Touring Gear Ratios for Wimps!

[BigG]

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. Incidentally, although I continue to be a fan of half-step systems (all doubles were half step when I started using them in the mid '50s) I found a one-and-a-half step arrangement wholly unacceptable. I tried one many years ago to try to extend the gear range by just changing a 44 tooth inner ring for a 36 tooth ring. The change sequence was too awkward and I abandoned it after only a few months.

[foxyrider]

The more I read on this thread the more I feel like Hee Man!

My Airnimal runs a 10x3 Campag set up with a gear range of 103.9 to 19.7 inches. The small 26 front ring can only be used with about four sprockets as the chainline causes fouling on the 52 outer or 39 middle although it is just about possible to use the 52x29 bottom. The Xenon triple would take a 24 ring but I didn't think it worth doing.

OTOH my big wheel tourer runs an 11x2 with a 50/34 up front and essentially the same 12-29 range at the back. I can select and use all 22 gears if need be.

Both bikes run long cage Campag mechs although, in theory at least, the double set up would work with a standard cage mech.

My biggest (ho ho) issue with the super low gears is actually not making them work but actually keeping the bike upright! The very few times i've resorted to the lowest gear on the Airnimal, balance at such low speeds has been the problem, I'm talking 25% grades fully loaded. Even loaded I can crawl up just about anything with the 34x29 combo and still retain the mid-hi range gears I use for the other 99% of my riding.

[CJ]

CJ/Brucey - a question from a novice, after playing around with the Gear Calculator:

For a nominal 27in wheel, why not have a triple with closely spaced outer and middle rings - say 48T and 44T and and a 30T or 28T inner, coupled with a 9 speed 12-36 cassette?

My first triple, which I made in 1972 by machining longer bolts in after-school metalwork club for a Nicklin double chainset, on which I fitted 'C34' rings of 32,44,52 as I recall, shifted outer-middle-outer just fine, but middle-inner-middle not so good. Being of an analytical bent, I realised that the middle and inner of a triple are to all intents and purposes a double, but with the front mech WAAAAY too far above them! So I 'invented' (simultaneously with Frank Berto and after countless others I daresay) the half-step triple. By making the middle ring only a few teeth smaller than the outer, this solves the lousy middle-inner shift (since the middle is now much closer to the mech) whilst also providing in-between gears on the rather gappy wide-range 5 and 6-speed freewheels that were the best you could get for touring in those days (and shifted very well in spite of going like 14,17,21,26,32, thanks to the slant parallelogram just invented by SunTour).

That arrangement stood me in good stead for three decades: great shifting, great range and finer tuning when I needed it in the middle to top of that range. I still run a couple of half-step triples: 24,42,46 on 13-34 7-speed 'K' cassette on my old (1973) King of Mercia tourer and 24,43,47 on 11-34 8-speed (see above) on our 1978 Mercian tandem.

Meanwhile Mr Shimano did all sorts of awful things to make the un-thinking man's "add-ten-twice" triple work a bit better. First he deepened the inner cage of the front mech, so it did actually push the chain against the teeth of the middle ring. That greatly improved the dodgy inner-middle shift, but only if the middle was just the right number of teeth smaller than the outer - which incidentally made half-step triples a thing of the past. Then he knocked out some of the middle ring's teeth and shortened the rest, so the chain was wanting to fall off to the left already and needed only the lightest of touches to finish the job. This half-worn-out-already middle is the most-used ring on any touring bike, but heigh-ho, at least it shifts now.

It is still possible to work a half-step triple with a road double mech, but the cages aren't deep enough to fit as small an inner as you're likely to want and besides: when you get to 9-speed, even with 11-34 or 12-36, the jumps between the sprockets aren't really big enough to justify the half-step. So frankly, unless I were restoring an old bike with 7-speeds or fewer, even I, pioneer and decades long exponent of this sytem that I am, would not bother.

As I've said before: with 9 in back you don't really need three in front. All you need is a much bigger difference than Mr Shimano thinks is good for you (but works perfectly well) between outer and inner rings! So the bike I actually tour on now has a 12-36 9-speed cassette with rings of 22,42 on Middleburn cranks equipped with their 'Incy spider'. Whilst the Audax bike has 11-36x10 with rings of 26,42 (plus a guard-ring in place of the outer) on a noughties vintage 105 triple crank. Helen's new bike meanwhile also has 11-36x10 driven by a 24,38 Deore XT double. This is future way of touring - or would be if any of the shakers and movers gave any serious thought to touring!

[CJ]

The more I read on this thread the more I feel like Hee Man!

My biggest (ho ho) issue with the super low gears is actually not making them work but actually keeping the bike upright! The very few times i've resorted to the lowest gear on the Airnimal, balance at such low speeds has been the problem, I'm talking 25% grades fully loaded.

I think you would probably find it easier to balance and steer if you were not pulling so hard on the handlebars, as you must be in order to counter the force you're applying to the pedals. So if you could possibly bear the shame of a 'dinner plate' sprocket and a 'granny' ring (and other equipment made by some firm unhampered a road-racing superiority complex), you might balance somewhat better at those low uphill speeds. I know I do. But also there is an element of practice makes perfect. You've got to WANT to ride slowly uphill in order to get really good at it.

Even loaded I can crawl up just about anything with the 34x29 combo and still retain the mid-hi range gears I use for the other 99% of my riding.

But "just about anything" literally isn't anything!

Ho ho

[foxyrider]

The more I read on this thread the more I feel like Hee Man!

My biggest (ho ho) issue with the super low gears is actually not making them work but actually keeping the bike upright! The very few times i've resorted to the lowest gear on the Airnimal, balance at such low speeds has been the problem, I'm talking 25% grades fully loaded.

I think you would probably find it easier to balance and steer if you were not pulling so hard on the handlebars, as you must be in order to counter the force you're applying to the pedals. So if you could possibly bear the shame of a 'dinner plate' sprocket and a 'granny' ring (and other equipment made by some firm unhampered a road-racing superiority complex), you might balance somewhat better at those low uphill speeds. I know I do. But also there is an element of practice makes perfect. You've got to WANT to ride slowly uphill in order to get really good at it.

Even loaded I can crawl up just about anything with the 34x29 combo and still retain the mid-hi range gears I use for the other 99% of my riding.

But "just about anything" literally isn't anything!

Ho ho

I'm quite prepared to walk if I have to! And what's this pulling on the bars thing?

Maybe with my advancing years I need to gear down, oh hang on what tosh am I thinking! I'll let you know when i'm back to fitness whether I need something lower than a 29 sprocket. Whether I can 'clean' a climb doesn't bother me but it's a point of personal honour to ride the full length - there's too much of me to climb well, always has been so I perhaps rely on brute strength more than is ideal.

[pwa]

When I was in my twenties I crossed the Col de la Bonnette on 39x28 with panniers and camping gear. It was a struggle, involving stops, but I did it. A few years later I discovered chainsets that allow smaller rings and since then I've aimed to have a smallest of 26. And that means I can haul a four pannier bike up a long ascent without feeling that I am on the limit. On a proper tourer (one that declares itself to be that on the label) I can spin those pedals without a wobble at less than walking speed if I really want to. In practice those really low speeds are confined to those super steep ramps you get on some hairpin bends.

I wonder if the problem some have with triples is down to using STI levers. I don't, and for me a triple is as easy to use as a double, but with more choice. All plus and no negative.

[De Sisti]

I wonder if the problem some have with triples is down to using STI levers. I don't, and for me a triple is as easy to use as a double, but with more choice. All plus and no negative.

I've never had a problem with triples. Setting up the gears is just as easy as on a double (in my opinion).

[Brucey]

....I wonder if the problem some have with triples is down to using STI levers. I don't, and for me a triple is as easy to use as a double, but with more choice. All plus and no negative.

I basically agree. Lack of readily available bits has forced me into using an 'alpine double' arrangement before now, and it has fringe benefits of apparent simplicity and low weight. But I've arguably yet to see an arrangement of this sort that wouldn't be improved by addition of another chainring, usually a middle one. With this setup the argument that the middle ring becomes the most-used chainring is difficult to maintain; if this becomes the case, surely the big ring in the double setup was the wrong size...?

If you add a third chainring the main difference is that the triple system has

- about +100g weight (which is chicken feed on a touring bike, surely?)
- more useful gear ratios, on the middle ring, typically suitable for gradual inclines
- a double-shift of very greatly reduced size

In addition there is likely to be

- a small change in Q (which may be non-existent with some chainsets as per Valbrona's earlier comment)
- gear ratios that give you the choice of 'filling in' the gaps in a wide-range cassette
- some care required in selection of a suitable FD, specifically to get the inner-middle shifts to work OK
- better chainlines in the most-used gears.

Taking CJ's preferred (2x9) 22,42/12-36 setup as a starting point

http://www.gear-calculator.com/?GR=DERS&KB=22,42&RZ=12,14,16,18,21,24,28,32,36&UF=2185&TF=90&SL=2.1&UN=MPH

(NB if you click the ' Display....gear inches' button you will see more, useful, info if you are used to gear inches)

you can see that there is a good range, but there is a huge double-shift, some large gaps in the cruising gear range, and the gears require a chain angle of ~2.1 degrees to allow them all to be used.

However this page

http://www.gear-calculator.com/?GR=DERS&KB=22,42&RZ=12,14,16,18,21,24,28,32,36&UF=2185&TF=90&SL=1.7&UN=MPH&GR2=DERS&KB2=22,34,42&RZ2=12,14,16,18,21,24,28,32,36&UF2=2185

compares the previous setup with another which has (very simply) a third chainring added, of 34T. You can see that the double-shift issue is greatly eased, there is more choice of gear ratios for slight gradients, and the full gear range can now be accessed when using a chain angle of ~1.7 degrees on the triple.

This page

http://www.gear-calculator.com/?GR=DERS&KB=24,46&RZ=13,15,17,19,21,24,28,32,36&UF=2185&TF=90&SL=1.8&UN=MPH&GR2=DERS&KB2=24,36,46&RZ2=13,15,17,19,21,24,28,32,36&UF2=2185

compares a 24,46/13-36 double with a 24,36,46/13-36 triple. The chain angle 'gate' has been set at 1.8 degrees in this case, which still demonstrates how the triple setup can be kinder to the chain and more efficient. I'd personally prefer this double setup over the previous one, since I spend a lot of time using gears of sixty-something inches, and there is more choice, because the big gaps in the cassette are reduced/in different places in the gear range. But the addition of the third chainring has again eased the doubleshift issue and provided more gear ratios for use on slight gradients. Also the use of a slightly larger chainring has increased the size of chainring and sprocket that are used when cruising, which gives a slight improvement in efficiency and a usefully improved wear life.

This page

http://www.gear-calculator.com/?GR=DERS&KB=24,46&RZ=13,15,17,19,21,24,28,32,36&UF=2185&TF=90&SL=1.2&UN=MPH&GR2=DERS&KB2=24,36,46&RZ2=13,15,17,19,21,24,28,32,36&UF2=2185

compares the same setups as above, but this time the chain angle gate has been set to 1.2 degrees. It shows that (barring the lowest and highest gears) the triple setup allows a good gear range (20" to 84") to be accessed without running appreciably cross-chained.

It is currently very fashionable to pooh-pooh the effects of running cross-chained (witness those who advocate basically ghastly 1x systems as 'the future'.... ) but there is evidence there for those that seek it; both efficiency and transmission life are likely to be compromised at greater chain angles. To put this into context, if you are slogging up a climb and working hard at ~200W, a ~1% change in efficiency is likely to be comparable to the difference between running nice easy-rolling tyres and pretty ordinary ones. Whilst 1% is maybe on the top side of estimates of efficiency change, it does suggest that if you are bothered about tyres, you perhaps ought to be bothered about the possible effects of chainline, too.

cheers

[reohn2]

As per other threads I've posted on about gearing/triples comparisons,etc.
For me the triple is king and not without reason,as PWA posts above "all positive without the negatives"
My 3x9 = 14,15,17,19,21,23,25,28,32 or the bottom three ratios can be 26,30,34 for loaded touring,coupled to a 22/32/44t or 24/34/46t, chainset or 48 or 50t outer ring to offer taller ratios for the flat stuff.
I've run all combinations of the above on tandems and solos with STI's and Kellys/DT levers with 9 and 10sp front road mechs.
Brilliant acurate and crisp shifting with a very useful spread,I cant think why I'd want to lose the middle ringas it's sooo useful.

[gloomyandy]

It is currently very fashionable to pooh-pooh the effects of running cross-chained (witness those who advocate basically ghastly 1x systems as 'the future'.... ) but there is evidence there for those that seek it; both efficiency and transmission life are likely to be compromised at greater chain angles. To put this into context, if you are slogging up a climb and working hard at ~200W, a ~1% change in efficiency is likely to be comparable to the difference between running nice easy-rolling tyres and pretty ordinary ones. Whilst 1% is maybe on the top side of estimates of efficiency change, it does suggest that if you are bothered about tyres, you perhaps ought to be bothered about the possible effects of chainline, too.

cheers

You can play around with these things all day to support various arguments, for instance if you set the limit to be 1.5 degrees then both systems have the same gear range (both losing the bottom and top gears), but the triple only has 5 "usable" gears for each chainring while the double has 6. Does that calculator accurately model how the chain will catch on the front mech or other chain rings when using some (not always that extreme gear) combinations? A modern double setup will happily let you use all of the 22 gears available (should you wish), without any trimming, not something I've ever been able to do with a triple.

As to your example above, is there really any evidence that shows that a change in 0.3 degrees of chainline (which is the difference between the two systems when operating in the lowest gear) actually results in a 1% loss of efficiency?

[reohn2]

It is currently very fashionable to pooh-pooh the effects of running cross-chained (witness those who advocate basically ghastly 1x systems as 'the future'.... ) but there is evidence there for those that seek it; both efficiency and transmission life are likely to be compromised at greater chain angles. To put this into context, if you are slogging up a climb and working hard at ~200W, a ~1% change in efficiency is likely to be comparable to the difference between running nice easy-rolling tyres and pretty ordinary ones. Whilst 1% is maybe on the top side of estimates of efficiency change, it does suggest that if you are bothered about tyres, you perhaps ought to be bothered about the possible effects of chainline, too.

cheers

You can play around with these things all day to support various arguments, for instance if you set the limit to be 1.5 degrees then both systems have the same gear range (both losing the bottom and top gears), but the triple only has 5 "usable" gears for each chainring while the double has 6. Does that calculator accurately model how the chain will catch on the front mech or other chain rings when using some (not always that extreme gear) combinations? A modern double setup will happily let you use all of the 22 gears available (should you wish), without any trimming, not something I've ever been able to do with a triple.

As to your example above, is there really any evidence that shows that a change in 0.3 degrees of chainline (which is the difference between the two systems when operating in the lowest gear) actually results in a 1% loss of efficiency?

On my 3x8 and 9sp triples I can and do in outer and inner rings use 7 of the 8 or 9 rear ratios and all 8 or 9 ratios in the middle ring.

[Brucey]

As to your example above, is there really any evidence that shows that a change in 0.3 degrees of chainline (which is the difference between the two systems when operating in the lowest gear) actually results in a 1% loss of efficiency?

If only it were just 0.3 degrees, some of the time. It is (depending on the set up and the distribution of gear usage) quite likely that the average chainline used on a double is worse than that vs a triple. The reason is that in very many cases one is forced to run cross-chained (small-small or big-big) a good portion of the time, whereas with a triple there is a similar gear ratio with a much better chainline.

For example with the double setup we looked at before, a gear of around 40" requires that you use a chainline of around two degrees. With a triple, that could be less than 0.5 degrees.

There are various measurements of changes in efficiency with chainline; as previously mentioned a 1% change is on the high side but is not out of the question.

cheers

[gloomyandy]

If only it were just 0.3 degrees, some of the time. It is (depending on the set up and the distribution of gear usage) quite likely that the average chainline used on a double is worse than that vs a triple. The reason is that in very many cases one is forced to run cross-chained (small-small or big-big) a good portion of the time, whereas with a triple there is a similar gear ratio with a much better chainline.

For example with the double setup we looked at before, a gear of around 40" requires that you use a chainline of around two degrees. With a triple, that could be less than 0.5 degrees.

There are various measurements of changes in efficiency with chainline; as previously mentioned a 1% change is on the high side but is not out of the question.

cheers

Yep as I said you can play around with those numbers in all sorts of ways. But if we are going to play that game then given that double setup you could use the 39" provided by the 24x17 rather than the 40" from the 46x32, using that combination lowers the offset to 1.5 degrees so a difference between the two systems of 1 degree.

But wait, some of the papers on this subject seem to suggest that chain offset is not as important as sprocket size (some even claim it is negligible), so perhaps in both cases it would be better to use the 46x32 combination to achieve the 40" gear, though that would require a worse chainline from the triple (over 2.5 degrees).

Good game this!

[Brucey]

Yep as I said you can play around with those numbers in all sorts of ways. But if we are going to play that game then given that double setup you could use the 39" provided by the 24x17 rather than the 40" from the 46x32, using that combination lowers the offset to 1.5 degrees so a difference between the two systems of 1 degree.

But wait, some of the papers on this subject seem to suggest that chain offset is not as important as sprocket size (some even claim it is negligible), so perhaps in both cases it would be better to use the 46x32 combination to achieve the 40" gear, though that would require a worse chainline from the triple (over 2.5 degrees).

Good game this!

hours of innocent fun for all!

But the middle ring gives you a ~40" gear that is less than 0.6 degrees chainline and the 24/17 gear gives you a gear that is ~1.9 degrees chainline.

You are correct in that there are various different results from practical tests but IIRC

- in one set of tests the results were skewed by the use of a brand new transmission with a heavily manipulated middle ring (which is probably why all the middle ring results were anomalously worse than expected) and

- the chain used was PTFE coated, which made the effects of cross-chaining less bad (temporarily) than they would be otherwise.

My take is that when you look at the wear that occurs in chains that are heavily used cross-chained, it is strongly suggestive that there has been a lot of work done in wearing the sides of the side plates down; I doubt that this is without penalty.

I'd also suppose that the 46/32 gear might not be as bad as you might expect; very roughly the beneficial effects of increasing sprocket size may be about as great as the downside of worse chain angle, so downshifting on the big ring, up to a point, often produces a set of gears with about the same efficiency (with a new, well-lubricated/PTFE coated chain). By contrast upshifting on the small chainring (to ever more cross-chained gears) there is a double whammy (smaller sprocket and worse chainline) so the efficiency gets worse at a fairly rapid rate. I'd expect the middle ring 36/24 to be much better than 24/17 and as good (or better given average levels of chain lubrication) than the 46/32.

cheers

[gloomyandy]

But the middle ring gives you a ~40" gear that is less than 0.6 degrees chainline and the 24/17 gear gives you a gear that is ~1.9 degrees chainline.

I'd expect the middle ring 36/24 to be much better than 24/17 and as good (or better given average levels of chain lubrication) than the 46/32.

cheers

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?

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.