Touring Gear Ratios for Wimps!

[Brucey]

Well using your favoured app 24x17 is still a valid gear (for the double case) with the setting at < 1.5 degrees so I can only assume it is less than 1.9 degrees?

sorry yes you are quite right, my mistake

You might expect that 36x24 would be a better combination than 46x32 based on the chainline, but this may not be the case when you actually test it. For instance in:
http://www.ihpva.org/HParchive/PDF/hp52-2001.pdf
using the 44x34 35" gear (which in this case is big/big) is 92.1% efficient but using the (closest equivalent in this test) 32x26 33.5" gear it is only 90% efficient despite having a much better chainline. So by using the bigger gear combination you get a gain of 2.1% and of course in this case (big/big) a double will actually offer a better chainline than the triple.

Triples may be a good solution for some people, but I'm not convinced that the chainline efficiency argument is really that compelling.

IIRC that data showed low efficiency for all tests including the middle chainring which was anomalous behaviour. I would expect that was because it was a heavily manipulated chainring (i.e. with lots of funny-shaped teeth) which hadn't been run in yet. You can draw some good conclusions from that data (better than they did, in fact) but comparisons of results from individual chainrings are not reliable indicators in this case.

This image


shows that if you run small chainring to the third sprocket in you are running a rather lossy gear by comparison with a larger chainring and sprocket, more favourably aligned. The losses due to variations in chainring/sprocket size are comparable with the losses that are caused by cross-chaining. As you downshift on the big ring, the cross-chaining losses start to overtake the benefits of larger sprockets. In the big/big combination you are losing over 2W in cross-chaining losses alone.

cheers

[gloomyandy]

IIRC that data showed low efficiency for all tests including the middle chainring which was anomalous behaviour. I would expect that was because it was a heavily manipulated chainring (i.e. with lots of funny-shaped teeth) which hadn't been run in yet. You can draw some good conclusions from that data (better than they did, in fact) but comparisons of results from individual chainrings are not reliable indicators in this case.

Do you have a link to a paper or anything else that supports this notion? My understanding was that they used a standard Shimano triple (without any special manipulations) for the test.

Also do you have a link for the paper containing your graph? The image does not contain any information as to what the test conditions are nor does it contain any information as to what the data points represent or how they have been obtained. Your conclusions may well be correct but I'd prefer to see the data for myself.

[Rhothgar]

Kore do that ridiculous looking Mega range rear sprocket that is a 42T but the jump on the rear would be massive.

Obviously there must be something fundamentally wrong with that if people here aren't mentioning it. It is designed to convert a 2x10 to a 1x10 setup but what would prevent it from working?

[pwa]

I wonder if there might be problems with a huge sprocket moving around in ways it ought not to. Some people (not me) worry about chainrings not being stiff enough. When sprockets get to the size of middling chainrings the same must apply. And would very large sprockets adversely affect the freehub in any way?

[Brucey]

IIRC that data showed low efficiency for all tests including the middle chainring which was anomalous behaviour. I would expect that was because it was a heavily manipulated chainring (i.e. with lots of funny-shaped teeth) which hadn't been run in yet. You can draw some good conclusions from that data (better than they did, in fact) but comparisons of results from individual chainrings are not reliable indicators in this case.

Do you have a link to a paper or anything else that supports this notion? My understanding was that they used a standard Shimano triple (without any special manipulations) for the test.

IIRC the authors commented on the anomalous behaviour of the middle ring in that paper; there is clearly 'something odd' about that chainring. There are only a few possible explanations for this and the heavily/weirdly shaped teeth, and the way they are made are right at the top of the list.

If you look at a typical shimano triple chainset then you will find (as CJ has also commented) that the teeth on the middle ring end up being unlike those on other chainrings. That such chainrings can behave oddly when new is well-known; for example the middle ring of a new shimano 8s triple chainset often will not run with a 9s chain, because of the manipulated teeth. By contrast one that isn't new is usually OK with a new 9s chain. Mass-produced chainrings (like many shimano ones) that have pressed (rather than CNC'd) teeth are (IME) often noticeably less smooth until they are run-in; you can see by the first wear marks that appear that the teeth do not always share the load well when the chainring is new, for example.

Also do you have a link for the paper containing your graph? The image does not contain any information as to what the test conditions are nor does it contain any information as to what the data points represent or how they have been obtained. Your conclusions may well be correct but I'd prefer to see the data for myself.

Well, that figure is publically available to view via their website (thus I've just hotlinked it) but you'll have to buy the reports from the friction facts website to 'see everything'. However you can read on their website how the test apparatus is claimed to work. I can tell you that the data in the graph is obtained in four ways (in order as listed in the figure);

1) using a 39T chainring on the sprockets of an 11-28 cassette, with the each sprocket set perfectly aligned in turn (i.e. no cross-chaining)
2) as above but with the sprockets offset as they would be in a 2x11 system (i.e. with cross-chaining as it would normally occur)
3) as 1) but with the 53 T chainring instead
4) as 3) but with the sprockets offset as they would be in a 2x11 system

The vertical scale is Watts lost in transmission (after other losses have been allowed for). IIRC the test conditions use constant 'full' tension which isn't quite like real pedalling, but is do-able and (they say) repeatable.

[edit; just in case it isn't obvious the horizontal axis is net gear ratio, e.g. 53/28 = 1.89 and so forth.]

This shows quite clearly the effects of poor chainlines on efficiency, in isolation from other variables.

In point of fact I'd expect the effects of cross-chaining to only get worse over time as the chainrings wear; when they are new they usually have anodising on the sides of the teeth (which has a relatively low coefficient of sliding friction) but this soon wears off and leaves you with a surface that will not slide so freely under the side loads of cross-chaining.

cheers

[Brucey]

I wonder if there might be problems with a huge sprocket moving around in ways it ought not to. Some people (not me) worry about chainrings not being stiff enough. When sprockets get to the size of middling chainrings the same must apply. And would very large sprockets adversely affect the freehub in any way?

I think you have a good point. Anytime you are running cross-chained the sprocket and the chainring will flex laterally and this won't improve the efficiency. The manner and extent of the flex (and the consequent losses) will be different in real pedalling from that achieved during constant torque tests that are commonly carried out.

In the current crop of cassettes the usual thing is that the largest few sprockets are riveted together in a cluster, presumably to get as much stiffness as possible without incurring a large weight penalty.

cheers

[djnotts]

And it's not just wimps and tourers. I need 22" max if I'm to get home on my daily ride without walking (which I'm avoiding for as long as possible for largely psychological reasons). Age and more crucially health mean that a modest hill which I rode not all that many years ago on >65" fixed now tests me to the limit.

I for once contemplated a new, nice bike, but the available gearing was of no use to me. Yes there are ways around it with drops, but I don't like them sufficiently to spend the effort, time and money. Flats and bar ends are fine for my purposes - and indeed were even when I rode medium to longish distances (but not touring). My Trek "hybrid" with 3 x 9 XT stuff has a low of <18". Does for me.

[mig]

it's funny but i'd find low gears such as these very much not for wimps. i'm just not used to turning them and would probably stall on a steep hill.

[CJ]

it's funny but i'd find low gears such as these very much not for wimps. i'm just not used to turning them and would probably stall on a steep hill.

How slow one can ride up a hill is a probably mostly a confidence trick. Someone I know very well, though just as experienced as I, cannot go slower than about 4mph without feeling they are about to lose control and putting a foot down at a moment of their own choosing, rather than risk a fall. They might not be wobbling any more than me, but are not happy with how that feels anyway. And I think the risk of a fall is minimal, but definitions of minimal are personal. Whatever: this means they really have no use for their lowest gear.

But personally, I'm happy to briefly roll to a standstill on a 1 in 4, without even unclipping. Because since coming through the learning phase (but lightly scathed) I've never again had any trouble escaping my SPDs in a hurry. It now seems to happen automatically without my even thinking about it. And besides, given such a low gear, at any position of the pedals I am confident that one push will always produce enough torque to restore what mariners call 'steerage way'.

Experience of using gears like this undoubtedly helps. I'm not at all sure I could have used such low gears properly when I was younger. I do recall one or two inadvertent wheelies when I first geared down. That never happens now even though I've now geared down much further.

[iandriver]

it's funny but i'd find low gears such as these very much not for wimps. i'm just not used to turning them and would probably stall on a steep hill.

Sounds like you need to give mountain biking a go. Some of the skills you pick up are invaluable as I found out with the Belgian cobbed hills this year. You've either tried to ride them of you haven't. Everything you thought you knew about gears and momentum blown out of the water. Why I still have gears I practically never use in unknown areas.

[mig]

maybe.

i grew up in the hills of the west pennines - uppermill, oldham, delph, saddleworth and so forth with the occasional foray to the dark side - holmfirth etc so i was always used to climbing hills. the thing is i took something like 44 x 21 to be low so just got on with that, doubtless developing techniques to accomodate the gear. i then moved on to TT riding and used chainrings in the order of 55 on solos and 61 on a tandem. i suppose i got 'worse' in that direction then. now my riding is commuting and leisure but i've never moved away from similar gears even with kit on the bike and on my back.

i have tried super low gears with rings in the mid 20s but i just can't use them effectively. my heart rate rises too quickly and i'd labour on a hill that i'd ordinarily cruise up on 39 x something.

[Trebor]

Brucey's suggestion for a 6 x 3 gives good results but a little tweaking can improve it further. I use a 14-17-21-26-32-40 freewheel (a standard 5-speed with a TA Cyclotouriste 40 tooth inner bolted onto the back) with a 52/46/24 chainset (Suntour Superbe Pro 130/74 mm bcd) and find the ratios close to ideal for me. They are evenly spaced and the relatively narrow freewheel (25 mm chainline shift across the sprockets) makes the system relatively chain and sprocket friendly.

A huge range with only 6 sprockets but it does make you question if much larger gear steps, for touring/leisure, are actually OK providing that they occur below a particular gear inch threshold?

http://www.gear-calculator.com/?GR=DERS ... 2.4&UN=KMH

The BigG configuration has gear steps of 25%,23% and 24.6% at the bottom end but all of these occur below 31.5 gear inches (or you could shift from 32T on the middle ring 39.4 gear inches to 21T on the inner, which would add another step of 25.8%). These steps are smaller than some found on hub gears: Sturmey-Archer AW 33.33%, Shimano Nexus 3 36.4% and even the Sturmey-Archer 4 speed FM, fitted to Raleigh Lenton, Super Lenton and Clubman bikes had a step at the bottom of 28.57%.

If a cyclist is already power-limited, climbing on a particular gradient but then it gets even steeper then, if it is an extended section that can't be tackled with a brief supreme effort, doesn't it make sense (from a physics basis) to have the gear step approximately corresponding to the change in local gradient? For a change in gradient from 1:5 to 1:4 this would require a step of 23.67% (round it up to 24% or 25%). So why do conventional cassettes often have a small gear step at the bottom end: 30T-34T = 13.33%, 32T-36T - 12.5%, when they normally have considerably larger steps, further up the cassette and perhaps in a more critical gearing range?

Shimano recently introduced (in February 2017) a 9 speed 11-36 trekking cassette, that goes from 30T to 36T (a 20% step) manufacturers ref. CSHG4009136. This may be a superior option than the 11-34 for many users?

https://www.rosebikes.com/article/shima ... aid:890363

[iandriver]

i have tried super low gears with rings in the mid 20s but i just can't use them effectively. my heart rate rises too quickly and i'd labour on a hill that i'd ordinarily cruise up on 39 x something.

Kind of sounds like you've got so used to the grind that you're not trusting the gear to do the work and easing off on the leg pressure. As CJ was pointing at, relaxing and letting the bike do the work is the key I think. Bit like riding with no hands on the bars. If you don't relax and let the bike track itself, it all goes wrong. Stay calm and loose in the shoulders etc. and the bike will do so much more for you in my experience.

[mig]

i have tried super low gears with rings in the mid 20s but i just can't use them effectively. my heart rate rises too quickly and i'd labour on a hill that i'd ordinarily cruise up on 39 x something.

Kind of sounds like you've got so used to the grind that you're not trusting the gear to do the work and easing off on the leg pressure. As CJ was pointing at, relaxing and letting the bike do the work is the key I think. Bit like riding with no hands on the bars. If you don't relax and let the bike track itself, it all goes wrong. Stay calm and loose in the shoulders etc. and the bike will do so much more for you in my experience.

could be but i am relaxed in 'my' gears. just habituated to them i think. maybe the onslaught of time will gear me down.

[gloomyandy]

IIRC the authors commented on the anomalous behaviour of the middle ring in that paper; there is clearly 'something odd' about that chainring. There are only a few possible explanations for this and the heavily/weirdly shaped teeth, and the way they are made are right at the top of the list.

If you look at a typical shimano triple chainset then you will find (as CJ has also commented) that the teeth on the middle ring end up being unlike those on other chainrings. That such chainrings can behave oddly when new is well-known; for example the middle ring of a new shimano 8s triple chainset often will not run with a 9s chain, because of the manipulated teeth. By contrast one that isn't new is usually OK with a new 9s chain. Mass-produced chainrings (like many shimano ones) that have pressed (rather than CNC'd) teeth are (IME) often noticeably less smooth until they are run-in; you can see by the first wear marks that appear that the teeth do not always share the load well when the chainring is new, for example.

Yes the authors do draw attention to the lack of efficiency when using the middle ring, but the reasons for this given above are your own conclusions not theirs. They do not provide any indication (either way) as to an increase or decrease in performance over time. Given that they are testing a standard triple which is similar to many that people will be using, perhaps those users might be surprised to learn that the efficiency when using the middle ring is not as high as perhaps they have assumed?

With regard to the Friction Facts tests a few points are worth noting...
1. The situation we have been talking about used a larger rear sprocket (34 teeth) rather than the maximum 28 tooth in the FF test. In general larger sprockets run more efficiently. This is likely to mean that the cross over point where friction due to chainline exceeds the gains from using larger sprockets is shifted and so the graph is likely to flatten out somewhat.
2. We have been discussing the use of a 9 speed cassette not the 11 speed one use by FF in their tests, the 11 speed cassette will be using worse chainlines for the larger sprockets.
3. They used constant tension on the chain. Other authors have found that this tends to produce a less efficient drive system than the real life drive system using an actual rear mech. My understanding is that the chain in effect becomes slack when running through the rear mech allowing a redistribution of lubricant. I can't say if this will impact cross chaining more or less, but it should probably not be ignored.
4. The 2W loss is only true when comparing an ideal chainline using the same chainring and sprocket sizes. When using a more realistic comparison of the same gear ratio using a smaller chainring (39 tooth in this case) this falls to around a watt and if you use the data point using the slightly offset chainline when using the 39 tooth chainring then it is around 0.75 Watts.

Given the above and that these tests are conducted at 250W I would say they they show a real world loss in efficiency when cross chaining that is significantly less than 1%.

Without an actual real world test that compares a double to a triple having similar range I don't think it is really possible to say which is more efficient. There may be other benefits to using a triple, but personally I'm not convinced that efficiency is one of them.