Ring Size!!!

For discussions about bikes and equipment.
pickerd1
Posts: 36
Joined: 31 Mar 2014, 5:05pm

Ring Size!!!

Postby pickerd1 » 7 Jul 2015, 7:46am

Over time I have tried many different gear combinations for touring, and think that what I have now is pretty good for me. But I'm still not satisfied.

I started with a 12-30 10 speed cassette and a 30-39-52 triple chainset. I found that was not using the 52T and was spending most of my time riding on the middle ring and the 3rd, 4th and 5th cogs of the cassette. That is to say the ratios I ride are around 1.6:1 to 2.1:1. However I also felt that the jumps between gears where may be a bit too large.

So I made a change ...
I customised the cassette to make it 12, 13, 15, 17, 19, 21, 23, 25, 27, 30. Note: this has slightly adversely affected the shifting.
And installed a chainset with 24-36-46 rings.
This gave me usable ratios on the large ring and reduced the jumps between cogs to 2 teeth instead of 3 in the usable area of the cassette. I didn't help with the close ratios though as I changed to a 36T middle ring.
As a bonus this set-up also gave me a very low gear that I have found useful for loaded touring.

My plan now is to try the following ...
A 12-27 cassette (12, 13, 14, 15, 16, 17, 19, 21, 24, 27) with 22,32,46 chainrings.
This seems to give me all the things I am after:
- usable ratios on the large ring
- close ratios in the correct range
- a very low gear

The only thing that is concerning me is that I have reduced the size of the middle ring, which I use most, from 39T to 32T.

Is there any adverse impact of this?

Brucey
Posts: 36763
Joined: 4 Jan 2012, 6:25pm

Re: Ring Size!!!

Postby Brucey » 7 Jul 2015, 9:22am

pickerd1 wrote: ....The only thing that is concerning me is that I have reduced the size of the middle ring, which I use most, from 39T to 32T.....Is there any adverse impact of this?


yes, there will be a larger double-shift between the middle and big ring, and when you are using the middle ring, you are more likely to be using smaller sprockets (which will wear faster and be less efficient) at any given speed.

I know some folk who would reject your gearing scheme because there is no useful gear between 65" and 70" but for others this would be of no concern whatsoever.

You might find this page useful

http://www.bikecalc.com/speed_at_cadence

Really, with a 10s/triple setup getting a workable set of ratios is like shooting fish in a barrel; it is far more of a challenge if you have about half as many sprockets at the back.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

pickerd1
Posts: 36
Joined: 31 Mar 2014, 5:05pm

Re: Ring Size!!!

Postby pickerd1 » 7 Jul 2015, 9:45am

Thanks for your input.

Why would my proposed set-up be 'less efficient'? Is this down to ring and cog size being small for a given ratio?

hamster
Posts: 3296
Joined: 2 Feb 2007, 12:42pm

Re: Ring Size!!!

Postby hamster » 7 Jul 2015, 10:17am

In general larger cogs are more efficient as the chain tension is lower. However we are talking about the odd 0.1% or so, and chain lubrication is a far bigger factor. Also, this is dwarfed on a tourer by the aerodynamic disadvantages of panniers. :wink:

pwa
Posts: 11254
Joined: 2 Oct 2011, 8:55pm

Re: Ring Size!!!

Postby pwa » 7 Jul 2015, 10:30am

It's just my feeling (no science involved) but I think the inefficiency of smaller cogs is more significant at the back, where an 11t sprocket feels to me, on a workstand, very rough and rumbly in operation, whereas a 13t feels a lot smoother. I've never had a chainring small enough to feel rumbly.

pickerd1
Posts: 36
Joined: 31 Mar 2014, 5:05pm

Re: Ring Size!!!

Postby pickerd1 » 7 Jul 2015, 11:04am

hamster / pwa - thanks for your replies.

Ok, I can sort of see what your saying - a bend round a lower diameter, being sharper, doesn't spread the pressure / force as evenly as round a large diameter. SORT OF?

I expect this is offset by the reduction is weight.

I can also see that, unless the diameters are very low, aerodynamics is more significant.

As I expect to be riding mainly in the 15T to 19T range so hopefully this won't become too much of a problem.

reohn2
Posts: 37461
Joined: 26 Jun 2009, 8:21pm

Re: Ring Size!!!

Postby reohn2 » 7 Jul 2015, 11:39am

IMHO what's needed is a progression of cogs on the cassette that suit the rider,whether you prefer close or wide ratios is an individuals choice.
With a 10sp cassette that shouldn't be a problem
The next thing is to find chainring sizes that provide high enough and low enough sets of the chosen cassette's ratios.
To be ideal,the most used cruising ratios should have a straight chainline for longer chain life.
If that can be achieved with bigger cogs the whole drivetrain will last even longer.
The cruising range should be in the biggest/outer chainring and middle cogs of the cassette equalling most chain wrap straightest chainline,which in turn will provide a smoother longer lasting and most efficient drivetrain.
-----------------------------------------------------------
I cycle therefore I am.

User avatar
al_yrpal
Posts: 8185
Joined: 25 Jul 2007, 9:47pm
Location: Cully
Contact:

Re: Ring Size!!!

Postby al_yrpal » 7 Jul 2015, 12:35pm

Ratios are very personal IMO. You are obviously searching for what suits you and you will no doubt find it.

On my Salsa Vaya I have a strange drivetrain, a 50 39 24 triple chainring (the 50 and 39 are original, the 24 was added by me) and a 12 to 32 cassette. This means I can zoom along mostly on the 50 , and grind up the steepest hills by selecting 24 32 on a 21" lowest gear. I ride my bike unloaded mostly, and tour loaded 2 or 3 times a year. Unloaded I rarely use the 24 tooth chainring.

Its all a very personal choice.

Al
Touring on a bicycle is a great way to explore and appreciate the countryside and towns you pass through. What do you do to make a difference?

Brucey
Posts: 36763
Joined: 4 Jan 2012, 6:25pm

Re: Ring Size!!!

Postby Brucey » 7 Jul 2015, 12:49pm

pickerd1 wrote: Ok, I can sort of see what your saying - a bend round a lower diameter, being sharper, doesn't spread the pressure / force as evenly as round a large diameter. SORT OF?


the losses are called chordal losses and they are not insignificant per se, just much smaller than aero and rolling resistance losses of the bike. I might accept these extra losses and the faster wear that results but only if it confers a palable advantage of some kind. I'm not sure your gearing scheme does that, unless you think you need lost of high gears on your touring bike.

I expect this is offset by the reduction is weight.


Not really. With your setup the chain is the same length within one link or so and two chainrings (like for like) might be 10-20g lighter, each. There is a weight saving in the cassette but this likely won't make up for the losses incurred by (say) using 2T smaller sprockets everywhere.

As I expect to be riding mainly in the 15T to 19T range so hopefully this won't become too much of a problem.


if you pedal at 90 rpm mostly using 15-19T sprockets then that means you will be doing 12-15mph on the middle ring and 17-24mph on the big ring, with lots of double shifts to look forward to if you routinely transition through that speed (which I do).

It is as much a matter of personal taste as anything; what suits some folk might drive others nuts...

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

pickerd1
Posts: 36
Joined: 31 Mar 2014, 5:05pm

Re: Ring Size!!!

Postby pickerd1 » 7 Jul 2015, 1:03pm

Thanks everyone - great info.

I like the approach set out by reohn2 and the detailed technical stuff from Brucey.

In the end as you all say it's a case of personal preference with a few provisos - mainly chainline in common gears to reduce wear and improve efficiency.

Ideally we would use very large lightweight rings and sprockets that all had a direct / straight chainline.

Brucey
Posts: 36763
Joined: 4 Jan 2012, 6:25pm

Re: Ring Size!!!

Postby Brucey » 7 Jul 2015, 1:42pm

I just ran some sums in a bike calculator and for example

if you can improve transmission efficiency by 0.1%;
- you are still faster on the flat if you carry an extra 200g
- you are the same speed on a 1% grade even if you carry an extra 100g.
- you are the same speed on a 10% grade if you carry slightly over 90g extra weight.

If you can improve transmission efficiency by 1%
- you are faster on the flat even if you carry an extra 2kg
- you are faster on a 1% gradient even if you carry an extra 1200g
- you are still faster with almost 1kg extra weight on a 10% gradient

So a more efficient transmission is 'worth paying for', even with a fair bit of extra weight on board.

IIRC one set of transmission tests indicated that the middle chainring on a modern shimano triple chainset was markedly less efficient than it should have been (for its size) vs the large chainring and the small chainring. Possibly this is because the tooth shapes are very far away from what they should be these days, on a middle ring, with a view to better shifting. So bad chainrings, small sprockets, bad chainline... they could easily add up to 1%, which in round numbers has about the same effect as adding a kg onto the bike in quite hilly country; more than you might expect.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

User avatar
simonineaston
Posts: 2309
Joined: 9 May 2007, 1:06pm
Location: Bristol - work in... Yate!

Re: Ring Size!!!

Postby simonineaston » 7 Jul 2015, 2:06pm

Brucey wrote:IIRC one set of transmission tests indicated that the middle chainring on a modern shimano triple chainset was markedly less efficient than it should have been (for its size) vs the large chainring and the small chainring. Possibly this is because the tooth shapes are very far away from what they should be these days, on a middle ring, with a view to better shifting. So bad chainrings, small sprockets, bad chainline... they could easily add up to 1%, which in round numbers has about the same effect as adding a kg onto the bike in quite hilly country; more than you might expect.
Now that is a surprise... Glad I've fitted a 'twinset' for my next tour! :D
ttfn, Simon in Easton
(currently enjoying a Moulton TSR & a nano Brompton...)

reohn2
Posts: 37461
Joined: 26 Jun 2009, 8:21pm

Re: Ring Size!!!

Postby reohn2 » 7 Jul 2015, 5:44pm

simonineaston wrote:
Brucey wrote:IIRC one set of transmission tests indicated that the middle chainring on a modern shimano triple chainset was markedly less efficient than it should have been (for its size) vs the large chainring and the small chainring. Possibly this is because the tooth shapes are very far away from what they should be these days, on a middle ring, with a view to better shifting. So bad chainrings, small sprockets, bad chainline... they could easily add up to 1%, which in round numbers has about the same effect as adding a kg onto the bike in quite hilly country; more than you might expect.
Now that is a surprise... Glad I've fitted a 'twinset' for my next tour! :D


Sounds like 'pearls' of wisdom.
My middle rings are all either TA,Stronglight or Spa and are all 7075 alu,so are the outers and inners FTM.Tooth profile is the same as on the inner and outer rings,obviously the middle and outers have ramps and pins,but I can't see that making any difference to the efficiency :? .
-----------------------------------------------------------
I cycle therefore I am.

Brucey
Posts: 36763
Joined: 4 Jan 2012, 6:25pm

Re: Ring Size!!!

Postby Brucey » 7 Jul 2015, 6:16pm

I don't think that ramps and pins ought to make any difference but if they come with distortions in the tooth form (which I think is likely with shimano ones) and/or very truncated teeth (ditto), then I can see how losses can creep into the system.

My logic is thus; under normal circumstances the chain bushings only move a small amount when they are under load. Mostly this is when the link attached to the loaded roller articulates through a small angle (360/n degrees, where n is the tooth count) as the next roller comes into engagement. It is as well to note that, on an ideal tooth form, the roller is pretty much static on the tooth face for all the time it is loaded, even though the link is rotating within it. Even so a good deal of the normal friction in the chain is caused by this movement.

Now, if the tooth form is less than perfect (through manufacture or wear) then the roller is more likely to move on the tooth form when it is loaded. This could significantly increase the losses in the chain.

The good news is that after a few hundred miles of real world use, I think a poor chainring ought to settle down some.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~