Shimano 105 5700 RD

[Andrew_P]

Could someone help, I have a Roubaix comp running 105 5600 my RD looked like a what I would call short cage (the cage for jockeys?) Any how I bought a 105 RD 5700 SS (Short Cage) and it solved the problems I had been having but the cage where the jockeys are held is longer and so on the top front chainring it stretches the chain in gear 1/2/3 gears 2/3 are okayish but it completey locks the chain on the jockey cage if I accidently go to far up.

So did I buy the wrong one? I cannot find much data on the GS version 105 5700RD but since buying have seen that the SS has a max 16t difference which I am exceeding. Front 50/34 rear 11-28 so am I right in thinking 28+16 = 44 so I am over stretching x 6t.

Anyway the original 5600 RD worked fine in all gears, I know cross chaining is frowned but it worked fine on the old one..

 

[PaulSecteur]

Have a look in the instructions or online for the "b screw adjustment". it adjusts how far the jockeys sit from the cassette.

 

[RecordAceFromNew]

Could someone help, I have a Roubaix comp running 105 5600 my RD looked like a what I would call short cage (the cage for jockeys?) Any how I bought a 105 RD 5700 SS (Short Cage) and it solved the problems I had been having but the cage where the jockeys are held is longer and so on the top front chainring it stretches the chain in gear 1/2/3 gears 2/3 are okayish but it completey locks the chain on the jockey cage if I accidently go to far up.

So did I buy the wrong one? I cannot find much data on the GS version 105 5700RD but since buying have seen that the SS has a max 16t difference which I am exceeding. Front 50/34 rear 11-28 so am I right in thinking 28+16 = 44 so I am over stretching x 6t.

Anyway the original 5600 RD worked fine in all gears, I know cross chaining is frowned but it worked fine on the old one..

Sorry it is hard to tell from the bold bit exactly what the problem is, but the official spec of the RD5700 is here and it should have no problem working with a 50/34 chainset and 11-28 cassette. Your capacity requirement is (50-34)+(28-11) which is 33T exactly per spec.

I would be somewhat surprised if the problem is the top jockey wheel hitting the largest cassette sprocket, but if so the cure is as PaulSecteur said B screw adjustment. I suspect it is more likely that your chain is a little short, if so that could have devastating consequences and must be avoided. If in doubt, use this calculator to work out what length you need and compare with what you have.

 

[the_mikey]

If the cage is longer you may need a longer chain, two to four links perhaps, if it is a longer cage it should accommodate the extra few links without any issues.

 

[Andrew_P]

Thanks for your replies, I think you are right looking at the cage it getting pulled toward the front of the bike, it is not happening on the 26t only the 28t cannot figure out why shimano would have changed the length of the cage.

 

[Andrew_P]

Sorry it is hard to tell from the bold bit exactly what the problem is, but the official spec of the RD5700 is here and it should have no problem working with a 50/34 chainset and 11-28 cassette. Your capacity requirement is (50-34)+(28-11) which is 33T exactly per spec.

I would be somewhat surprised if the problem is the top jockey wheel hitting the largest cassette sprocket, but if so the cure is as PaulSecteur said B screw adjustment. I suspect it is more likely that your chain is a little short, if so that could have devastating consequences and must be avoided. If in doubt, use this calculator to work out what length you need and compare with what you have.

Really hard to explain but the cage goes at an acute angel toward the front, and the chain is clipping the jockey cage as it rotates. Only happens now on the 28t

I have only been using the 39 for most of the day and the 50 downhill I guess I am in danger of chain snapping

 

[RecordAceFromNew]

Really hard to explain but the cage goes at an acute angel toward the front, and the chain is clipping the jockey cage as it rotates. Only happens now on the 28t

I have only been using the 39 for most of the day and the 50 downhill I guess I am in danger of chain snapping

Acute angle, if like the pic near the bottom of this page, is due to chain being too short.

 

[PpPete]

Acute angle, if like the pic near the bottom of this page, is due to chain being too short.

Correct ... but if only the RD is changed I'm surprised it is happening - even if run 50 front 28 rear, which you shouldnt anyway.
Has chain been changed ?

 

[Andrew_P]

Correct ... but if only the RD is changed I'm surprised it is happening - even if run 50 front 28 rear, which you shouldnt anyway.
Has chain been changed ?

Thats the weird thing, same chain same everything only difference is 5600 version was the old RD and 5700 new one, and the difference is so exaggerated, like I say the new Short cage looks like a standard cage to me, must add 1/2 an inch.

 

[PpPete]

Well they have increased the capacity between versions so longer cage makes sense.... but a longer cage doesn't need a longer chain IMO.

Time to go back to first principles.... disconnect the chain ( I assume it has got a quicklink ?) and wrap it round 28 rear and 50 front without going through the derailleur. It should have at least an inch (one inner and one outer link) to spare. If not add links until it does.

Just going back to the last sentence in your OP....

I know cross chaining is frowned but it worked fine on the old one..

It's not "frowned upon" - it is suggested not to do it for a very good reason, it wears out chains especially quickly, and puts a lot of un-necessary strain on other components, like RD springs for instance. Can I ask why you needed to change the 5600 ?

 

[Andrew_P]

Well they have increased the capacity between versions so longer cage makes sense.... but a longer cage doesn't need a longer chain IMO.

Time to go back to first principles.... disconnect the chain ( I assume it has got a quicklink ?) and wrap it round 28 rear and 50 front without going through the derailleur. It should have at least an inch (one inner and one outer link) to spare. If not add links until it does.

Just going back to the last sentence in your OP....

It's not "frowned upon" - it is suggested not to do it for a very good reason, it wears out chains especially quickly, and puts a lot of un-necessary strain on other components, like RD springs for instance. Can I ask why you needed to change the 5600 ?

Rarely use the 28 on the 50 except maybe when I get caught out by changing lights coming downhill.

I get about 2-2.5k out of a chain which I am happy with. The 5600 got caught in the spokes, the reason I think was I had the bike in the back of the car taking it to and from the bike shop to have a couple of bottle mount rivnuts replaced and I would think as it had been running ok before that I must have knocked the hanger out of line (replaced this with the RD change)

 

[RecordAceFromNew]

Time to go back to first principles.... disconnect the chain ( I assume it has got a quicklink ?) and wrap it round 28 rear and 50 front without going through the derailleur. It should have at least an inch (one inner and one outer link) to spare. If not add links until it does.

The problem with this method, imho, is that there is some subjectivity and risks. E.g. if this method is used on a used chain, and if subsequently said chain is replaced by new with the same number of links then one could end up with a broken mech, hanger, chain, rear wheel, or worse.

The minimum chain length needed to wrap around two sprockets between a rigid* chainstay is as exact a science as pretty much anything in cycling, unless one is on the road and has no choice using a calculator is probably better/safer.

* for completeness, none of these methods work perfectly for bikes with most rear suspension systems

 

[PpPete]

Good point there RAFN, I suppose I've shied away from the calculation methods because most of the family's bikes have old school horizontal drop-outs making chainstay measurement imprecise at best.
But, for the OP's problem, the wrapping method will adequately determine if this particular chain, which any stretch it has already accumulated, is too short, or if there is some other strangeness going on.