How to calculate new chain length after sprocket change

[TravisBickle]

Haven't ordered the sprocket yet but looking to swap a 45t on a 390 Duke for a 70t sprocket (custom made in Poland).

Is there some magic formula for calculating the required chain length for the new setup? Or is it a kinda offer it up kinda deal once the new sprocket is in situ and have a best guess or use a piece of string or something?

Might go with a 60t Supersprox Stealth if I can get one to fit, I think 60t is the biggest they do but I'm not sure they do one of that size to fit the 390 Duke??? Anyone able to confirm this? And if not, could one for the wrong bike be made to fit by drilling holes in the right places? Taking it to an engineering place to get it done with precision, obviously, not just done slap dash on my workbench at home LOL! And would a 60t be enough to make it effectively into a stunt bike? Or would 60t just be some hideous setup halfway between having an unusable gearing ratio for daily riding but not quite good enough for easy stuntery?

I'm lead to believe that the original spec is 520 112 15 45 but I could be wrong (I was once). Not able to physically verify this myself at this stage as I'm buying the bike from a friend when I get back from holiday and he gets some shore leave in a couple of weeks but I wanted to get a head start on getting the parts on order if poss!

Seems logical to me to:
Take the tooth difference between the new and original sprockets (70 - 45 = 25)
Divide by 2 = 12.5
Add to original 112 chain length = 124.5
Making it a 124 or 126 I need

Or have I just WAY over simplified this process???
____________________
2024 BMW M1000XR
2021 Honda CMX500 Rebel S
2016 Triumph Tiger 1050 Sport

[Pete.]

Look at it more simply.

If you don't move the sprockets but add 2 teeth to the total count (doesn't matter which sprocket they are on it's the total count that matters) you need a chain with one extra pin count. This is because the chain is wrapping around half of the total teeth.

It's also the reason why people get caught out by removing links from a badly worn chain and often discover they don't have the adjustment to join the ends back together.
____________________
a.k.a 'Geri'

132.9mph off and walked away. Gear is good, gear is good, gear is very very good

[TravisBickle]

Yeah that's basically the logic behind my thinking except for a VERY small increase in distance from the top edge of the front sprocket to the top edge of the rear sprocket when the rear tooth count is increased, for example, it increases the angle upwards of the chain from the front sprocket to the rear sprocket so by some Pythagoras crap I seem to remember from school can only have the effect of increasing the chain length (even further than simply adding one link per two teeth increase) as that one side of the triangle gets increased ever so slightly... however in most cases I doubt it equates to more than one link's worth of difference, even going from 45t to 70t... and I'm guessing we round things up to the nearest even number of links? i.e. 124.5 links becomes 126 links?

Like I said... just wondering if I had over-simplified it or whether there was some magic formula involving pi and Pythagoras' theorem etc?
____________________
2024 BMW M1000XR
2021 Honda CMX500 Rebel S
2016 Triumph Tiger 1050 Sport

[TravisBickle]

[Zen Dog]

https://i.imgur.com/UF5dRP1.png

I see what you're saying. Basically that A2 will be longer than A1 because of the radius increase in the rear sprocket B changing the angle of the chain between the front and rear sprockets. And while this would only make a very small increase in the chain length needed if you were only going up a couple of teeth, it's going to make much more of a difference when you're going extreme (which you are...).

I don't have any magic calculation, but if it was me I'd probably use Pete's approach, and then add 4 links (2 for the top and bottom chain runs) for the change in angle. Almost certainly overkill but it's a lot easier to get it too long and shorten it than the other way around. Or, you know, just wait till you've got all the bits and measure how much extra you need using the old chain as a guide.
____________________
Current - '94 VFR750FR (Dead), '00 VFR800FI, 2011 CBF125 Previous - '10 Street Triple R, '92 MZ ETZ301, '05 TTR250, NSR125R, KMX125, "Honda" Win (chinese copy of an old Honda design with a C90 engine)
My bike trip around S.E. Asia 2010/2011

[stinkwheel]

I usually buy one I know is a bit too long, set the adjuster to its minimum setting, lay the chain over the sprockets then wind it out until I find where the pins line up, cut it there and join it.
____________________
“Rule one: Always stick around for one more drink. That's when things happen. That's when you find out everything you want to know.”
I did the 2010 Round Britain Rally on my 350 Bullet. 89 landmarks, 3 months, 9,500 miles.

[TravisBickle]

Yep, fair point!

I've never fitted a new chain myself before, but when I've ordered them in the past for my mechanic to fit I'm pretty sure I've always ordered the correct length and it's come as an already-connected loop? Possibly meaning the swingarm having to be removed for fitting? If I were to do this job myself, would a clip-type master link be ok for a 390 Duke? Or would it be necessary to get the correct tools and use a rivet-type master link?

Being only 40odd bhp, I'm guessing clip-type is probably ok? Although with the change in tooth count that's gonna put a lot more torque through the chain than standard...
____________________
2024 BMW M1000XR
2021 Honda CMX500 Rebel S
2016 Triumph Tiger 1050 Sport

[jaffa90]

[TravisBickle]

[stinkwheel]

I preferr a rivett link. However a mate of mine used to run a 300bhp drag bike with two split links in.

An old trick was to fit the chain with a split link then ride the bike to someone with a rivetter and get them to fit the rivett link.
____________________
“Rule one: Always stick around for one more drink. That's when things happen. That's when you find out everything you want to know.”
I did the 2010 Round Britain Rally on my 350 Bullet. 89 landmarks, 3 months, 9,500 miles.

[xX-Alex-Xx]

I'll just leave this here...

https://store.schnitzracing.com/motorcycle-chain-length-calculator/
____________________
DILLIGAF

[Pete.]

[GettinBetter]

I'd just put the old chain on, tie the ends together as tight as possible, then measure how short it is. Add that to the original chain and hey presto you have a length. make sure you get one a bit longer, tweak it yourself.

[Pete.]

[GettinBetter]

[P.]

As a person who used to run a dinner plate sized rear sprocket...

Are you sure you want 25 extra teeth.

I went 41 to 49 and knocked 20mph off and just increased the revs at cruising speeds. Sure it wheelied a lot in any gear but..

This led to the engine shitting the bed, it was a solid single cylinder 660 Yamaha engine.

Yours is a cheaply made engine by a less than reliable company. I wouldn't be sticking more than a few teeth on at max, going up 8 really fucked it.

[redeem ouzzer]

Vastly undergeared, this is an autistic idea.
____________________
Be a REAL MAN!

[ThunderGuts]

It's a pretty extreme change - 45T to 70T! The fact it's a custom sprocket says something about how extreme it is. Other than the ludicrous change in gearing, I'd also check the chain line will still work - that's moving the chain a fair amount both up and down.

What's the reason for the change?
____________________
TG.

[Kickstart]

[TravisBickle]

Thanks for the input Paddy.

I was wondering that myself whether that's going way too much or not, even for a stunt bike? I've only ever gone up a couple of teeth at the rear and one down at the front before now so I've never really had anything to compare it to to gauge how much to go for. Is 70t more than even most dedicated stunt bikes go for?

-1 front and +2 rear on my old FZS600 did make it very stressed at casual motorway cruising speeds... but this isn't intended to be a bike I'll be using as a daily ride by any stretch of the imagination or taking it any real distances. It's intended as a cheap(ish) second bike to be a bit of a toy to have some silly fun with to see if I can learn to do a few stunt, doesn't matter if I bin it or get bored and revert it back to standard and sell to get most of my money back.

I thought KTM had come a long way I recent years? Especially with the engines? I was lead to believe the lumps are solid and that they really only suffer from the odd electrical gremlin here and there? But then again, compare anything to a Jap bike and it's never going to hold a candle to it, is it?

I haven't done the deal on the bike yet so it's not too late to change the bike, or back out of the idea completely, I just thought a 390 Duke seemed logical as it's the lightest 40bhp+ bike you can get in my price range (ideally under £2k, the one I'm looking at is £2.5k, I'd stretch to £3k if I had to). I had done my research and had this bike at the top of my shortlist and it helped that someone I know just happens to have one for sale having just upgraded to a Street Triple having turned 21. Plus an added bonus is as it's an A2 bike my son can ride it (with or without the dinner plate) when he gets back from Canada in a couple of months if he wants to as he's 19 now and passed his A2 and sold his 125 before he went to Canada for 6 months. Although he has ideas of getting a CBR500R when he gets home but if there's already a 390 Duke sat on the driveway he may settle for that instead and save himself a few bob.

So yeah just all seemed to stack in favour of a 390 Duke really.

Was originally looking at a 2007+ Hornet 600 as, again, it's pretty much the lightest bike in its power range you can get, so that might be an option if you think the KTM is a poor move? But still, despite it still being a very light bike it's still got about 25kg on the Duke so the KTM just seemed the better option.

Also considered a DR Z400 but they're not plentiful on the used market and can't find anything in my price range. Looking minimum £3.5k for a 2008 model as opposed to £2.5k for a 2015 390 Duke. Duke is only a couple of kg more and a fair bit more powerful and torque than the DR Z400. 7 years newer for £1k less. No brainer???

Also considered a WR450. Hens Teath!

I've since discovered that SuperSprox advertise on their website that they will custom make any bespoke sprocket with any tooth count up to 70t for any bike. As well as two Polish companies I found on Ebay but they only make standard steel sprockets but they're relatively inexpensive at around £70. So yeah, waiting for SuperSprox to come back to me with a quote for a bespoke sprocket, if it's under £200 for a Stealth I'd consider that a good price for a bespoke engineered item.

But yeah going back to your original question, am I sure I wanna go up 25t? I don't know, I was hoping you could help me there now I've gone into a bit more detail as to what I'm expecting from the bike. But given that custom made sprockets are an option, the sky is the limit regarding number of teeth.
____________________
2024 BMW M1000XR
2021 Honda CMX500 Rebel S
2016 Triumph Tiger 1050 Sport

[xX-Alex-Xx]

[TravisBickle]

[ThunderGuts]

[TravisBickle]

At the very least the chain guard will be going for a Burton LOL.
____________________
2024 BMW M1000XR
2021 Honda CMX500 Rebel S
2016 Triumph Tiger 1050 Sport

[xX-Alex-Xx]