that caught on - or maybe paid any attention. Hehehe.

   Oh man, math makes my head hurt. Getting.....dizzy

   Am guessing that the earlier K swing arms are a tad shorter than my F2 ones,in fact I've probably read it many times over on these forums and never paid attention

[jtb]

I think there might be only about 1 degree separation between conventional wisdom and old wives tales.

[eldar]

sssnnnkkk snort sneeze!
What is going on? Sorry I think I fell asleep there for a bit!

[Tower]

For those not fully asleep yet, maybe this will grab your attention...

We're making the claim that by going from 17/48t to 17/47t not only will you save on gasoline, but your 106-link F chain and sprockets will last 2x longer and your 100-link K chain and sprockets will last 4x longer!

To get the same benefit without going to a different rear sprocket, you would need to jump a tooth every 1/4 chain lifespan on the K, or at about half lifespan on the F bikes.

Now tell me this info is not worth cheering about 

[Raul CB750K1]

For those not fully asleep yet, maybe this will grab your attention...

We're making the claim that by going from 17/48t to 17/47t not only will you save on gasoline, but your 106-link F chain and sprockets will last 2x longer and your 100-link K chain and sprockets will last 4x longer!

To get the same benefit without going to a different rear sprocket, you would need to jump a tooth every 1/4 chain lifespan on the K, or at about half lifespan on the F bikes.

Now tell me this info is not worth cheering about 

Where do you get such an assumption?

In order to cover a mile, the rear wheel must make X revolutions. As the rear sprocket is bolted to the rear wheel, the rear sprocket must make the same X revolutions to make the bike cover a mile, irrespective of the number of teeth. Roughly half of the teeth will be in contact with the chain all the time.

The front sprocket is different. The ratio between front and rear affect the number of revolutions the front one must take to make the rear wheel cover a mile. Again, roughly half of the teeth will be in contact with the chain at all times. The gearing will mainly affect how many engine revolutions does it take for the rear wheel to cover a mile, but I don't think that giving or taking one or two teeth here and there can double the chain life in the course of 30.000 miles -30.000 times X rear wheel revolutions-. If we go further, a bigger rear sprocket means more teeth and according to your logic, more wear. But it also means bigger radius and therefore the chain goes more "straight", pins and rollers doesn't need to turn that much and therefore you are also saving wear in the rollers.



And BTW, you can have a chain with an odd number of links. There are special links for that, I remember having seen a picture but don't remember when.

[SteveD CB500F]

And BTW, you can have a chain with an odd number of links. There are special links for that, I remember having seen a picture but don't remember when.

Not heard of that Raul.

Now there's a challenge for the bored on this board...

[HondaMan]

As I understand it, tooth count also plays a pivotal role. 17 is a prime, but 34t is 17x2. Does anyone know the equation that one can plug tooth count and link count to determine how many revolutions it takes for the same link to contact the same tooth?

Divide the chain links by the sprocket teeth. Truncate the mantissa, multiply by decimals to see how tired you get of chasing it, then you'll have an idea. 

It can be a lot!