The Curious Case of the 3-Speed Hill Climb
For a good few years after I first began cycling, I was quite weak at climbing hills. For steep gradients in particular, I needed low gears, a lightweight bike. And by 'needed' I don't mean preferred; I mean that I would be walking otherwise.
Three years of living in Ireland changed that. I am not the strongest cyclist out there by far. But I've adapted to my surroundings. And my surroundings are hilly! If I'm riding long distance, I sill prefer to have (and use) low climbing gears. But when it comes to each hill on an individual basis, I no longer strictly speaking 'need' a super-low gear to scale most of the ones within commuting distance.
As I look at the road that rises ahead, no longer do I wonder anxiously, 'Will I be able to make it, on this bike?' Instead I find it interesting to notice differences in how different bikes will behave. For bikes, not unlike cyclists, can certainly have different climbing styles! Some skip up like mountain goats. Others will float like helium filled balloons. Others still will feel like dead weights, requiring effortful dragging. But most interesting of all, is that almost living-breathing feeling of accordion-like flex that I experience on certain steel upright bicycles.
They are bikes which are neither too light, nor too heavy, for their type. And they are nearly always 3-speeds, of the 'sports roadster' variety.
For casual cycling in ordinary clothing I prefer to ride them in a low gear - because I don't want to sweat, and because an easier gear just makes for a better 'smell the flowers' experience. With the flimsy looking 3-speed shifter I switch to 1st at the slightest provocation. And truth be told, it kind of feels like I need to. Whereas a roadbike might crest the bumps of gently rolling roads unnoticeably, on the upright 3-speed I feel resistance even on mild inclines.
And if I am already in 1st on this gently rising road, it hardy seems a bike like that would manage up the 14% gradient mountain lane?
Yet when the harmless looking 3-speed is truly challenged, it flexes its muscles! I could not appreciate this until I had the strength to actually make it happen enough to feel it. But as the gradient steepens and I push down on the pedals harder, I can nearly feel the bike contract under my effort, and then tense and relax in sync with my pedal-pushes, thus heaving the pair of us uphill.
The sensation is like a wringing, a squeezing-out of 'performance,' drop by drop. And of course it is but my mind's translation of the raw sensations. What actually happens - to the metal, to my body - I do not entirely know. But that it is flex I'm feeling, I am pretty certain of. I am also fairly certain that this flex helps, rather than hinders, my climbing efforts. At least on these particular bikes.
Whereas the lightweight roadbike bestows performance upon me, the heavy upright 3-speed must be made to yield it. It yields it coyly and reluctantly, sometimes begrudgingly. But it does yield it inevitably. And as we crest that hill, we are friends again, ready to smell the flowers.
Redefining the "vertical" component of the old cliche, "laterally stiff but vertically compliant". ;)
ReplyDeleteI honesty think it may be a case of 'there is more than one way to skin a cat.' I have ridden bikes that feel as if they climb well because they are stiff, and also bikes that feel as if they climb well because they are flexible. What it is we are actually feeling, and whether flex even really has anything to do with it, is another matter!
Delete" that almost living-breathing feeling of accordion-like flex..."
ReplyDeleteThat's what we would call "noodly!" Yikes.
To me noodly is a bike where you can feel the flex constantly, and the flex interferes with your pedaling efforts. I've ridden bikes like that too. What I mean here though is a bike that does not feel at all flexy until you mobilise that flex, as it were, on a steep incline - and then the flex helps rather than hinders your efforts.
DeleteI certainly know what you mean and wear as at my age I prefer Multi-geared bikes I am constantly amazed at how well a properly geared single speed does going up a hill. What they lack in gearing they make up for in light weight and super efficient drive train. I find that 3 speeds (& other upright bicycles) practically beg you to climb out of the saddle and reward you for it, assuming you are in good enough condition. - masmojo
ReplyDeleteSeems like you've experienced a beneficial kind of frame flex that Jan Heine calls "planing": https://janheine.wordpress.com/2014/11/23/what-is-planing/. He has become quite scientific about trying to sort out how it works. He finds that a heavier bike with good planing can climb better (require less power output from the rider) than a lighter bike without planing. He seems to find the best planing most often in steel randonneuring frames. He feels that planing can also improve performance on flat sections of road, giving you the sensation of almost-flying, but apparently that doesn't happen with your 3-speed.
ReplyDeleteI am well familiar with the 'planing' concept, but I doubt JH would ascribe it to the sort of bicycle I mean here : ) A standard upright 3-speed would in fact feel quite stiff most of the time.
DeleteWhat I mean is more like, if you push on the pedals hard enough, you can squeeze out some degree of flex out of these bikes, that you'd otherwise not sense. Maybe...
Reading your, and his, posts over the years I have to wonder how much the idea of planing assumes a powerful and rather heavy rider.
DeleteAt 5'3" and 125lb, no bicycle I have ever ridden has either "planed" or "flexed" for me!
At 5'6" and 135lb I do feel flex on some bikes, and I felt it even as a much weaker rider in my first years of cycling. But the thing is, when I feel flex in a bike it does not necessarily translate to 'planing.' With some bikes it feels as if the flex is in sync with my pedal strokes, in which case it helps. With other bikes the same amount of flex will somehow feel wrong, interfering with my effort. What determines this, I do not know.
DeleteIn general, I think these ideas are all quite subjective and rider-specific, and therefore difficult to prove. It's not only weight and strength that come into it, but perhaps also the rider's pedaling style, the way they sit on the bike, their body's center of gravity, and also of course rider sensitivity. Some people just do not notice; others are hyper-aware of tiny differences.
There's always a marriage with bike and body and sometimes the riding style needs to change in order to suit the bike. Different bikes flex in different ways under different circumstances. The fact that you're able to cross over to so many bikes is, well, what you are…Bless you! I've tried your suggestions and attitude before…Simply doesn't work for me, my style, my brain, whatever…My poetry is equally wonderful and is what keeps me and my bike happily together. The upright three speed was a nice date but I never was able to resume smelling the flowers after topping a hill ;)
DeleteJans article is interesting, butI was surprised to see him comparing steel bikes to Titanium bikes and the Steel being "better" because some of the best "planning" bikes I've ever ridden were titanium and they would do that even one flat surfaces and moderate pedal pressure. Now in fairness Jan was strictly looking at climbing and living in a relatively flat area I never did enough climbing to tell, but I guess it's conceivable that the Springyness I've always associated with titanium might give up some of it's advantage under hard climbing!? Of course the wild card is the riders themselves - masmojo
DeleteI think a fair number of commenters here are familiar with tube dimensions. Our blog host is. Consider the Eddy Merckx hour record bike. Built from Reynolds 531 drawn to 22/28 British Wire Gauge. That is 0.71mm at the ends and 0.376mm in the middle. Just slightly lighter than current 0.7/0.4 tubing. And we are talking about old school skinny tubes, no oversize tubes. No one these days would even consider tubes so light for someone so large as Eddy. I am about Eddy size with half the power. Twenty years ago I got a frame with those tubes by ordering across the continent and lying about my weight. When I arrived to pick up the bike they almost wouldn't give it to me. In 2017 it wouldn't happen.
DeleteBut it worked for Eddy. He had lots of other bikes with those tubes. He had heavy bikes too. He knew what he wanted for the ride that day and he knew how to make it work. If a modern rider only accustomed to ride unyielding carbon frames were to try the Eddy bike no way would he get effective power transfer. Possible he wouldn't even get the bike pointed straight under power. Current riding style does not work with light steel. But light steel can work.
Bikes are half alive. They respond. They know who is riding them. What a novice gets from a bike is not what a cyclist gets from a bike. The same bike is very different for different riders.
I wonder to what extent "planing" is a myth. I don't recall ever having seen a quantitative analysis of it, or even a good description of what planing is and does. Further, as far as I can tell, this is a quality that only Jan Heine seems to recognize. I assume I'm wrong and that planing really is a thing and that it has some value in bike performance, but I would have said the same thing about aether before Michelson and Morley.
DeleteI have to confess I haven't experienced "planing" in such a way that I can isolate the phenomenon from other factors that make a bike seem "fast" -- and I certainly do find some bikes "fast" and others "meh" and others still "dogs".
DeleteBut on the iBob list (started, and still named, as the "Bridgestone Owners Bunch" list) there are very many very experienced riders who swear they experience it, and have isolated what makes it work for them. As a very longtime member of this list, I know that there are too many very experienced and intelligent cyclists who claim to feel it, to be able in good conscience to deny that it exists.
Only, I think "planing" as described by Jan is relative not only to tubing, but to weight, pedaling style, position, gearing choices, and many other things.
Is this what Jan Heine calls "planing," perhaps?
ReplyDeleteI wonder if, when you stand and put out a great deal more torque than usual, something the "planing" phenomenon doesn't show itself? I find something similar when standing to climb in relatively high gears (say 65" to 75") on many different bikes -- climbing thus seems easier than it should, given the feel of these gears when sitting. At any rate, I've for a long time wondered if "planing" isn't relative to rider weight and pedaling torque (itself due to pedaling style and gearing), as well as to frame design.
ReplyDeleteBy "3 speeds" you mean the perennial AW hub? What I don't understand is why Sturmey Archer chose the ratios they did -- 33% overdrive, 25% underdrive. The 25% makes sense; with a 60" middle gear that gives you a 45" climbing gear. But a 33% increase? What is that good for? And it's really too bad that SA/SunRace chose more or less the AW gaps for the S3X hub, instead of the apparently much more reasonable ASC ratios with high as direct and -10% and - 25% reductions. Thus, while an AW might give you a nice (for a city bike) 45'"/60" low and middle, it also gives you a 88" high -- far too great to be useful. The ASC would give you 45/60/66" -- much, much more useful.
Does anyone have an explanation?
I think I do. 3-speed hubs are really simple. You have your choice of "direct-drive" or "not", but in the "not" gear, you can reverse the input and the output, which is the difference between 3rd and 1st. If the gearing is a simple 4:3 reduction than your low gear is 3/4 or 75% of direct-drive, for a 25% decrease. Reverse input and output and your high gear is 4/3 or 133% of direct-drive for a 33% increase. I think that's how it works.
DeleteIt's a simple design. 25% down and 33% up are 3/4 and 4/3; the hub can use the same set of pinions to create both ratios, simply alternating which side gets locked to the drive shaft and which side gets locked to the hub.
Deletehttp://sheldonbrown.com/internal-gears.html#how
I haven't used IGHs before. Is it really that critical to use the direct-drive as your primary gear? Based purely on the ratios, I'd be inclined to try using the high gear as primary and having two levels of bailout.
I understand the inverse ration restrictions on epicyclic gear systems, but what I want to know is why SA chose this one, for utility bikes of all things, when there are other patterns available, like the ASC's direct/-10%/-25% (from direct), or the AM's bottom/+15.5/+15.5 from second, or the very simple TF direct/-25%. What does that + 33% add to an already overgeared (68 or 72" for Indian roadsters) direct?
DeleteTupolev: no, you don't need to use direct as primary. With the modern S3X fixed hub, where direct is high, many use direct/3d/+33% as a downhill gear (again, this hub allows no coasting unless you screw on a separate freewheel), use 2nd/-25% for cruising, and bottom/-25% from direct as climbing. I tried this: 93" high, about 72" cruising 2nd, about 52" 1st. (Still not as nice as the direct/-10%/-25% from direct of the ASC fixed hub.)
>"no, you don't need to use direct as primary."
Delete>"for utility bikes of all things"
If we're talking about being able to ride at low-intensity on a heavy or possibly loaded bike in a hilly area, an AW with high as primary sounds considerably preferable to the alternatives you mentioned. Less cadence control, but you have to get into a much worse situation before the ride is forced to become strenuous.
ASC's DD/-10%/-25% looks like race gearing.
I own two Raleigh Twenties and two Raleigh Sports. As seems to be the case with Raleigh AW three-speeds they were all over-geared from the factory. I solved he problem by replacing the Twenties' 15 tooth sprockets with seventeen teeth units. Similarly both Sports,fitted with 16 teeth were replaced with 21 teeth units. These low cost mods have transformed their design characteristics, and made them much more practical and pleasant to ride. I don't miss the original "top" ratio one bit -which was too high anyway.
DeleteI have several different types of bikes, but the 3 speed is my go-to ride for general fun. I put a larger cog on the rear Shimano hub-a 22 I think. Cost less than $10 and 15 minutes.I can climb over most local hills and bridges easily. Coaster brake and one shifter cable. Simple. May add a front brake- seems like a good safety measure for a chain break, Love a 3 speed.
DeleteThis has reminded me of something in relation to the previous blog entry which mentioned surprisingly light bikes from the 1930s (and the gain in weight from ~the '60s before dipping again with modern materials and technologies): those old steel lightweights may have got impressive weights from artisan working of high-quality steel alloys and butted tubing, and 8kg compares well to many modern bikes (though the current UCI minimum sets an effective floor which could easily be breached if it weren't there), but perhaps at the cost of flex? And I mean here the bad kind of flex. Difficult to make direct comparisons though while respecting the integrity of antique lugs – and not necessarily greatly relevant on the road, as opposed to track.
ReplyDeletevery cool. i wonder how your bella ciao fares in this respect. because certainly we have designed them exactly with the experience in mind that you so wonderfully describe.
ReplyDeleteI didn't want to make it brand-specific, but yes sir the Bella Ciao is exemplary in this regard. The other modern 3-speed I felt this on was the Bobbin Birdie. Otherwise, modern bikes tend to be a little too stiff, I think, be it in frame or wheels, compared to the original 3-speeds.
DeleteIt sounds to me like you are absolutely experiencing "planing". The frame flexes during the peak of your power stroke, and then gives back most of that stored energy (since steel is elastic for small deflections) during the weaker part of your pedaling rotation.
ReplyDeleteThis is actually more likely to happen on a mixte or step-through since those frames are more flexible, owing to a lack of direct connection between the top of the seat tube and the top of the head tube, and thus much more likely to flex under pedaling load. Furthermore, your lack of a low gear forces you to put more oomph into the pedals, which also increases the flex of the frame compared to spinning uphill in a lower gear.
Mixtes and step-throughs tend to be a bit heavier than diamond frames because they use heavier tubing to combat some of that flex. Furthermore, that type of frame is more likely to be ridden by a weaker rider, so most users never experience the flex you are experiencing.
That is my theory, and it is mine.
Daniel M
Berkeley, CA
This reminds me of an amusing and inspiring attempt at this year's Dirty Dozen: every year, the Saturday after Thanksgiving, we race up the thirteen steepest hills in Pittsburgh. Some people compete "for points," (i.e., to be one of the first few people up the hill), while the vast majority compete to complete all thirteen hills.
ReplyDeleteThis year, we had a rider (Jeremiah Sullivan) attempt the hills on a bike share bike, which is 7-speed (internally-geared hub, i.e., Shimano Affine), upright-positioned, and quite heavy. He made it up ten of the hills before his legs literally locked up. (See http://triblive.com/local/allegheny/11538272-74/sullivan-dirty-challenge.)
I have found that having very low gears prevents the learning you're talking about. When I head up a hill I find myself shifting way down, though, technically, I probably don't have to any more. But OTOH I'm not racing anyone, so why not.
The fixed gear is probably the best tool I've ever found for becoming a proficient climber, evenbetter than a freewheel single speed. After a couple of months you'll surprise yourself at the hills you can climb in a mid seventies gear. Before the fix was a hipster fashion accessary it was a serious early season training tool.
ReplyDeleteWhen a fixed wheel was a universal training tool no one fitted a gear higher than 48x18, also known as medium gear, normal gear, or standard gear. That would be 72". Many competitions had a gear limit of 72". Only fast riders on flat ground even used a gear that high. The whole idea of fixed training is to spin.
DeleteFor someone who uses a fix as a daily driver I think the classic 42x16 (71") is hard to beat. I ran that for years. My current set up is 48x17 (76"). I get a little better chain and sprocket life out of the bigger gears and my chubby little legs still get enough spinning on the back sides of hills. I know what you're saying about the old UK timing clubs limiting to a medium gear. I think it would be fun to bring back a time trialing event on classic steel bikes. No carbon, no clipless etc. Sort of a L'Eroica for old British style time trialing.
DeleteOh yes, many miles on 42x16. I fancy that as 70.9", to make myself feel better about flirting with limits. Never, even when young and strong enough to do 25 miles inside the hour, could I have managed a 48x17 fixed. And on hills it sounds so unlikely.
DeleteCurrent fixed bike is 48x19. Gives an easy 20mph cruising speed. Recently on a club ride I was chided for pulling too hard in that gear, 28mph they said. But on hills even the 48x19 sends me straight to the back.
"The whole idea of fixed training is to spin."
DeleteWell, I must be an outlier, because 20+ years ago, just before I got my first fixed geared bike, my favored cruising gear was 64 or 65 inch, and I could maintain 20 to 23 mph in this gear on the flats. I used Coppi's style: pedal, coast, pedal, coast. Then I started riding fixed, and my favored cruising gear gradually increased from 63" to 67" to 70" to -- on my light gofast -- 76". My cadence dropped from 104-112 while cruising on the flats to under 90. OTOH, climbing in a 70 inch plus gear -- no problem!
As long as the gear is somewhere close to appropriate, fixed-gears offer the nicest climbing, in my experience. In particular, my older than me Nishiki Olympic 12 that's a little too big for me and running 42/14 is perfect for the limited hills in my area--the largest of which I have to climb to come home from any direction. I'm a wuss for climbing, though.
ReplyDeleteAnything that wobbles or oscillates or flexes has to match itself with the user. As a kid in my neighborhood we used to pretend we were in the Summer Olympics and set up a number of running, throwing, and jumping events. It was curious that I could throw the javelin (an old fishing rod) much farther than kids who were superior in strength and athleticism. For them it flexed too much. They hated me.
ReplyDeleteI've been riding my 3-speed in my hilly area to work and back, and on errands and to meet friends, on average 8 miles a day, and my hill-climbing muscles have definitely gotten stronger. Those include stomach, back, side, and arm muscles! I tend to rock the bike side to side as I'm climbing hills standing on the pedals — this is probably bad form, but I can't help myself. I keep wondering if I should get a bike with more gears to have an easier time, but I like the simplicity of the 3-speed shifter. I also really like being able to change gears when I'm stopped at a light.
ReplyDeleteWhat's your gear ratio on the outer sprockets? If you're still using the stock 18-tooth cog on the back, try switching it out for a larger one. You'll need to lengthen your chain (or get a new one ... $9). I tend to prefer a 20-tooth just to maintain some top speed, but many prefer the 22. And standing up to climb hills is probably worse for the hub internals than it is for yourself - internally geared hubs put their load in the middle of the axle where the sun gear is, far from the bearings. On derailleur-geared bikes, you'll notice the freewheel is at one side of the axle, on top of the wheel bearings, so standing up is not so much of a problem.
DeleteI followed Velouria's early advice and put on a 20-tooth cog, with two more chain links, soon after I bought the bike on Craigslist.
DeleteI like how you make stories out of every little occurrence on the bike, I do too. My kids and friends politely nod, at best, or sometimes just turn their attention elsewhere. I don't get it. This thing we do on a bike and the insights are important, right?
ReplyDeleteSpare the innocent children, that's what the bike blogs and forums are here for : )
DeleteHah! I'm thinking the other way, so grateful for these kids to keep me grounded. What I've learned about kids…They're not so innocent and they're way smarter/insightful than we give them credit. I'll ditch my bike and my stories in a heartbeat in exchange for time doing anything with them. Blogs, I'm thinking are part of the problem, not part of the solution ;(
DeleteFirst place to look for flex on an old sports three speed is the wheels. Since you are now doing your own wheel work try pulling up the three speed wheels as snug and straight as the wheels on your good bike. Should make an immediate difference.
ReplyDeleteThe rear fork ends on a sports are thin. They will flex. Used hard in hilly terrain eventually they break. When the time comes replace with something heavier. The chainring also bends, especially going slow uphill. The only way to fix that is a complete new crank. When the teeth on the ring are getting thin and sharp start looking for a better crank.
On a 14% grade the stays will be bending. The main frame is built of thick heavy steel. Steel is steel, all one modulus. Compared to lighter bikes the main triangle will bend so little you could not notice.
Nice to see ol' three speeds getting some love! Like you, Velouria, I have gotten used to cresting big enough hills on a three speed. Lower gearing (say, a 22 tooth sprocket in the back) and alloy wheels do help, but even before that, I never felt limited in my day-to-day riding on a three speed.
ReplyDelete-Shawn
https://societyofthreespeeds.wordpress.com/
Steep climbing with three speed IGH is best done in second gear, wich is direct drive. You should have either an extra smaller chainwheel in front or an extra larger cogwheel in the back, plus an derailleur. Climbing a steep hill in first gear puts an enormous strain on the hubs internal components.
ReplyDeleteNot necessarily. Pawls drive the hub shell in all gears. Using 1st or low gear to climb engages the planet assemble which is a robust mechanism. There are just more moving parts engaged. Proper lubrication is more critical than choice of gears.
DeleteI agree with Chris. Do remember that in an AW hub, in 2nd or 3rd gear, the clutch is being held in place by the spring tension on one side so prolonged excessive pedaling force can cause the clutch to "walk" out of gear. This is not the case in 1st gear, as the clutch is held in place by the shifter cable with the spring compressed all the way. It has nowhere to go.
DeleteAlso, the real factor that causes excessive strain on the internals of a gear hub is the external gear ratio. Sturmey-Archer does not recommend a gear ratio of lower than 2:1 for this reason. This also why it is much harder to find rear sprockets with a tooth count of >24, as most internally-geared bikes back then came with 46 or 48-tooth chain rings.
Yes necessarily!
DeleteOnce again, you've accurately worded one of the peculiar things about certain bikes that I thought I was alone in thinking ... the unusual ability for a Raleigh Sports or similar bike to climb hills. And I love how you describe it as being not-too-heavy, not-too-light because that's exactly how I feel about them too. I have had faster bikes and lighter bikes, but none feel "just right" in the sense that they are so well-rounded, for me at least, as to handle everything a daily errand can demand. And the flex! The thought of the flex helping my pedaling crossed my mind, but I shrugged it off immediately thinking it was my superfluous obsession leading me to believe that my Sports (now 5 of them) were so conducive to acceleration.
ReplyDeleteAnd are you saying that your newly-acquired Humber or Rudge were the subject of this post, or do you have another Sports roadster?
Two ergonomic considerations. A Sports frame has a shallow seat angle, places the rider well behind the cranks. Pedaling from behind the cranks is quite different than sitting directly on top of the cranks. IMO the rearward position is always better, for low rpm, high torque riding such as hill climbing the benefit is even more apparent.
ReplyDeleteSecond factor is Sports riders are less concerned about perfect fit and perfect position. The attitude towards a Sports is get on and ride. Thus Sports riders are spared the endless nonsense of modern fitting theories and are blessedly free to go ride a bike.
Allow me to propose a testable hypothesis. I am of the opinion that the hillclimbing ability of 3spd sports bikes is mostly a biomechanical benefit of the slack seat angle. I have done this experiment with my own body as test subject and repeated it yesterday. For me the result is plain, but I am not everyone.
ReplyDeleteReverse the seat clamp. Go for a ride. See what happens on the hills. With reversed clamp the saddle is about 1-1/2" forward, which neatly approximates a bike with three or four more degrees of seat angle. It changes the riders interface with the bike from sports roadster geometry to roadbike geometry. Give it a try and tell us what you think. Generalizing my experience to the whole population is not valid, I have tested my opinion and I know what I know.
Now this is an interesting opinion. I've long suspected that one very big factor in what makes a bike seem "fast" or "lively" is how your body fits on it, and of course this involves geometry. Second, I can't for the life of me begin to see why old roadsters were geared as they were, unless somehow the extremely slack geometry itself encourages slow, torquey pedaling. FWIW, I installed a S3X hub'd wheel on my Rivendell road gofast, which in its usual fixed form (76" cruising, 66" hill gear) is, in terms merely of speed output for effort input, my funnest bike. Since I hate the lash of the S3X downhill, I installed a 12 t sprocket on the hub, to give me a 95" downhill **only** gear, with a 71" cruising/2nd and a very draggy 59" climbing gear. The 95" high isn't too much bigger than the typical 44 X 18t roadster gearing with a AW (91"high).
DeleteEven on this very light (sub 18 lb) gofast, the 95" gear was very awkward to use in any situation except long downhills. Even riding it hard on the flat with a strong tailwind, it felt too high. And again, this is a light bike with the best of Compass tires on it, not a 50 lb roadster with primitive rubber.
It has to involve the roadster's geometry.
Anyway, very interesting, and I'd be very interested to hear more on the subject.
We've recently (5 weeks ago) had our first child. As a result I've been riding exclusively on a direct drive trainer, which is a novel experience for me. Pretty much the opposite to what you've eloquently described above :-). It's taken 5 or 6 hours to get used to the lack of give and change in technique required to be effective...
ReplyDelete