What's your favourite gear ratio?

You getting back into roller racing bmmf?

Nope. I've had enough of 0 to >200rpm in ~2s.

I'm just interested in hearing about the physics involved in a snap situation, not a standing start, as anyone who claims to accelerate better on a smaller ring is talking about jumping from something like 15/20mph to 35/40mph, which happens in the space of a couple of seconds, with the first massive peak burst happening in one or two crank revolutions. There's nothing 'snappy' about a standing start compared to a jump from around 80rpm up to 160rpm or whatever.

The physics doesn't really change much, regardless of the start and finish speeds or the time over which it happens. Because the inertia of the bike+rider is huge compared with that of the drive, the power needed to accelerate the whole plot is vast compared with the power needed to accelerate the drive components, so a small change in efficiency in the drive will correspondingly swamp any change in drive train inertia. With the whole drive train (ring+sprocket+chain) being about 0.5% of the system mass, and the rotation of the chain only about 0.05% of the eventual kinetic energy, it's just irrelevant.

Taking your jump, by the way, at the lower end of your example, accelerating from 20mph (9m/s) to 35mph (15m/s) in 2s needs 36W/kg plus aero and rolling drag. You can get about 15W/kg by dropping 3m from the top to the bottom of the track in those 2s, about half of which will be used just to maintain constant speed leaving the rider to find about 28W/kg to drive the acceleration. That's just about within the scope of elite track sprinters. 20mph to 40mph in 2s needs >50W/kg even after using the potential energy from dropping down the track.

On the rollers, assuming the usual cheating set up at Rollapaloser, the driven load is just a back wheel, tyre and a light roller, which only has about 25-50 (depending on the mass and geometry of the roller) times as much kinetic energy as the chain on its own, so the change from a very small drivetrain to a very big one could conceivably make a difference of 1-2% to the power required to accelerate, which is of the same order as efficiency difference during the 2s spin up. However, even if your small-ring drive gets you up to speed a couple of hundredths of a second sooner, you lose it all again in the subsequent 18s constant speed phase of the competition, where the higher efficiency big-ring drive wins.

Thanks. Regarding the 20-40mph jump, surely it's still within the scope of elite trackies if they've dropped back a couple of bike lengths so that they can jump into the slipstream of the rider ahead?

I don't think the snap capabilities are limited to the elite anyway, just the genetically predisposed with a moderate dose of specificity in their riding (but yes, 40mph is a world apart from 35mph). The elite part is the ability to consolidate a number of skills/abilities: snap, top end speed, sustained power at higher cadences, optimising biomechanical aerodynamics, tactics, mental attrition, blah blah blah.

Regarding the rollers, there's plenty of times where that 'getting up to speed a couple of hundredths of seconds sooner' is enough to mentally break your opposition when they sense you're ahead (or sneak a glance over their shoulder at the clock) - but it's moot as RPLZ use identical set-ups apart from frame size.

Giving 44/16 a go tomorrow as 44/15 was feeling a bit much for pulling away from lights without digging deep.
Fuck me... this thread.

I'm a Physics failure, but maybe we should just sneak a look at the British Cycling machines and presume that in their pursuit of marginal gains they might have got it right. Then deduct several gear inches on the basis that they're really very good/strong/terrifying and we're a bit shite by comparison....

Suggestions for a fixed ratio for Alpe d'Huez please, I've drunk too much wine and this idea has occurred.... 39/25 has been kind to me in the past on a road bike, but should I go higher or lower on fixed?

Going up or down?

Regarding the 20-40mph jump, surely it's still within the scope of elite trackies if they've dropped back a couple of bike lengths so that they can jump into the slipstream of the rider ahead?

No. They can hit that speed, obviously, but they can't go from 20mph to 40mph in 2s. Whatever the slipstream deducts from the aero drag, it does nothing about the sheer amount of kinetic energy which has to be gained.

Also, the bloke in front has no slipstream to hide in, and your benefit drops off very fast as you get further back, so the bigger your speed differential compared with the bloke you're behind, the shorter the time you have in effective draught. To be in his slipstream from two clear lengths back to running into his back wheel, you have 2s with a speed differential of 2m/s, or around 4mph. So, if you're following 2 clear lengths back at 20mph, he puts the hammer down, uses the banking to max effect and has the ability to put out 28W/kg, he should hit 35mph in 2s. You precisely match his jump and experience zero aerodynamic drag until you hit him, you still need to put out 45W/kg to run up to his wheel in those 2s. It's just not credible, given what we know about the power outputs of elite sprinters over that sort of duration.

Great thread, I still don't know what side people are on but I like big rings, not for performance because mine's shit anyway but for how smooth they feel to me.

Oh and new ratio for town will be 53-20

new ration for town will be 53-20

^^^^^^ Lynchman; up only, sod descending the Alpe fixed! Would be part of an Alps & Dolomites trip with a car and She who must be obeyed. Other passes would be geared only, I'm mad not completely bonkers....

The guy who climped The Bald Mountain or was it AH? on fixed (3 times including downhill - smoking brakes) think did it on 48/19 on the first two climbs and 48/20 on the last effort (sounds 20GI to high,argh). There is a blog about the attempt somewhere on the internet.

Tester.

I;m sorry but I haven't got my ration book but I do remember my National Registration Number. KSVU 120 5

Neil

I thought 49x18 felt like a big jump from 47x18... Turns out my speedy commuting hasn't been down to the gym training I've been doing, more the fact I put the wheel back in the wrong way round and have been running 49x16 by mistake... I quite like.

I really like 49/16... at about 79 gear inches and 6.3m of development on racing tires, it's a little high for frequent acceleration and deceleration, but great for descents and cruising.

Choosing the "right" gear ratio is often an academic decision, and in that respect it's quite ironic that you found "your perfect ratio" by accidentally flipping your wheel the "wrong" way round... ;-)

53/20 70GI, very nice.
Finding it hard to hit anything over 35 though.

Looks good too aesthetically speaking...

53x20 is what I've got on the Sirius... Struggling to hit 35 on the flat or downhill? Kph or mph?

sheldon says 49x16 is 82gi with a 28c tyre, s'ok, but I wouldn't fancy a long, hilly or windy ride on it. Great for quick blasts though (commute is 8 miles)..

35mph on slight downhill, I'll find the % in a sec.
I have hit 46mph there though, can't remember if that was 53/16 or 53/18 though.

Like a boss...

*Speed is measured in km/h.

http://app.strava.com/segments/3272036
This segment, so -0.9% gradient, highest speed I reached was before I got strava.

Surely you mean -9 percent...

Speed is measured in various units depending on what you're measuring, I like knots, mach and using the speed of sound in pure water as a unit, failing those mph is fine.

There's always one...