Friday, September 12, 2014

Snowbike geometry.

A few weeks back I shared some pics and minutia relevant to my new snowbike.

For years I've been meaning to write out the pertinent numbers explaining what I like for snowbike geometry, and more importantly how and why I've arrived there.

I finally took an ~hour a few nights ago and did it.

Disclaimer: Below I've shared a bit about the journey I've taken with respect to riding fat tired bicycles on snow.  I do not purport that any action I've taken or design I've settled on is the only, one, true, undeniable way forward.  The world is big, conditions vary drastically from region to region and even hour to hour, and I am just one person with limited finances and time, doing what I can to notice things, think about them, improve upon them.  A lot of the conclusions I've drawn and solutions I've found were based on extensive trial and error on a series of ever-improving bikes, tested in extremely variable conditions.  I've verified that the changes I detail below are indeed substantive improvements by hopping onto most of the commonly available fatbikes and taking them out on the same trails, in the same conditions, to compare.  Still, I'm just one rider with my own set of experiences and biases, and I don't expect anyone to take these tests or conclusions as indisputable proof of anything.  Rather, I hope that the ideas give you reason to thinker on and experiment with your own setup, in your own conditions, and to draw your own conclusions.  

* * * * * * *

Snow.  Kabloona say that eskimos have a hundred (a thousand?) words for it, but that's sort of inaccurate because what they really have are a heap of descriptors for the different ways that snow manifests itself in their daily lives: round and styrofoamy snow skittering across the ice, heavy snow that's wet from overflow, wind-driven snow that gets up into your eyes, snow that bends then breaks branches, snow that snowmachines get stuck in.

Of these, the one that I am most interested in is that light, dry, airy, almost moistureless snow that falls in the early winter in my backyard--the mountains of Colorado.  This snow falls in decent quantities--a foot or two at a time--and then the skies clear for a day or a week, or even two, and the dry, cold air above sucks what little moisture there is in the snowpack right *out* of the snowpack.  Skiers refer to what's left as 'sugar' or 'rotten'.  Drop a ball bearing into it and that BB will keep moving downward until it hits bedrock, as there is absolutely nothing to slow it down--no crust, no moisture, no layers of thicker and thinner.  

Run over it with a snowmachine or even a snowcat, and it packs down somewhat, but it does not stick together.  Try to make a snowball out of it and you'll soon find yourself either frustrated or laughing, because unless you add water this snow will simply not adhere to itself.

Why belabor this point?

Because if I'm riding a bike, on snow, these are the conditions I get ~90% of the time.  This kind of snow is difficult to ride, at best, and more often difficult to wade through while smacking your shins on the useless bike you're dragging next to you.  And of this moistureless snow my backyard mountain gets copious quantities: 33 feet a year, on average.  Riders trying to learn this medium need to think about it three dimensionally, for they will be within it more often than atop it.  

To be clear, when I speak of snow I'm not referring to that moist, packable stuff that you lucky bastiges get in Anchorage or Minneapolis (aka the twin centers of the fatbike diaspora) or coastal BC--the kind that quickly sets up into white concrete and that you could ride a MTB or even a CX bike on.  That simply doesn't happen here.

Not coincidentally, the snowpack of my backyard is also somewhat common in...

(wait for it...)

...Interior Alaska.  They don't get as much of it up there as we get down here, but the end result is the same: snow that doesn't pack well, blows around often, and has little to no base beneath. 

I've spent a good portion of my adult life fanatically (not too strong a word) obsessed with finding ways to be efficient when riding on and in this medium.  Specifically, I set as a goal over 20 years ago, before I'd ever been to Alaska, that I wanted to ride every inch of the Iditarod Trail.  To date I've ridden roughly 6500 miles on that single trail, including 4 complete Knik to Nome traverses, all in winter conditions.  With each passing year I delved deeper into learning how to be safer, faster, and (most importantly) more efficient so that I could go still further, with less of a safety net, and still feel confident that I'd emerge out the other side.  

The net result of that fanaticism is an acute awareness that what works to keep a rider upright and pedaling through this kind of snow is very different from what the major players are pushing right now.

They've got the unenviable task of trying to please all the people, all the time.  Think about that for a second: They have to compete on price first and foremost, now that everyone's vying for a piece of the pie.  And they don't want to alienate a potential customer, so right off the bat they're convinced that they need to make their bike fit 6 different racks and 12 different bags and 7 different front derailleur standards, plus have 13 different bottle cage mounts as well as remain compatible with every crank and chainring and q-factor option.  Plus fenders!  In trying to please everyone they're making too many compromises, chief among them is that in order to fit 3 chainrings *and* a 5" tire, they have to lengthen the rear center by over an inch.  An inch is a significant number when it comes to bike geometry, and in this case it means that the rider's center of gravity is another inch removed from the rear axle.  That arrangement works fine on hardpack and singletrack.  But this is a fatbike, in my case a snowbike, and how it handles on hardpacked singletrack is of little interest.  You can ride *any* bike on a hardpacked surface, but if you take just any bike to the above-described soft surfaces you will be disappointed.  And you won't ride much.

I'm grossly overgeneralizing on this next sentence, simply to make a point.  What the manufacturers are doing is making average bikes for average people.  That is, the bike that doesn't offend anyone's sensibilities while still remaining somewhat attractive and reasonably affordable.  

There's nothing inherently wrong with that.  In fact from their perspective it's just good business to make a product that appeals to the masses, not a teeny, tiny niche.

They're also gambling to a large extent that few buying these bikes will ever ride them on snow and discover that:
1. Snow riding is difficult, slow, and unexciting, and,
2. The geometry they just sold you sucks frozen monkey ass.

And while length (chainstay, wheelbase, cockpit) is not the only consideration, the amalgamated blend of those big three is the top of the heap.  (For those of you speed-reading through this in hopes that I'm going to come right out and say "X head angle with Y BB drop and Z chainstay length is *the* magic ticket, just stop.  It's not ever that simple.)  

When I wrote the initial post introducing this bike, I stated that:

'I had a custom frame built because although everyone seems to make a fatbike these days, none of them come anywhere close to geometry that really, truly works on snow.  99.9% of the people buying and riding fatbikes these days don't know any better, and 90% of them don't care.  Most are simply happy looking down on their gee-whiz bulbous tires, thinking that the tires are the most important thing.

No way.' 

In that statement I was referring to the fact that few people are riding these bikes on snow to begin with.  And of those very few that are, almost all of them assume that tire size and pressure are 'the whole deal' when it comes to being able to ride instead of push your bike.  It's becoming more understood that pressure *is* hugely important ("when in doubt, let air out") but what happens when you're struggling along at 2psi, virtually riding on the rims, and that's still not low enough?

In the last two decades I've been in that scenario countless times, and each time I've asked myself 'what can I do to make the conditions underfoot rideable, given current rim and tire technology'?

To the end of finding answers to that question, I've designed, paid to have built, and extensively ridden 8 different snowbikes.  By extensively I mean tens of thousands of miles in the last 18 years, mostly in the above described conditions.  With each new bike I had to go in knowing that no matter how much we tweaked the design, what we would arrive at this time was still going to be a compromise in some way.  Every bike is.

The first custom had 18.9" chainstays.  If you ever want to know if a certain change in geometry will make a difference, exaggerate it.  The next one had 17.2" stays, and although the rims and tires were identical between the two, on the shorter bike I could maintain traction effortlessly by comparison.  Soon rims got fatter and tires got more volume, yet not until almost a decade later did tires get fat and have reasonable tread.  In that interminable "Remolino/Endo/Larry/BFL" vortex it was a given that while your tires would have some float, they'd have zero effective traction at any pressure.  My succinct way to describe the handling of a bike with Remo's or Endo's on snow was, "It goes sideways almost as fast as it goes forward".  So I did everything I could within those constraints to make my bikes float, dig, and track better.

Wider rims helped a lot, by squaring off the profile of the tire. You still didn't have actual edge knobs with which to lean and/or dig, but by removing the round profile you could at least gain a measure of consistency.  Wider rims meant that in sugary or wind-affected snow, your wheels were less likely to squirt out from beneath you.

Shortening the rear center was big.  Think about it this way--the medium on which you are riding is dynamic: Shifting, changing, moving beneath your tires.  You, as a human, are pathetic and weak, able to put out a whopping average of one horsepower on a warm, sunny, scantily-clad June day.  But in January, entombed as you are in layer upon layer of Windshopper, Poor-Tex, and Primacost, and with subzero air on offer to your torched lungs, no way you're putting out even that much.  So you have to maximize what you can put out, and you do that by bringing the rear wheel up under your center of gravity, the better to keep it from slipping and spinning when the snow can barely hold itself together.  Put differently, by bringing the rear wheel more underneath your body weight, you minimize the amount of body english needed to maintain forward momentum in marginal conditions, which means you move forward more, using less energy.   

The thing about shortening the rear center is that while it gains you massive amounts of traction out back, it also changes the handling of the front end of the bike.  Now (all else equal) there's not much weight over the front, so the front wheel wants to wander and wash.  No free lunch.  What to do?

You do two things.  

First, drop the BB.  This brings more weight forward over the front wheel, without having to resort to long/low stems, or excessively short front center measurements.  Snow is quite tolerant of a low BB, with the ancillary benefit that a low BB means added standover.  Next time you're riding in soft snow, and you're forced to dab, and your foot goes more than ankle deep, and keeps going...

...then you'll probably realize how important a healthy amount of standover really is.    

Next, pick your favorite blend of HTA and offset to net yourself a LOT more trail than you're used to.  I like a trail number of about 100mm on my snowbikes.  That amount is a compromise like any other number picked out of thin air, but it's a compromise that allows the bike to track straight on flats, carve corners when properly weighted, and remain neutral in ruts and off camber.  At the risk of redundancy, we're talking strictly about soft, baseless, marginally-rideable-at-3psi snow.

Once we'd experimented with these three big changes and saw the general direction we wanted to go, it was time to do further experimentation to determine how far to go.  To that end we experimented with seat tube angles to mimic the effect of shortening the rear end--and learned that getting weight directly over the rear axle is probably the Most Important Thing.  How you achieve it (slack STA with straight post?  medium STA with a setback post?  steep STA with super setback?!?!) is debatable, and probably always will be. 

This is getting long winded and there are probably all of 3 people left reading, so I should wrap it up.  How much to shorten the rear end, how low to go on the BB, how slack to make the front end--these are all questions to be answered by individuals and smaller builders, the people that actually see the value in faffing about with bikes, and then going out to see what their faffing has achieved.  Leave those questions (and answers) up to the bigger players and you'll continue to get average bikes that struggle where the hardpack ends.  Beaten men follow beaten paths, and all that.

For my part, I have yet to find the point of diminishing returns with respect to a short rear center.  The bike pictured here has a chainstay length of just a hair over 17".  That was as short as we could go while still maintaining clearance for a 5" tire on a 100mm rim, as well as a 29+ tire and rim.  Someone should go shorter still.  

Eventually, I'm sure I will.