I have decided to get some shorter cranks for my stoker so have bought the SJSC crossover crankset (160mm + 175mm crank lengths) and now need to buy rings and a chain. I have 24-36-46 rings already (its an 8 x 3 = 24 gear tandem) however I am not sure what to do for the crossover drive. I think the cranks are 110BCD.
What is best thing to do ? Get a normal set of "middle" 36 tooth chainrings and a normal 8 speed chain ? Is it possible to buy a "fixie" or singlespeed 3/16" chain, and if so which one should I go for ? Or could I buy some 1/8" chainrings and have a really rugged 1/8" chain setup (I stand on the pedals a lot and am hard on the components). I found some 44 tooth 110BCD 1/8" BMX chainrings on eBay (pictured) - how about these for the crossover drive ?
(I currently have a newish KMC 1/8 crossover drive chain on the (very) cheap pressed steel original crossover chainrings - it has already stretched loads but apart from that seems pretty ok in use .. despite being too wide for the chainrings ???)
Any experience or comments on this welcome - I would really value your views on this .. it is uncharted territory for me
Attachments
Last edited by zenitb on 22 Mar 2019, 10:18pm, edited 1 time in total.
years ago I converted a tandem from 32T chainrings in the crossover drive to 46T chainrings in the crossover drive. The transmission immediately felt a lot stiffer from the captain's seat.
You could go for 1/8" chain but remember the front chain in a crossover drive only takes one rider's worth of aggro so it isn't the weakest link; with a powerful crew the rear bottom bracket and the rear drive chain are the things that you usually need to worry about.
As Brucey says, bigger rings feel stiffer. I have tried both 1/8 and 3/32 chains on the original 38 tooth rings I had. I was disappointed with the kmc 1/8 chain which stretched quite rapidly. ( Not a B1 btw). And resorted to 1/8 bushed chains salvaged from a couple of dumped 3 speed, which were an improvement. I now have these rings above with a 3/32 kmc bushless chain. And you can rotate the rings and turn them around to even out the tooth wear. As I have identical left cranks I swap the chainset front to back to even out the wear so the puling crank becomes the pulled one. simples.
At the last count:- Peugeot 531 pro, Dawes Discovery Tandem, Dawes Kingpin X3, Raleigh 20 stowaway X2, 1965 Moulton deluxe, Falcon K2 MTB dropped bar tourer, Rudge Bi frame folder, Longstaff trike conversion on a Giant XTC 840
Brucey wrote: you mean 3/32" not 3/16", don't you?
cheers
...fractions were never my strong point !!!
Thanks for the advice though... I understand your point about 8 speed stuff at the back being the weakest link Brucey...and jumping gears at the rear of a tandem is really not a good scenario of course. However we have unshipped the crossover chain a few times mid hill (due to stretch/ hitting the adjustment limits of the EBB/me not bothering to take a link out) so I am open to a chunkier setup thsn normal for this chain. One of my motivations in customising the crank length is to make it easier for us both to stand on the pedals for short sections....interesting that the larger chainrings felt stiffer...
As Brucey says, bigger rings feel stiffer. I have tried both 1/8 and 3/32 chains on the original 38 tooth rings I had. I was disappointed with the kmc 1/8 chain which stretched quite rapidly. ( Not a B1 btw). And resorted to 1/8 bushed chains salvaged from a couple of dumped 3 speed, which were an improvement. I now have these rings above with a 3/32 kmc bushless chain. And you can rotate the rings and turn them around to even out the tooth wear. As I have identical left cranks I swap the chainset front to back to even out the wear so the puling crank becomes the pulled one. simples.
Interesting post RJB ...I had a similar stretching experience with my KMC chain (pictured) which I think has the bushes Brucey mentioned..I bought two of these from Decathlon and they are just called "BMX chain" on the packaging...maybe these are the B1? After a few rides the stretch seems to have stabilised so I am hopeful they will last. I just dont want it unshipping like the Viking original did in the end.
SJSC 36 middle rings look interesting...thanks for that....and I take all points about rotating/reversing etc..
chain above does look a lot like a B1. I think they are a bargain for what they are, but they are not very hard-wearing. FWIW for a crossover drive I would vote for
- big chainrings - full height teeth on the chainrings (no ramps or gates or truncated teeth) - chain with cambered inner side plates - chain with relatively low lateral flexibility
In a long chain run, the static tension ends up being quite high if there appears to be a relatively small amount of slack; it is all to do with the mathematics of catenary curves.
Brucey wrote:chain above does look a lot like a B1. I think they are a bargain for what they are, but they are not very hard-wearing. FWIW for a crossover drive I would vote for
- big chainrings - full height teeth on the chainrings (no ramps or gates or truncated teeth) - chain with cambered inner side plates - chain with relatively low lateral flexibility
In a long chain run, the static tension ends up being quite high if there appears to be a relatively small amount of slack; it is all to do with the mathematics of catenary curves.
cheers
I have ordered the 44 tooth "BMX" chainrings and will try the KMC bushed (a.k.a. Decathlon "BMX Chain") with them .. will ping back at some point once its all bedded in ... thanks to all for the help/advice.
I cannot see a split in the bushing in the pictured chain. Brucey pointed out that the bushing in the B1 is a rolled piece of steel and is not especially hard. If the bushing in the BMX chain is differently formed it might also have a different hardness and wear resistance. It does otherwise look very similar.
My Dawes Discovery tandem doesn't allow sufficient adjustment in the eccentric to avoid the use of a 1/2 link in the chain. The eccentric allows approx 20mm adjustment. 5mm more and I could have removed a whole link.
At the last count:- Peugeot 531 pro, Dawes Discovery Tandem, Dawes Kingpin X3, Raleigh 20 stowaway X2, 1965 Moulton deluxe, Falcon K2 MTB dropped bar tourer, Rudge Bi frame folder, Longstaff trike conversion on a Giant XTC 840
Brucey wrote: you mean 3/32" not 3/16", don't you?
cheers
...interesting that the larger chainrings felt stiffer...
Presumably because the same torque (eg pedal force x crank length) is transmitted with lower chain tension. This is also why large chainrings wear slowest - more teeth sharing the load AND reduced total chain load for same rider effort.
rjb wrote:My Dawes Discovery tandem doesn't allow sufficient adjustment in the eccentric to avoid the use of a 1/2 link in the chain. The eccentric allows approx 20mm adjustment. 5mm more and I could have removed a whole link.
I have just hit the limit of EBB adjustment RJB..lets hope I dont have the same issue!!
a good amount of the feel of stiffness increase with larger chainrings comes because the lower chain tension causes a smaller bending moment in the 'drainpipe'. This is also reduced if the crossover drive chainline is brought inwards.
Removing a whole link from the chain shortens both the top and bottom runs by 1/2" so requires a BB movement of 1/2": if you have 20mm BB movement available, and you run out of adjustment, you should never need to use a half-link.
Oops I meant 10mm! On the eccentric, that's 20mm on the overall chain length. When it's at the forward limit, turning to the slackest doesn't allow sufficient slack to remove a whole link.
At the last count:- Peugeot 531 pro, Dawes Discovery Tandem, Dawes Kingpin X3, Raleigh 20 stowaway X2, 1965 Moulton deluxe, Falcon K2 MTB dropped bar tourer, Rudge Bi frame folder, Longstaff trike conversion on a Giant XTC 840
if you have a free choice of the chainring sizes, it is maybe worth choosing one that leaves you with the maximum adjustment possible when the chain is new. Unfortunately a good deal of this is set by the cunningness of the manufacturer; you only have limited control over this.
If the centre to centre distance of the BBs is 'B' (in inches) and you use chainrings of 'N' teeth then the total chain length 'C' (in whole links) is given by
C = N/2 + 2B
So varying the chainring size by one tooth is the same (length-wise but not timing-wise) as adding or removing a half-link to the chain, or moving the BB centres 1/4" further apart.
If you are lucky then the minimum dimension 'B' is a fraction under a whole 1/4" dimension. However ideally it should be a fraction under a whole 1/2" dimension, because 1/4" dimensions (e.g. 20-1/4" or 20-3/4") mean that there may be unwanted restrictions on the position/tooth count of the chainrings if you use the normal (same phase) crank positioning.
It is actually a lot more complicated than it might at first appear to be, (esp for alternate crank phases) but if you are fussy about 'normal' crank phasing (and IMHO you should be) then the closest BB centres ought to be a fraction under a whole 1/2" size (for simplicity), and you should consider using a chainring that doesn't have a tooth count that is exactly divisible by the number of fixing bolts. This constraint allows you to vary the phase of the front and rear cranks slightly. This is best (and easiest) achieved if you have five-bolt fixings and (say) + or - 1T from a multiple of five (eg 44 or 46T) chainrings. You can get the same phase adjustment with other tooth counts (that are not multiples of five) but it isn't as easy to see what is going on.
The reason for fussiness about crank phasing is that if one set of cranks leads the other even slightly, it feels to that rider like they are doing more of the work. This may be desirable if one rider is a lot stronger than the other, but in a well-matched crew, if you want to make the best progress, the workload is best shared, or at least felt to be shared.
If the cranks are 'perfectly phased' then it won't have escaped your notice that, under load, the lower run of the timing chain sags slightly. This also causes the front cranks to advance slightly. Moving one of the chainrings on the (five bolt) spider allows the front cranks to be retarded a fraction, so that, under load, given the usual chain adjustment, the cranks return to true 'perfect phase' and the workload is felt to be well-shared again.
Probably there is a better way of saying what I've said above, but I've rarely seen a very good discussion of such matters.
[edit; BTW unless you have chainrings with an odd number of teeth and one can be turned 180 degrees, putting a half-link into the timing chain is guaranteed to put the cranks out of phase with one another.]
cheers
Last edited by Brucey on 23 Mar 2019, 8:09pm, edited 1 time in total.
A thought that has just crossed my mind & don't know if it is something that is done.
If you run out of EBB adjustment before you could remove a whole link, would it not be possible to swap chainrings to ones with a single tooth more to take up some of the slack. Or start with rings one tooth smaller if that works better.
Former member of the Cult of the Polystyrene Head Carbuncle.
As I have identical left cranks I swap the chainset front to back to even out the wear so the puling crank becomes the pulled one. 
