chucky

I would like to share with you an excellent technique I stumbled upon for tensioning single-cog drivetrains (ie with horizontal dropouts or eccentric/hubs).

First off, you are probably wondering that if we are talking about bikes with horizontal dropouts then isn't the only thing we need to do to tension the chain is pull the wheel back with the left hand while tightening the axle nuts with the right hand? Actually no, because one of the primary benefits of single cog drivetrains are that you can run the chains much much longer by really cinching down the chain tension to prevent severely stretched chains from jumping the cogs. By doing this you can forget about measuring chain stretch with a micrometer (and throwing out perfectly good chains) because no matter how stretched a chain gets you never have to discard it until it literally snaps in half! However, once the chain gets stretched to the point that it's more inclined to jump off the cogs than be driven around them, then you require a much higher chain tension than can be achieved with springs or arm strength and, until now, the only way I knew to achieve it was with a rigid tensioner/tug like this:


But chain tugs are annoying because:
1 Bikes almost never come with them...you have to buy it separately and many bike stores don't carry them.
2. They always come with the wrong kind of nut for tightening...you need a wing nut so you can cinch down the chain on the road without tools after it starts jumping off the cogs in the middle of your 30 mile winter commute on slushy salted roads. Of course you can get a wing nut, but it's yet another thing you have to buy separately and it's not like every store carries metric wing nuts.
3. The quality of many of the chain tugs is not good. I bought a fancy MKS chain tensioner for $30 and the tensioning bolt snapped in half before my chain did...kinda defeats the purpose of extending chain life when it would've been cheaper to just replace the chain.
4. There are many axle standards for single cog drivetrains...so whatever you end up getting probably won't fit without some filing. Even this is just a partial list of bike axle standards:
M9x1.0- Front "fixie" (fixed gear road bikes) hubs as well as front dynamo hubs (front generator hubs for bicycle head lamps)
3/8"x26T - Most axles including most department store bikes; Shimano Nexus and Alfine Series; 8FUN e-bike front hub motors. Very commonly used on road or bikes purchased at department stores that are not fitted with traditional quick release hubs.
M10x1.0 - European brands, SRAM, NuVinci, Rohloff; Bionx and some e-bike front hub motors
3/8"x24T - Rear coaster brake hubs found commonly on beach cruisers, Huffy brand, older department store bicycles.
13/32"x26T - most Sturmey-Archer hubs.
M12x1.25 - E-bike hubs - either 1.0 or 1.75 thread pitch.

So what's the alternative? Well, the technique is quite simple:
Step 1- Tension the chain with arm strength by adjusting the axle position with your horizontal drops or eccentric BB/hub. Carefully observe the axle position so you can remember it.
Step 2- Take the chain off the cogs, move the axle to a slacker position so you can derail the chain or whatever you gotta do to get it off.
Step 3- Set the axle position or eccentric a couple mm tighter than you had it in step 1.
Step 4- Put the chain around the rear cog and over a few teeth of the front chainring. You won't be able to get it all the way around because there's not enough slack.
Step 5- Turn the cranks with your hands so that the chain is forced to climb itself all the way onto the chainring. You might have to also push the chain laterally hard with your other hand to help guide it just as a derailleur guides a chain towards climbing onto a bigger cog.

So, in summary, the technique is to set the axle to a position that would be a little tighter than you can tension by hand (without the chain on) and then force the chain to climb onto the cogs like a derailleur does. With this technique you can get the chain just as tight as any rigid mechanical device.

Looigi

Sounds all wrong to me and rather brutal. The chain shouldn't be tight.

Airburst

+1. Running chains tight can't be good for the rear hub and bottom bracket bearings, let alone the driver bearing in an internally-geared hub or one with a coaster brake (I know those sieze up under high chain tension, I've seen it happen)

I suspect your technique could bend a chainring or even break a tooth off the sprocket or chainring. Not to mention the potential for snapping the chain or possibly a chainring bolt.

Also, how do you propose using a chain tug on a forward-opening dropout?

chucky

This is true, but the chain is what gets the most abuse because it is exposed to the elements. Therefore, it's better to shift the burden to the enclosed portion of the drivetrain which will last longer even with the extra load.

In other words, you're damned if you do and you're damned if you don't...so you might as well address what is literally the "weakest link".

Steel cog? No way. It's possible with an aluminum chainring, but I haven't had it happen and if it does they're about the same price as the chain you'd definitely have to replace if you don't take the risk. Besides, aluminum chainrings are disposable which you'll realize is silly for a single cog drivetrain after you go through a few with the same chain.

That's the entire point of this thread...chain tug is the my old method, but I am teaching you the new method I discovered when I couldn't use a chain tug.

Slack chain that gets replaced when stretched is the expensive method for spendthrifts who want to send the bikeman's children to college.

chucky

Much less brutal than using a derailleur. Just ask the drivetrain which lasts much much longer this way.

Airburst

Oops, I may have missed that bit, I thought you were saying to pull the wheel back with the chaintug before forcing the chain back onto the sprocket.
I've seen steel cogs that broke from being shifted under load on a derailleur bike. This has the potential to put a lot more force on a cog.



Well, you can throw away a Ł10 chain when it's worn out, or you can damage the bearing races in a Ł100 SHimano internally geared hub by adding stupid amounts of tension to the chain to try and run it when it's completely worn out.

Not to mention replacing a chain is a damn sight easier than replacing the hub.

Retro Grouch 

I've got an easier way:

When you try to tighten the drive side axle nut the hub wants to "walk" forward in the dropout so it's hard to keep the chain tension snug while doing this. I pull the hub back in the dropouts, align it with the frame and tighten the non-drive side first. After doing that it's easy to align the wheel in the frame with one hand and tighten the drive side axle nut with the other.

fietsbob

Yea kind of like RG ...

Tighten the left side nut, and then pull the rim sideways , to 'goldilocks' tighten the chain, take the slack out..
then tighten the right side nut, to keep it that way..

chucky

Definitely not. Sure the force is levered by the cranks, but it's arm strength pushing the crank instead of leg strength and you're not even giving it 100% like you are when pedaling. I think if there were any possibility of steel teeth snapping off then aluminum ones would not survive even a single trial as they have.

If it damages the races...I don't think it does. Maybe the bearings don't last as long, but the wear is far muted compared to the chain which otherwise becomes disposable due to being immersed in road grit. If you don't do it you'll probably go through 10 chains and 1 set of bearings instead of 1 chain and 2 sets of bearings. And in an ideal world all maintenance would be at once because time is money even if it cost less for the parts to replace 10 chains, labor is x10 (x10 cleaning off the bike, x10 going to the store, x10 getting out the tool box, etc, etc).

I've put many many thousands of miles (maybe tens of thousands) on a Sturmey Archer 8-speed using the chain tug method and it doesn't seem to have any damage and I don't expect different results from my new "force-on" method because the tension is the same.

Yeah I used to do that before I started experimenting with extending the life of stretched chains. It's not tight enough. When the chain becomes so stretched that it won't even seat on the cog at rest then it needs to be really really tight to stay on when pedaling.

And if you can get it tight enough pulling side to side then you're going to bend the wheel out of true....and every time you need to take the tire off for a full inspection of what's causing flats, etc

FBinNY 

+1,000,000

There's an entire new generation of folks riding fixed and free SS & IGH bikes. Unfortunately the basic knowledge about these systems was lost in translation and headed the way of Damascus Steel.

I shudder whenever folks discuss how to tension these chains, because you're not supposed to tension them at all. SS chains need to be slackened, with ideal slack (tension) such that at the tightest position (slack varies because the sprockets are usually slightly eccentric) there's free vertical play in either the upper or lower loop. I test for this by holding the wheel and rocking the crank forward and back and checking that I can see and feel the slack transfer between the upper and lower loops.

Running a tight SS chains is death on the hub's axle and/or bearings depending on how tight it is.

chucky

In my experience there's a lot of people that say you "should" do this or that when it comes to bikes and other things, but in light of actual testing in the real world it's downright shocking how often I find that you really shouldn't.

How many axles and/or bearings have you personally killed like this? Or is it that you were too afraid to ever try it because you didn't want to lose the basic "knowledge" that someone passed on to you?

All I know is that by increasing chain tension way beyond the advice of conventional wisdom I've not had any bearing problems and my chains have basically ceased to be disposable wear items. I've had bicycle use interrupted by worn out saddles, frame failures, out of true wheels, rubbing fenders, brake maintenance, etc but not by worn out bearings or worn out chains or bent axles. So if it is, in fact, "death to axles and/or bearings" then I can only conclude that "death" isn't even as much of an inconvenience as replacing brake pads and, therefore, not worth worrying about at all.

Was it Velocio who said about derailleurs: "it's brutal, but it works"?...well this may sound +1,000,000 wrong and brutal to you, but it works and in my opinion it works better than even the now ubiquitous derailleur concept which I personally find too brutal and wrong to use on my own bikes.

FBinNY 

I never ruined any chains, hubs or cranks by running chains too tight, for the very simple reason that I never ran chains tight.

There is a wealth of knowledge about what works and what doesn't. That doesn't mean anyone isn't free to do whatever he prefers, or think he's smarter that the cumulative knowledge of over 100 years.

I wouldn't have posted, but I see so many damaged hubs, and a few folded chainrings on SS bikes, which resulted from overly tight chains, that I felt folks were entitled to a counter balance to your opinion.

So now readers have both opinions, and are free to choose whichever method, or level of chain tension makes the most sense to them. That's the beauty of the forum format, lots of opinions, some agreeing, some disagreeing, and in the end each reader can make an informed decision for himself.

Continuity

Maybe do this as a 'get you home' technique, but after that, for Bog's sake - surely just spend the Ł5-10 on a new chain...

chucky

Thanks I appreciate your input and am not at all offend by opinions or observations contrary to my own because I never know which puzzle piece might be the one needed to complete the puzzle (even if it's not the one for the section of the puzzle I'm working on now).

As far as thinking that one is smarter than the cumulative knowledge of over 100 years...that's certainly not how I feel, I just think that a lot of times one person imitates another who imitates another who assumes (or claims) that the first person did it because they knew something even when it was just coincidence. So I simply don't agree that 100 years of practice is, in fact, 100 years of knowledge....not that I'm smarter than those who came before me as I'm sure there were many smarter than myself who simply didn't have the time to try everything; Doesn't mean they were dumb, but doesn't mean they knew better either, just that they ran out of time and left the rest to you and me.

And not to get too far off topic, but my professional background is in the science of analyzing information in general, and the accumulation of information is a difficult thing because what may be true in one context may not be true in another and I suspect that different tolerances in modern manufacture probably changes quite a few things when it comes to bicycles and undermines quite a bit of the knowledge that was accumulated under different circumstances. We like to think when we know something then that's the end of it, but when things change the unintended consequences can be very surprising and even when they don't change that doesn't necessarily mean that what remains true is still relevant to the best course of action at present (which is what I think is better to focus on rather than knowledge itself lest we find ourselves being right about the wrong things and wrong about the right things).

Ironically, as a rule, it is the use of knowledge that often renders that knowledge irrelevant or even invalidates it altogether (for example, the more we act on our knowledge of global warming the less important it is for us to know about it because we'd already be on the right course and if we act decisively enough to stop it entirely then what we know will no longer be true).

ThermionicScott 

Just speaking for myself (since I haven't yet worn out a bottom bracket, or broken a chain), the increase in drag from really cranking down on the chain tension is perceptible -- you can see it slowing down your free-spinning cranks on the stand. How much does a chain have to stretch before it'll actually skip on a single-speed? I didn't even know such a thing happened.

reddog3

Well Chucky- I'm glad you're in some science field, rather than working in a bike shop out there somewhere. I'm not gonna argue the benefits of proper chain tension except to say that "too tight" ain't good, and that's what you advocate. Ever heard the term "track slack?" There's a good reason for it. Then I get to the last paragraph in your last post, and realize that just maybe, a tight chain will slow global warming. Don't even go there dude!

vredstein

Your advice seems to be targeted towards
1. Those whose chain is already jumping off the rings frequently.
2. Those who don't want to find out the root cause of #1.
3. Those who can't install a bolt-on wheel properly.
4. Those who don't want to learn to install a bolt-on wheel properly.
5. Those who don't know how to maintain their chains.
6. Those who don't want to learn how to maintain their chains.
4. Those who want to delay replacing their chain until the point where is snaps during a ride.

I suppose any discussion of bottom bracket or rear hub bearing wear is irrelevant to this target audience.

Sixty Fiver

Wow... just wow.

The mind boggles.

Sixty Fiver

So in summary...

As very few drivetrains have perfectly round chain wheels and cogs the tension on a chain varies throughout the rotation of the cranks and by setting the chain as tight as possible one might turn the cranks over to find that the chain tension is suddenly high enough to shift the drive side axle.

Running a chain at maximum tension also decreases efficiency as it prevents one from being able to turn the crank over smoothly because the drivetrain will bind... in essence the chain becomes too short for the drivetrain when it hits the largest point in the rotation the chain will then be working to pull the chainwheel and cog toward each other. If the wheel does not shift out of place the stress will be transmitting to the other drivetrain components.

It will be noisy.

Using this method to prolong the lifespan of what is a consumable item will cause premature wear on the chainwheel and rear cog and can affect the bearings in the hub and bottom bracket due to the high repetitive stress on these components.

Running a chain until it snaps due to fatigue is not a safe practice.

I typically see 8000 - 10,000 km of life from a single speed chain which costs $10.00 - $15.00... at this point the rear cog is usually too worn to handle the new chain but a good chainwheel can be run 40,000-50,000 km.

I am constantly tuning up people's bikes and showing them how to adjust the tension on their singlespeed and fixed gear bicycles... they are always pleasantly surprised on how much better their ride becomes when they have been shown how to tension a chain properly and proper chain tension includes a small degree of slack.

This is something folks figured out well over 100 years ago when chain drives were first utilized on bicycles.

sreten

Hi,

It is a lot better than your ill-informed opinions you
are trying to masquerade as some form of knowledge.

What is clear is you don't seem to know the fundamentals
that have existed for a 100 years that your trying to rubbish.

There is very little point trying to discuss something that
is basically wrong in many ways, with someone who is
being disingenuous in all their "arguments".

rgds, sreten.

jsharr

So do one use wrenches to do this work, or just bang on the axle with a large rock to knock the wheel back until the chain is way too tight?

Retro Grouch 

I'm calling BS and here's why:

Chainrings on the crank are rarely perfectly concentric. If you tighten a single speed chain too much, it'll make a popping noise and the crank will be noticeably harder to spin at one point in the rotation.

FBinNY 

Some times it's better to let go. This is a open forum and everybody is entitled to his opinion, right, wrong or otherwise.

I agree that this is far from best practice for any number of reasons, but it's entirely possible that it works in some cases. Some chainrings are pretty round, so the problems of eccentricity aren't severe. In that case the flex in the rear axle and elsewhere in the system might be enough to accommodate the tight chain without bending.

As I said, I think it's poor practice, but calling it BS just fuels the fire. As it is readers of the thread have seen the spectrum of opinions, and can judge for themselves what and who makes sense.

gregjones

All of you guys on this forum are a bunch of gentlemen.

This thread managed to get locked by the mods over in SSFG!!!

FBinNY 

Give it time. Actually, I think we aren't getting worked up over this because it's too far out for serious debate.