Author Topic: Re: Ambiguous semantics and chain tension -- what minimum range?  (Read 9784 times)

Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Hi All!

I've been running a 40x17 combo on my Nomad Mk2, and with Rohloff's relaxed guidelines, I changed that to a smaller 36x17 today.

The bike is newish and the chain has never been adjusted. I swapped in the new chainring -- a 36t in place of the 40t. This made the chain slack, as expected. So, I removed two links -- as many as possible -- and checked the chain tension. It was way out of spec -- too loose at 70mm.

So, I turned to the manual; the guidelines are written so they can be read two ways.

On the one hand, it sounds like the chain needs 40-60mm of slack. Any less, you add some. Any more, you take it up. Simple.  

On the other hand, the first sentence makes it sound as if the chain should have something *less* than 40mm of slack...and then you readjust it when it reaches 40 but not let it go past 60mm. The confusing part is the first sentence on pg.9 of the manual (http://www.thorncycles.co.uk/thornpdf/ThornBikeOwnerManual2Web.pdf ), which reads...
Quote
Once the bottom bracket screws have been tightened, do not re-adjust chain tension until T1-T2 reaches 40mm (T1-T2=40mm).

A careful look at every single post on chain tension in the Forum archives has failed to clear this up, so my question:

What is the minimum desired chain tension (T1-T2) in mm?

Thanks in advance for your kind efforts.

All the best,

Dan. (...who may be taking things too literally)
« Last Edit: January 07, 2013, 08:55:40 AM by Danneaux »

julk

  • Hero Member
  • *****
  • Posts: 976
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #1 on: January 07, 2013, 01:05:33 AM »
Dan,
My interpretation is the 'getting slack measurement' is 40-60mm, after which chain tightening is recommended.
although I have run slacker than this until the chain jumped off over a bump.

No slack is too tight for the good of the hub, but only 10mm is enough to avoid tight spots and wear on the hub bearings.
Julian.

Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #2 on: January 07, 2013, 01:14:20 AM »
Ah! Julian, bless you!

So you are saying that starting essentially "new" as I am, a good minimum would be around 10mm?

Then start monitoring the chain for stretch at 40mm of slack, but don't go past 60?

Best,

Dan.

JimK

  • Hero Member
  • *****
  • Posts: 1652
    • Interdependent Science
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #3 on: January 07, 2013, 01:59:58 AM »
starting essentially "new" as I am, a good minimum would be around 10mm?

Then start monitoring the chain for stretch at 40mm of slack, but don't go past 60?


I have two criteria for how much to tighten. 1) I try to hit an existing dent in the BB shell, or to create a new dent that is not too close to any existing dent; 2) I don't want the chain so tight that I can feel any kind of rumbling feel when I turn the cranks. I try to get the chain as tight as I can without violating those criteria.

Yeah, I think that once the chain gets past 40 mm of slack you can tighten it. Once you are past 60 then you probably won't have to measure any more because the chain will just fall off. I know because I have done that!

I have never measured the minimum slack to make that rumbling go away. I would guess that you want a bit more slack than 10 mm. Somehow there is a trigonometric tangent or some such - the tension starts to go really high as the slack goes to zero.

pdamm

  • Jr. Member
  • **
  • Posts: 94
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #4 on: January 07, 2013, 02:40:11 AM »
Dan

I second Julians and Jim's advice.  I would be impressed if you were able to adjust the EBB so you had only 10mm of slack.  I have never been able to get much less than about 30mm slack when I adjust it.

Peter
 

Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #5 on: January 07, 2013, 03:06:08 AM »
Thanks guys, very much; this is immensely helpful. On my tandem timing rings, I've had good luck with 0.5in of slack...12.7mm, adjusting back to that when slack grows to .75in/19.05mm. I aways used a straight edge across the tops of the 'rings and added the amount I could pull the chain above and push it below to get my total, essentially the same method as Thorn's. This has worked well once I "normalize" the tight spots across the chainrings, per Sheldon Brown's suggestion here: http://sheldonbrown.com/synchain.html

Sheldon suggested setting the eccentric so the chain pulls taut at the tightest part of the cranks' rotation, retorquing the chainring stack bolts to allow the 'rings to center, then rotating to each high spot and lightly striking the chain with a hammer to relieve the tension before working 'round to the next spot, finally tightening the bolts when done. Sheldon then added...
Quote
This technique is also good for high-end BMX bikes and fixed gears, as long as the crankset has replaceable chainrings.

Once you have the chainrings centered and secured, adjust the eccentric to make the chain as nearly tight as possible without binding. Notice how freely the cranks turn when the chain is too loose. That is how freely it should turn when you are done, but with as little chain droop as possible.
This sounds too tight for a Rohloff-equipped Thorn, from all I have read.

Since this was my first loosening of the EBB (and not for wear reasons), I decided to use a mirror to aid in photographing the eccentric dents so I would have a record of how they looked when Thorn/SJSC set them. I also photographed the "point" on the end of the setscrews. These are less conical than I'd imagined from descriptions, and differ from the grub screws on my tandem eccentric (which create a ring-like indentation, rather than a pit).

Do you think it is worth striving for around 1cm minimum slack if possible? This would seem to allow for tight spots while extending as much as possible the time between adjustments, useful when starting a long tour with a newish chain.

Best,

Dan. (...chain-slacker, but by how little?)

pdamm

  • Jr. Member
  • **
  • Posts: 94
Re: Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #6 on: January 07, 2013, 03:38:06 AM »
Dan

I would be happy as long as there were no tight spots.  I would rotate the pedals several rotations to make sure you have checked for the worst tight spot combination for both the rear sprocket and front ring.  As I understand it, it is the tight spots that can stress the hub's bearings and do damage.  If you have no tight spots then there will be no problems.

Thanks for the pointer to Sheldon's article on centering the chain rings.  I might try it some time.

Peter
 

martinf

  • Hero Member
  • *****
  • Posts: 1167
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #7 on: January 07, 2013, 06:00:14 AM »
Not yet much experience with the Rohloff, but many years and lots of kms with Sturmey hub gears. With the latter, I have once or twice set the chain too tight - I could feel extra friction when pedalling, it probably wasn't doing the hub gear and chain/sprocket/chainring any good. Too loose is when the chain falls off, but in my experience that is a LOT of slack.

On my new Raven Tour as delivered by SJS, when turning the cranks by hand before fitting the Chainglider, I decided that the chain felt a little too tight on the tight spots, so I loosened it a bit. I'd rather err towards "too loose" than have it too tight.

Don't yet know what a loose chain will do inside a Chainglider - I imagine it will make clattering noises, but suspect that the Chainglider will prevent it from falling off.

Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #8 on: January 07, 2013, 09:24:26 AM »
I see Koga call for about 1cm of slack in their Rohloff chains:
http://www.koga-signature.com/en/Rohloff-Alfine-chain-tensioning
After adjustment, they say...
Quote
The tension is fine, if you can push the chain up by about a centimetre.
Of course, this is with sliding dropouts, not an eccentric, but the minimum tension is beginning to sound about right. Given the chain in the demonstration is being pushed from below and gravity must cause a catenary curve, I'm guessing the "about a centimetre" is really more like 12.7mm or a half-inch, much like my standard for the tandem timing chain.

Best,

Dan.

Andre Jute

  • Hero Member
  • *****
  • Posts: 4128
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #9 on: January 07, 2013, 09:47:13 AM »
Dan, like Koga I have sliders rather than an eccentric bottom bracket to adjust chain tension.

However, I would consider your previous standard of 0.6in/12.7mm a minimum. I adjust my chain until I eyeball it moving about a centimetre or half an inch on both halves when squeezed between thumb and forefinger. This avoids any danger of tight spots, misalligned chainrings, etc, etc.

Simply as a matter of interest, the way I read the instructions, both in German in the printed book, and in the perfectly good English translation downloadable from the net, 10mm is an absolute minimum of slack to avoid problems. Furthermore, 40-60mm total slack, an inch each side plus an "eyeball allowance", say, is acceptable.

Herr Rohloff, in my interpretation, clearly does not want you to run with a chain that is tight at any part of its rotation. More, it is quite clear to me that he would rather you run with a slack chain than one even approaching tightness. On a piece of equipment as expensive as a Rohloff hub gearbox, it is worth paying attention to the maker's intention when it can be this clearly perceived.

This entire chain tightness discussion is, in my opinion, wrongly directed, a sure sign of the roadie antecedents and obsessions of the tourers on this board. Their rellex question is "How tight can I make the chain without doing damage?" Why? There is no perceptible advantage to a tight chain. A better question would be, "How slack should the chain be as a minimum to avoid any possibility of damage by over tightness?" And the corollary is, "How slack can the chain be before it starts falling off?" which I take up in my answer to Martin.

Sorry to be out of step.

Andre Jute

Andre Jute

  • Hero Member
  • *****
  • Posts: 4128
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #10 on: January 07, 2013, 10:14:00 AM »
Don't yet know what a loose chain will do inside a Chainglider - I imagine it will make clattering noises, but suspect that the Chainglider will prevent it from falling off.

At one point I had the chain skipping inside a Chainglider. This was shortly after I fitted a new crankset, chain wheel, and chain. On inspection the chain was clearly slack by a total of about three inches; I didn't need to measure, I could eyeball such grotesque slackness. It was probably caused by a combination of a miscount of chain links and setting a bit of extra slackness on the sliders.

I suspect that if the chain ran open or inside a big Dutch style chain case, instead of skipping it would have come off the cogs. In the Chainglider it skips.

Inside the Chainglider I run my chain with my normal 20-25mm of slack and it causes no audible drag inside the Chainglider or perceptible drag at the pedals. When you can't see your chain -- and let me stress for people who clean their bike with a toothbrush, the idea of the Chainglider is that you don't have to see your chain more than once or most twice a year, or even every few years if you're light on chains -- I reckon slackness in the middle of the desirable range is just right. I work this out as a maximum of 40-60mm before adjustment, so halfway is max of 50mm, and half of that is 25mm, which is easy to eyeball, and enough distance from the absolute minimum not to be dangerous to an expensive piece of equipment.

One of the important points of riding on a piece of German agricultural engineering like the Rohloff is low maintenance (you grease the EXT gearbox twice a year, you change the oil every year, done). To my mind, it is contrary to the spirit of thing to set it up with a chain so close to damaging tightness that it needs to be inspected and adjusted more than once a year. The Chainglider, another low maintenance tool, just makes it more necessary to give the chain a bit of slack, in both senses of the word!

Andre Jute

Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #11 on: January 07, 2013, 10:27:06 AM »
Quote
Sorry to be out of step.
Not at all, Andre, not at all. I value your opinion and know you'll give it plainly, without holding back; that's why I asked so I can weigh it as a data point in terms of my own experience. The guidelines were not clear as I read them, so I figured if there are maximums (in terms of slack), there must also be minimums (also in terms of slack). Couldn't find anything on the desired, non-damaging minimums, so it was worth tapping the collective hive-mind for opinions.

Your 20-25mm sounds reasonable to me. I know I can't afford to risk damage to a piece of equipment I am pretty much staking my life on in the extremely remote places I tour alone and unassisted, outside cell-phone and rescue coverage.

All the best,

Dan.


Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #12 on: January 07, 2013, 11:34:34 AM »
Ah! The Word from Rohloff, direct:
http://www.rohloff.de/uploads/media/Montage_en.pdf
Page 1:
Quote
Eccentric BB or adjustable dropouts on frames with the Rohloff Speedhub:...The chain tension should be set so that the chain has approximately 10mm of play when lightly pushed from underneath and under no pressure.
The accompanying diagram shows the chain deflecting at the midpoint and includes the words, "Max. 10mm".

So, while it does no harm to have it run considerably slacker, it appears this is the Rohloff's Official Word on how tight it can go without causing harm...at least measured from below at the midspan as they mention. It is late/early here at 3:39AM and I need to go to sleep, but when I get up for good at 7AM, I'll check to see how this compares to Thorn's T1-T2 method and will update with results.

Thanks very much for your thoughts, everyone. I'd still like to hear what you consider the "bare-safe minimum".

Best,

Dan.
« Last Edit: January 07, 2013, 11:41:45 AM by Danneaux »

Dave Whittle Thorn Workshop

  • Hero Member
  • *****
  • Posts: 582
    • Thorn Cycles
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #13 on: January 07, 2013, 03:24:06 PM »
In short, our experience is that about 10mm at the tightest point in the chain run is fine for a new chain, 20mm for a used chain.

As has been correctly stated too loose just means the chain may unship, too tight can damage your hub, its far better to run a chain too slack than too tight.

Its best to do large adjustments to chain tension than small ones to avoid all the dents joining up into a row making it difficult to adjust and secure the eccentric.

The entire user manual is due a re-work/update which we will hopefully get round too later this year.


Danneaux

  • Administrator
  • Hero Member
  • *****
  • Posts: 8281
  • reisen statt rasen
Re: Ambiguous semantics and chain tension -- what minimum range?
« Reply #14 on: January 07, 2013, 05:24:00 PM »
Quote
10mm at the tightest point in the chain run is fine for a new chain, 20mm for a used chain.
Thanks, Dave! Very helpful to know the limits before "too tight" as well as "too loose".

Best,

Dan.