marqueemoon

Does it help to use a bigger ring/cog combo?

abeyance

It works if your measurements are precise, but as soon as your chain starts stretching, it's time for a new chain or a new gear combo. Good luck with it, as I had to change the setup every 500 miles or so.

wroomwroomoops

Your questions have been answered, in detail (just enter your numbers):

https://eehouse.org/fixin/

exfreewheeler

I did a search... what is it?

Like using a 48T with a 18T?

Because I was considering it.

iVAG

what is a magic gear?

mander

A gear combo that gives perfect chain tension in vertical dropouts.

dirtyphotons

yes, it will work. measure carefully and your gear will have a very good tension. at first. if it's fixed, the chain will get loose too quickly for your taste. for ss, if you're ok with some slack, it'll probably last long enough to work.

when i was running a magic gear, i put the new chains on that. when the chain stretched out it went on the horizontal dropout conversion, still it's not ideal. over the long term of the wheel it's probably cheaper to go with the eno.

as far as chain wear, i think it would make sense that a larger drive train would last longer. larger radius means less tension on the chain, and longer chain means more links to share the wear.

for those who asked, here's an explanation of a magic gear

wroomwroomoops

I agree with everything you said, except the part I highlighted in red and italic: it is true that in a longer chain, each link will wear less, but this is offset by the fact that the wear will happen on more links, causing the exact same stretch in total! The two effects are quantitatively identical, so there is absolutely no difference in total stretch.

dirtyphotons

that's an interesting idea. i would need to think about it some more, but i can see where you're coming from.

mascher

Coming up on a year that I've been running a magic gear on my snow/rain/quick run to the dep bike, a Kona Smoke - 44x18 is a perfect fit. The chain tension does slacken a bit once the chain starts wearing, but I'm running a ring with an extremely tight spot on a cheapo sugino biopace ring, and it's still not enough to cause enough slack to raise the alarm, though it sure is bad for drivetrain wear.

When I first set it up I was running a really worn chain from another bike in the snowiest/saltiest month ($/chain conservation since it was going to be trashed that month either way), and shortly after that I started throwing chains. Put on a new one, no problem.

If you have "semi-horizontal" vertical drops (shallow ones that aren't exactly perpendicular to the chainstay, as is pretty common) you even have a couple of mms of adjustment, even better. I start off near the front of my "semi-horizontal" drops with a new chain, and by the time the chain's about done, the axle sits right in the drop.

I say ignore the naysayers and run it - you'll only wear stuff out faster and have tension problems if your chain is too tight (like mine is). Chain tension is the most overrated factor on a fixed bike imo, and it seems a lot people here must run their chains way too tight. It has to be tight enough to not get thrown, that's it.

Mike552

I ran a 40:14 on a CAAD4 Cannondale with vertical drops... worked really well, but as others say... 500 miles, 800 maybe before you get a new chain. FWIW, I commute short distance, so it never bothered me.

dirtyphotons

ok, i thought about this a bit. the longer the chain, the more links share the load. but the more links sharing the load, the less each one must stretch in order for the chain to become "too slack" by whatever definition one uses.

while i'd need to know a little more about exactly how a chain wears to quantify these, i'll agree that they have a cancelling effect and it is quite possible that they negate each other entirely, meaning the benefits of more links would be zero.

so this negates the argument that the length of the chain itself will be a real benefit in preventing wear in larger drivetrains. however, i still think that a chain will wear slower with a larger drivetrain.

at a larger radius, the amount of force required to turn the wheel at a given torque will be less if the chainring and cog are larger (in this case, torque = force times radius). so there will be less total force on the chain (applied over approximately the same distance and number of revolutions).

then there's the fact that with larger chainring/cog radius, the chain has to bend at less of an angle, as it wraps around them.

so this makes me think that a larger drivetrain will still cause the chain to stretch more slowly.

this is independent of (but related to) larger drivetains having longer chainring and cog life.

wroomwroomoops

Instead of answering point by point, I will simplify by just telling you how things seem to be.

At given sizes and type of chainring and sprocket, type of chain - and amount of pedaling: if the chain has n links, each of these links will "take" an equal amount of wear force W/n. Each link will stretch a bit proportionally to that wear force:
S1 = (W/n) * r
(r is some factor that depends on the geometry and material of the chain - the "softer" the chain, the higher the r). The total stretch
Stot = S1 * n = (W/n) * r * n = W * r

So, the total stretch depends on the wear force, and the proportionality factor r, which itself depends on the type of chain (geometry, construction, materials, lubricant). The wear force W depends on the strength of pedaling (basically, the energy "given" to the chain) and the drivetrain (size of chainring and sprocket). This is where I agree with your latest post: the larger the chainring and sprocket, the lesser the wear. Other factors that affect the wear force are chain tension and lubricant. (remember, r is a proportionality factor, that will determine how much will the chain stretch under the effect of a given wear force).

To return to the formula above, we get Stot = W * r which does NOT depend on the number of links in the chain - Q.E.D.


EDIT: due to my sometimes poor english skills, I think I might have used a less-than-ideal term for W. Instead of wear force, I should call it wear energy. Sorry. Hope the picture holds, regardless.

dirtyphotons

agreed, pleasure geeking out with you.

i feel bad for jacking the thread marqueemoon, even though we were kind of addressing your question. i think we agree the answer is "yes, larger drivetrain will cause slower chain stretch"

rpc180

brain. spinning.

lymbzero

I find it's just a bit of trial and error.

Put the chainring you want on there..
And a casette in the back..

Wrap a chain on the chainring and find the closest matching cog on the back.

If your links match up.. Presto. Magic gear attained.
If they don't, try a half link.

A couple of different combinations will work.

When you find the right ratio, just replace the cassette with spacers and a single cog.

I recommend 1/8" chain and Non-ramped cogs.

Use a realitively unstreached chain for best results.

kemmer

Until recently, I had two super ghetto single speeds with the chain going from the middle ring to the cog in the cassette that lined up best, magic gear style.. One is my tall bike, the other is a bike I keep at the office for lunch/errand runs for days I take the train to work.

The tall bike gear wasn't as magic as one might like. I had to force the chain on and it was super tight. I just ran it like that for a few miles and it stretched out enough that it wasn't hard to pedal anymore. I rode it for a couple hundred miles after that till the bike broke and met an untimely end.

The other bike worked out nicely for a few months of regular but short rides. I started dropping the chain which prompted me to seek a less ghetto set up.

I think it's an ok way to go if you can't afford or don't want to spend the money on a better set up. If you're willing to put up with a little more maintenance, go for it. Don't used ramped cogs from a cassette like I did though. I'd prefer to drop a little coin on a more trouble free set up if you'll be riding it a lot. I don't log a lot of miles on the aforementioned bikes, so it worked out ok for me.

dirtyphotons

also, if it's gonna be singlespeed, you could try a ghost ring

idea credit: progre-ss

wroomwroomoops

The ghost ring works fine for FG as well - if you think of it, it's a symmetrical tensioner, it works no matter which way the tension is applied, top or bottom run of the chain.

I had a ghost ring in the first FG conversion I made. I actually love the idea of a ghost ring, as I can put a logo or some such artsy thing in/on it, and have it rotate in the middle of the drivetrain, while I pedal. The drawback is, well: fg drivetrains are dangerous traps for your fingers as they are. The ghost ring just increases the likelyhood of getting something in there - more chan/chainring trap hotspots.

If you have a SS, your options for tensioners increase dramatically. There are all sorts of tensioners that will work, attached to the chainstay, derailleur hanger, BB, whatnot. It's so much easier in the single speed world... For FG, the ghost ring is pretty much the only show in town, as far as proper chain tensioning is concerned, not counting lovely gimmicks like the ENO Eccentric (I luv mine) or eccentric BB - but for that you ought to have a special frame, and this whole thread started from the need of addressing the FG conversion of a normal vertical dropout frame.


OK, I think I yapped too much already. Thousands of years from now, these discussions will be found in some obscure archives, and the people of the future will wonder WTF was this, making such a big and sophisticated deal out of a discussion regarding a hobby.

marqueemoon

No problem. I'm enjoying the geekout.

mascher

Now there's something I've never seen mentioned. Is there any way to calculate what size to use, or do you just stick a sprocket in there and check your tension?

wroomwroomoops

I'm surprised, I thought it was, more or less, common knowledge among FG riders, since it is the only viable way to tension the chain of a FG.

Well, you know, there's always a way to calculate things I just wouldn't know how to, in this case. What I do is, I stick a worn sprocket in there, of size based on how slack the chain is, and then push it back towards the rear sprocket as much as it will go. If the teetch of the sprockets touch, then the size is too small. Sometimes the chain is so slack, you need a middle chainring.

Careful with your fingers, while you do this, expecially while you spin the pedals/rear wheel to see if the drivetrain is OK. If they get caught between chain and chainring, chances are they will be cut off, expecially with brand new machined chainrings (was goofing around with a Raceface alloy chainring once - that's the closest I came to have a finger severed, so far).

LóFarkas

If you want a ghost ring to work,
1)you need a lot of room in your frame. Narrow chainline and generously wide room between the chainstays.
2) Choose a nice big one. For my 16t cog, a ~26t or bigger was necessary. You want the chain to wrap around it at least a bit.
3) Tall teeth help.

Mine fell out a few times and it was a bit noisy. I was thinking about making a 1/8 ring for the purpose with tall teeth, which would have hopefuly cured both problems. But then I went for a wider chainline and I had to abandon it completely.

Perhaps the biggest cog from an old freewheel would work. A bit thicker than multispeed cogs, and I have one that has crazy tall teeth and no ramps. Only likely to work with 16t or smaller cogs and not too slack chains.

Of course something huge like a 32+ would likely work, but I think it looks like crap.