Fixie chainline discrepancy?
Bennett Fischer
of a fixed gear bike I recently bought, that I hope someone in this
forum can explain to me.
The bike came with a Surly fixed gear hub (120mm spaced out to 130mm).
Now, according to the Surly speccs, this hub gives you a 42mm chainline
in the rear, which should match up pretty well with the 103mm bottom
bracket the bike came with. (Sugino 103mm bb, with a Sugino RD crank.)
However, using Sheldon Brown's chainline formula - front chainline =
distance from center of downtube to front sprocket (44mm); rear
chainline = 1/2 frame spacing, minus the distance from the inner edge of
the fork end to the rear sprocket (49mm) - I get a difference of 5mm,
which indicates that I'd be better off with a 113mm bottom bracket.
<http://sheldonbrown.com/chainline.html>
Any clues as to why the discrepancy between my measurement - 49mm - and
the 42mm spec from Surly? Am I doing something completely wrong? I
notice there's an approximately 5mm spacer built into the rear sprocket
- is that a common feature to all fixed gear sprockets? Before I change
the bb, should I just flip the rear sprocket to place the spacer on the
outside (and the sprocket teeth closer to the spokes)?
Cheers,
Bennett Fischer
Brooklyn, NY
Norman
So long as your additional spacers are the same width, the
Surly should give a 43 mm chain line. Are you sure some of
the discrepancy is not from measuring to the inside of the
front sprocket and the outside of the rear?
Th' General method that I use is:
1) Double check the rear spacing to 0.1mm
2) Know or measure the down-tube diameter.
3) Measure the front sprocket to downtube from the center of
the sprocket & calculate to the center of the downtube.
4) Using the number from item (1) above, center the measuring
device on the rear hub center @ some value & find the center of
the rear sprocket.
5) Subtractively arrive at rear chainline & compare.
You can treble check by laying a know good straight edge against
the front sprocket & seeing how close it lines up with the rear. Do
use more than one rotational position for the front sprocket, though.
Notes: the first sprocket I measured (some old suntour 16
tooth for 1/8" chain) looks to be ~2.8mm wide. 4mm of variance
in chainline doesn't seem important, at least anecdotally (NB all
evidence is anecdotal). Bicycles can have rather poor alignment,
use the Sheldon brown method to check that if your numbers are
not coincident with your observations.
Norman
> Any clues as to why the discrepancy between my measurement - 49mm - and
> the 42mm spec from Surly?
I just recalled another bit from an old thread on
fixedgeargallery.com:
some leaky-headed assemblers have been known to sneak spacers
under the sprocket.
Qui si parla Campagnolo
What does it look like? Is the chainline parallel to the top tube?
Have a look and go from there, instead of relying only on the
equation.
Lou Holtman
>
> - Tekst uit oorspronkelijk bericht weergeven -
Excactly. Or better hold a straight ruler with sufficient lenght
against the outside of the chainring towards your sprocket.
Lou
thirty-six
What about a bent ruler, or a taut string?
String is simple. Trap it under the chain at the front srocket and
turn the wheel, holding the string against the chainstay until the
string bridges the front sprocket. Keep the string taut and move it
in and out to check the natural chainline for the front sprocket. If
it is within 1/8" its perfect. 1/4" error is acceptable, 1/2" NEEDS
correcting.
Lou Holtman
I said a straight ruler. Geezz.
> String is simple.
It is also a good method, but it needs meer eyeballing/
fingerspitzengefuhl.
Lou
thirty-six
Do you usually have a bent ruler?
dusto...@mac.com
> String is simple. Trap it under the chain at the front srocket and
> turn the wheel, holding the string against the chainstay until the
> string bridges the front sprocket. Keep the string taut and move it
> in and out to check the natural chainline for the front sprocket. If
> it is within 1/8" its perfect. 1/4" error is acceptable, 1/2" NEEDS
> correcting.
"Perfect" is "where you can't measure an error", not 1/8" off.
Quickly adding: I've ridden FG bikes with at least 1/2" "off"
chainlines, and gotten away with it-- low gears down steep hills and
all.
Ignorance is bliss, as long as the chain stays on. I might have gotten
away with that because I was riding a "converted" touring frame with
17-1/4" nominal chainstays, where the middle of the long DO actually
gives 17-1/2", axle to BB centers.
Since those days, having seen a tossed chain or two at the track,
where pista-specific equipment is used and one might think that
chainline disparities would be largely, if not entirely, thus avoided,
I've converted to "perfect" as in "best I can measure", backed up by
sighting along the chain from cog to chainring (nod of agreement with
Lou H), both of which I freely admit worked a lot better before I
started having to wear reading glasses <g>.
Straightedge, string, laser device-- whatever. Get the thing measured,
and WTH, make it right and then you don't have to worry about it.
Also: straight means your chain and cog/chainwheel teeth will last
longer, and the chain will run quieter, too.
John Dacey's Business Cycles chainline page:
<http://www.businesscycles.com/tr-refspec.htm#chainline>
Sheldon "The-Good-That-Men-Do-Sometimes-Outlives-Them" Brown's
reference:
<http://www.sheldonbrown.com/chainline.html>
My current fixed, a Hans Schneider "sprinter" pattern, old school pro
frame, OS steel and steep and tight, has a chainline that was checked
by me and then checked again by a real pro mechanic with many years'
experience. It's quiet, smooth, and has been real fun to ride the last
couple of days since I declared Fixed Gear Season was open at this
address.
--D-y
thirty-six
> "Perfect" is "where you can't measure an error", not 1/8" off.
When the sun's shining 1/8" off is perfect, see you at dusk.
> Quickly adding: I've ridden FG bikes with at least 1/2" "off"
> chainlines, and gotten away with it-- low gears down steep hills and
> all.
But they wear unecessarily.
>
> Ignorance is bliss, as long as the chain stays on. I might have gotten
> away with that because I was riding a "converted" touring frame with
> 17-1/4" nominal chainstays, where the middle of the long DO actually
> gives 17-1/2", axle to BB centers.
Test that the chain stays on by spinning the wheel and pressing a cone
spanner against the side of the chain in an attempt to derail it.
dusto...@mac.com
> When the sun's shining 1/8" off is perfect, see you at dusk.
I don't understand that.
> But they [badly aligned chains] wear unecessarily.
People say "timer" chains on tandems can last many thousands of miles.
Do proper alignment, no shifting make the difference, compared to
derailleur chains?
> Test that the chain stays on by spinning the wheel and pressing a cone
> spanner against the side of the chain in an attempt to derail it.
I forgot to mention that sometimes I ran those old worn chains pretty
loose, too. I think I might have had one "try" to come off (clack
clack clack) during some high-rev pedaling over a lumpy surface.
Perhaps a "manufactured memory" from close to thirty years ago. I
never "lost" a chain on the road or on the track, is the good part.
That's all fine; like I said, now the chainline is as close to
identical distance from center on both ends as I can "see". And
carefully adjusted for tightness/looseness, too. And yeah, even though
I would think having a longer stay would mean more "offset" if the
chainline were off, the bikes I've seen toss chains were short-stay
track bikes, and that's what I'm fixing it on these days. So:
"Reasonable precautions".
I have no idea what caused the derailments I saw at the velodrome. The
trackie culture is picky picky picky IRT to setup, from what I saw.
But, I did hear a few bikes that "clacked" when ridden, which might
indicate something (very) out of line.
It might have something to do with at least some track chains being
the old pin/bushing style, said to be stiffer laterally than the newer
bushingless "road" chains-- but then again, so were the Regina black
and gold chains I used back in the day (made with bushings and pins).
Perhaps someone who knows will expound...?
--D-y
AMuzi
Hard to say definitively but yes all of those things. And
more. Had a Peugeot fixie in today with a "clicking noise".
Seat tube was broken clean off the BB shell.
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971
thirty-six
From what I heard locally in recent years, al lot of track riders have
switched from 3/32", which was previously common due to duality of
144pcd to road and track, to 1/8" chain because they were breaking the
chains. Problem was, they were using the bushingless type and I
suspect the failures were down to joining chains poorly using a
rivetting tool. When using a 1/8" chain the standard split link is
the joining medium. Of course the track racer will be keen to point
out that they are strong enough to break a 3/32" chain. The noises
you heard may have been from an overtight chain, which is not a danger
in itself. BTW it appears that 1/2" Regina chains are stilll
available, marketed through the motorcycle trade.
Qui si parla Campagnolo
Altho not a ton of velodromes around here I have worked on bikes like
those belonging to Colby Pearce and I have never seen a track bike, a
bicycle actually used on a track, have a 3/32 chain. In addition,
there are no current 3/32 chains that are connected using a rivet
tool. All have a snap/master type link. PLUS 144 BCD is common in
terms of chainrings, both 3/32 and 1/8 inch. AND you can use a 1/8
inch chain on a 3/32 chainring, no problem.
dusto...@mac.com
(I invited):
> > Perhaps someone who knows will expound...?
(AMuzi replied):
> Hard to say definitively but yes all of those things. And
> more. Had a Peugeot fixie in today with a "clicking noise".
> Seat tube was broken clean off the BB shell.
"That would be 'more'" <g>.
Thank goodness (the force that watches over us) for happy landings in
the bike shop.
--D-y
thirty-six
But being cheapskate, we used the cheapest setup which was to use road
chainrings (because someone in the club had them spare) and cheap
3/32" chain (which was bushed) still available three years ago, I was
given a few (was offered a gross) chains when a shop was closing. The
shops catering for the racer dwindled and the survivors did not
realise the significance of the cheap chain to the local track
riders. With only the bushingless chain or the 1/8" chain available,
the riders persisted with what was common at the local track until
this problem of snapping chains came to light. It had not been
experienced with the cheapo 3/32" . The track is now gone so I expect
that a standardization nationally on 1/8" chain to take place for all
disciplines in track use. Only sprinters would see the 1/8" chain as
essential. All other riders accepted you could use either and the
cheap chain could be replaced with a new one each week if desired.
The local track was open and not boards so an overly lubricated chain
was not seen as a problem.
dusto...@mac.com
wrote:
(first, thirty-six wrote):
> > From what I heard locally in recent years, al lot of track riders have
> > switched from 3/32", which was previously common due to duality of
> > 144pcd to road and track, to 1/8" chain because they were breaking the
> > chains. Problem was, they were using the bushingless type and I
> > suspect the failures were down to joining chains poorly using a
> > rivetting tool. When using a 1/8" chain the standard split link is
> > the joining medium. Of course the track racer will be keen to point
> > out that they are strong enough to break a 3/32" chain.
When I was racing (ahem) at Alkek (Houston V-drome) late 90's-2001,
1/8" seemed to be the overwhelming favorite.
One racer, I believe J-Me Carney (IMS) was known to use 3/32". The
night the "circus" of all the nats-level racers came to town and I
could get away at all, my time was v. limited and you bet I would have
loved doing equipment inventory in the pits.
I saw people I was racing against using chain tools when necessary;
part of the deal with carrying lots of cogs and chainrings was getting
a gear you wanted in a chain length that obviated breaking the chain.
One frame builder in particular, Hans Schneider, began using rear fork
ends with longer slots, which help in solving that problem.
I didn't see anyone using a split link but they well could have been
doing so.
Since then (ten years ago) there might have been a movement, at least
in some locations, toward the 3/32" "road" chain, and then a movement
back! <g>
As this general subject has been discussed here, I've seen at least
one reference to breaking point comparison between 3/32" and 1/8"
chains and IMS, there is very little or no difference; also, the man-
mountain kilo-dude leaving the blocks in his mongo big gear would put
a lot more actual "breaking power" into his driveline if he were on a
mountain bike trying to climb a steep hill in a little twiddle gear,
from demonstrations I have seen here.
That run of chain breaking might point to "poor assembly" and,
perhaps, a weakness in bushingless chains-- that (just supposing) they
lose reliability with (repeated, by luck of the draw, on the same
link?) pin removal.
> > The noises
> > you heard may have been from an overtight chain, which is not a danger
> > in itself.
Bingo, I forgot that possibility, maybe due to the observation of
*most* riders being very careful to adjust chain tension.
I never saw anyone break a chain due to over-tightness per se, but was
sternly (if kindly) warned about this by at least a couple of Alkek
regulars, people with experience at other 'dromes and some years "in
the game". IOW, a chain that is properly adjusted except for one v.
tight place can and will break, was the word.
Yes, I joined the long, long line of newbies who have looked,
thunderstruck at their instructors: "It's [chainwheel] not ROUND???
It's THAT BAD out of round???" Then rotated their pedals, and learned
of the base truth of this imperfect world we live in. St. Tulio, with
feet of clay...
(Peter replied to the call for expertise and waited as D-y stuffed two
replies into one):
> Altho not a ton of velodromes around here
Well, there is one pretty close, at the Olympic Training Center in Co.
Springs, I believe <g>.
> I have worked on bikes like
> those belonging to Colby Pearce
(excuse me) who has a pile of USA nats championships and placings
> and I have never seen a track bike, a
> bicycle actually used on a track, have a 3/32 chain. In addition,
> there are no current 3/32 chains that are connected using a rivet
> tool. All have a snap/master type link. PLUS 144 BCD is common in
> terms of chainrings, both 3/32 and 1/8 inch. AND you can use a 1/8
> inch chain on a 3/32 chainring, no problem.
Track racers are a pretty hidebound bunch; there was still a lot of
steel rolling stock ca. 2000, and toe straps, too.
J-me (again IMS) was kinda underground-famous for using 3/32" chain,
very unusual at the nats level per my understanding.
I'm way behind on 'drome info. Is the Boulder track up and running?
Last I saw they were still building.
Dang, I miss the velodrome, even leaky old lumpy Alkek. Riding my old
track bike on the road the last couple of days, even with a stupid-
long rake road fork stuck in it, what a joy. "It's not twitchy, it's
responsive" --D-y
thirty-six
> Dang, I miss the velodrome, even leaky old lumpy Alkek. Riding my old
> track bike on the road the last couple of days, even with a stupid-
> long rake road fork stuck in it, what a joy. "It's not twitchy, it's
> responsive" --D-y
Put a 24mm traing tub on it. Less tyre pressure lengthens the contact
patch and reduces unrequisite steering.
dusto...@mac.com
Good idea. It's got a 23 clincher on it; there's room in the road fork
for "bigger yet", too.
But, to tell you the truth, that bike is so sweet with the fork Hans
made for it, that (ah, memories!) I've become somewhat dissatisfied
with the road fork setup. Still a favorite ride as is, FWIW, but...
The OEM fork has enough room for a 23 tire, but the crown isn't
"thick" (not enough surface area) to drill for a brake, at least on
something I'm going to ride <g>.
Plus, the brake bridge is just a light tube, not drill-able, per me
and Mr. Schneider, too <g>.
In addition, the tire clearance between chainstays is very tight.
I can ride it as is, just for shorter rides on flat ground-- no rear
brake means safety and control issues in my book, which I am not open
to discussion about <g>-- and it's great. But this bike needs sewups,
and brakes on both ends.
I've seen reference to Dia Compe brake adapters, and seen pictures of
Japanese Keirin bikes set up with a couple of variations of factory-
made clamp-on brake mount adapters. Plus the home-made alu plate
adapters for rear brakes. Any port in a storm but I'm thinking the
factory-made equipment is the way to go.
(anyone): Is there a source for these? I've looked, a couple of times.
Hen's teeth.
BTW, I have an early-90's Tommasini Prestige (SLX tubes) that has
begged for years to be the "road fixer", for personal use plus going
on the larger, longer group weekend fixed rides that are about to
crank up here in Austin (Tx). Pump peg, water bottle bosses, holes for
brakes f/r, clearance for 23mm + clincher tires, and whatever else I
forgot on the list. Mostly, water bottle cage mounts, so I don't have
to clamp onto that nice red paint, or look like a tri-geek out there
with bottles on a saddle mount, which also might get in the way if I
need to eject.
Thanks for the suggestion on tires, obviously I was thinking "the
other way", and wth, that larger tire might work very well.
--D-y
Zog The Undeniable
Don't measure it. Get a long steel rule and check it properly. You'll
need to take the chain off, and ensure the rule isn't sitting on a
chainring bolt. use the straight edge, NOT the wide flat face. Within
2mm is OK with a bushingless chain, but with a full bushing chain you
want it to be within 1mm.
thirty-six
The bigger tyre contact up front just means you have to make forced
turns instead of relying on the bikes natural tendency to fall into a
turn. It's a much more comfortable way to learn the skill rather on
criterium geometry with a 20mm HP which will wobble and dive if you're
not paying attention. As for brakes, if you have a box crown, use it
to locate your thumb and pivot your hide gloved hand onto the tyre.
And you could fit a spoon brake, which is just a lever doing the
same. I can remember someone using their fist as a wedge to brake
rotating it to vary the force but I can't imagine this being
particularly safe when clearance is tight.
thirty-six
>
> > When the sun's shining 1/8" off is perfect, see you at dusk.
>
> I don't understand that.
The bike is for road use not competetive, so if the sun is shining
ride the bike instead of tinkering. If it is well oiled and has a
little slack in it, only then should you consider adjusting chainline
after it has proved to be a problem. Incorrect crank bracket
alignment or facing may throw out the chainline anyway, so dont be
fussy, if it works well, there is no need to adjust chainline. Just
check for wear on the sprocket sides. If it shows up, fix the
alignment.
> People say "timer" chains on tandems can last many thousands of miles.
> Do proper alignment, no shifting make the difference, compared to
> derailleur chains?
Derailler chain can last, use lower gears and grease the chain by
liquifying it, using heat, to penetrate the chain.
dusto...@mac.com
> The bike is for road use not competetive, so if the sun is shining
> ride the bike instead of tinkering.
"Tinker"? Nope. Adjust for correct chainline.
> If it is well oiled and has a
> little slack in it, only then should you consider adjusting chainline
> after it has proved to be a problem.
That can be done. Correct chainline is better.
> Incorrect crank bracket
> alignment or facing may throw out the chainline anyway,
Don't cop out here, OK? We're talking "correct", in spite of
variables.
> so dont be
> fussy, if it works well, there is no need to adjust chainline. Just
> check for wear on the sprocket sides. If it shows up, fix the
> alignment.
Am fussy already. Would rather prevent wear. And noise.
> > People say "timer" chains on tandems can last many thousands of miles.
> > Do proper alignment, no shifting make the difference, compared to
> > derailleur chains?
>
> Derailler chain can last, use lower gears and grease the chain by
> liquifying it, using heat, to penetrate the chain.
O my god, did I say I don't get long life out of my derailleur chains?
"Use lower gears"? "Use heat"?
I really really should have kept accurate mileage counts on those two
(Bike A, Bike B) Campy 9sp chains I used for years, with simple
TriFlow (which some people do not like) application-- generous, not
ridiculous-- and thorough wiping off.
The chain on Bike B checked out fine for "stretch" but eventually quit
shifting well. Inspection showed looseness to the sides. Replaced
chain, and the new one worked 100% fine on the same cassette cluster
as the old one worked on-- showing little-to-no wear on cassette teeth
due to running a "stretched" chain. Thousands of miles, close to three
years on Bike B, maybe longer on Bike A (A+B= my "road fleet").
On Bike A, the chain "broke" on the beginning of a steep hill. First
"real" chain I ever had fail, going back to the early 70's. The first
first chain that failed was on a Schwinn Varsity ca. 1968, doesn't
count due to "Schwinn Approved" factor. I think I might have snapped
a pin; it looked kinda short while I was removing that link and
shortening what was left to get home-- or actually, to the shop for a
new chain. Set the thing down carefully but it disappeared in the
grass. Obvious conspiracy by Campagnolo.
I thought that might have been an installation problem (by me),
although, as I said, I've never had a problem previously. Both bikes
have master links now, "just in case" and it was time to hop on the
bandwagon at long last anyhow.
This method uses shop rags (which I just wash in the clothes machine
and rinse well, both rags and machine <g>), and some elbow grease.
Both bikes get an occasional wash at the mechanic.
Not "fussy" there.
Chainline? I like "fussy". YMMV, as I stated from the outset or
pretty close in this thread. --D-y
thirty-six
> O my god, did I say I don't get long life out of my derailleur chains?
> "Use lower gears"? "Use heat"?
> I really really should have kept accurate mileage counts on those two
> (Bike A, Bike B) Campy 9sp chains I used for years, with simple
> TriFlow (which some people do not like) application-- generous, not
> ridiculous-- and thorough wiping off.
> The chain on Bike B checked out fine for "stretch" but eventually quit
> shifting well. Inspection showed looseness to the sides.
I've reset the rivets recently on an old Shimano bushed chain which
shows next to no wear (not counting plating removal and sideplate
wear). I think I had stopped using it for racing because It didn't
snap down onto the sprocket too sprightly with the grooved tip
sprocket teeth that Regina used. It was relegated to winter use and
was used with heavier lubricants hence, no wear. Looks poor even
without all the crud that it picked up in winter. Its the best chain
on flat tipped sprockets (compact 7) I have used. It ran best,
greased.
dusto...@mac.com
Yo, I had one, an old bushed Euro make if I can't remember which exact
one, that saw "rugged duty" on my foul-weather fixed bike, total
mileage unguessable. Heavy sludge lube accumulation.
One day, I felt a "tight link". The very first link I checked was
tight.
"Can't be that lucky". Nope. Kept checking, every single link was
tight. No visible rust or corrosion, even with links moved in a wide
arc of motion.
"Gave its all" <g>. --D-y
Bennett Fischer
Thanks all -
I have to admit that in my impatience, I did swap in a 113 bb, and
eyeballing the chainline using the top tube as a reference... it looks
fine with both the 103 bb, and the 113.
Now I'll dig up a good metal straight edge and check it that way.
thirty-six
> In article <he6ilm$kk...@energise.enta.net>,
Unless you sight it with a line first you'll need two other straight
edges to check they are straight which it wont be because you'll have
slammed the spade into it when you were digging. Best to just use the
line.
dusto...@mac.com
> Thanks all -
>
> I have to admit that in my impatience, I did swap in a 113 bb, and
> eyeballing the chainline using the top tube as a reference... it looks
> fine with both the 103 bb, and the 113.
>
> Now I'll dig up a good metal straight edge and check it that way.
Measuring off the center of the seat tube and the centerline of the
rear hub, if the frame is aligned, will show you if the chain is away
from the CL of the frame at chainwheel and cog.
You can find the distance for the cog CL to be from the inside of the
rear dropout with some measurement and arithmetic, too. Yes, I
prescribe "fussy" <g>.
IOW: Eyeballing is good. Measuring, too. "Do both" (OK, have repeated
and will quit for awhile).
I had an old metric steel measuring tape that cracked. I broke off
several lengths at convenient centimeter markings. Those shorter
lengths are easy to handle and measure with.
Hope we haven't told too many stories on your time, BF.
--D-y
thirty-six
A suitable 3/32" chain would be the Wippermann Connex 7R8 which comes
with a traditional reliable spring clip, not a snap off, connector.
You can use this chain no problem on 3/32" sprockets. This may be
most appropriate for low gear riders where the grease drag at high
chain speeds may be seen as a restriction (??). Never mind, suffice
is to say, you can use 3/32" on the track and this is the chain to do
it with.
Qui si parla Campagnolo
Hipsters use 3/32...are you a hipster..are you one of these guys?
http://www.xtranormal.com/watch/5684963
thirty-six
my hipsters are 34/32. That sounded like my nearest bike shop when
the fashion BMX kit was brought in.
dusto...@mac.com
wrote:
> Hipsters use 3/32...
Well, 'round here it seems 1/8" is hip.
However, some local groaty old roadies use 3/32, too. One guy I know
of uses 52x21, .2" bigger (66.9 v. 66.7") than the 42x17, nominal (27"
wheel). Which he says runs smooth, and people tend to look at the
chainwheel first, you know?
I look somewhat askance at the hipsters but do take the odd
opportunity to strike up a conversation with them, which includes "the
first time I rode a fixed gear, you probably weren't born yet"-- only
in the nicest way, of course <g>. --D-y