Saturday, May 25, 2013

KCNC Skewers

Sheldon Brown teaches about the difference between "closed cam" skewers and "open cam" skewers.  The simply perfect Shimano skewers are an example of closed cam skewers.  I'm sure there are many examples of open-cam skewers, but one is my KCNC skewers that I bought with my custom-made wheels from White Mountain Wheels.

In 2011, after we got the bikes ready, gave some unfortunate sound-bites to the documentary film crew, and had a great ride from Valdez over Thompson Pass and into the late-evening half-light, my KCNC skewer opened on my front fork, a bump caused the fork to separate from the wheel, and I dented the pavement with my (thankfully helmeted) melon.  That was the earlier version of the steel Gunnar fork without lawyer lips.

Since then I've been manic about checking and re-checking my skewers while riding.  I nonetheless convinced myself the first failure was operator error and I put the skewers to use with my Hed wheels on the rebuilt P2SL.  Today near the end of my ride I was descending a slight grade when the rear wheel separated from the bike, rolling out the back of the horizontal dropouts.  The bridge of the rear brake landed on the tire and I was able to stop without incident.  While replacing the rear wheel, I noted my front skewer had worked itself slightly loose as well.

Looking for others with similar problems, I found one advertising pitch:

"KCNC skewers are some of the most exotic and durable road skewers on the market. Get your KCNC products from Zen Cyclery."

Sheldon writes about open cam skewers that they were considered to be better because they could be made marginally lighter than closed-cam skewers.  But, he concludes, "The exposed-cam skewers are generally OK for vertical dropouts in back, and for forks with "lawyer lips", but should not be relied on with horizontal dropouts or plain forks."

I suppose I could put them back on the Gunnar, the one bike I have in regular use with vertical dropouts and lawyer lips (on the new fork).  I think I'll go a step further, however, and pitch them.  Saving a very few grams of non-rotating mass just is not worth it.

Thursday, May 23, 2013


There's something special about riding a fixed gear. It's cliche to say, but when your effort, cadence and speed are all correlated, you feel a little more connected to the bike, and to the road. Coming to a stop involves nothing more than gentle (or, occasionally, violent) backpressure, and accelerating afterward provides a nice opportunity to stand and stretch the legs.

But really the appeal I think is the silence. No clicking Ksyrium hubs or clacking shifting chains. For some reason seldom on a fixed-gear ride do I feel the need to ride fast or far. It's just a gentle romp through the countryside.
Romping Gently Through The Countryside

After several years without a fixie in the stable I procured a Surly Cross-Check recently. It arrived as a single-speed with flat-bars, which I immediately replaced with a fixed cog and drop bars. The bike came with some burly tires, but I swapped those out for the Ultremo ZXes that I originally bought for my Habanero. I went out for a short ride a week ago, and took time to see some sights that I've ridden past hundreds of times, but never stopped to absorb.

Gliding Past The Glider Airport

In some ways a fixie reminds me of a glider. I guess that's because I assume that a glider is quiet and peaceful. Of course in most other ways a fixie is absolutely nothing like a glider.

Off-Roading In Washington County
The roads near my house are perfect for riding a fixie. It's easy to go 10-20 miles on good quality roads, with gentle rolling hills and nary a stop sign or traffic light. The scenery isn't magnificent, but it is remarkably pleasant. If one really wants, there are plenty of long steep climbs too. But on this ride I stayed with the rollers.

I'm pleased to be back on a fixie. The Cross-Check is a fine ride. I think I'll be using it quite a bit this summer.

Wednesday, May 22, 2013

What's weight got to do with it?

I was intrigued to see this data from my ride today.  Average wattage was 208, which is about what I hope to put out in my next IM bike leg.  (Adjusted wattage of 230, according to some algorithm that surely accounts for the warm temperature.)

That wattage produced an average speed of 17.6 over a moderately hilly (3250 feet) 50 miles.  I anticipate closer to 20.5 mph in an IM race.

There are two obvious differences.  First, I was riding a road bike that is set up to be comfortable.  (It is comfortable, too.  After two weeks of riding the two Cervelos, getting back on the Gunnar felt like settling into a Lazy Boy and grabbing the remote.)  Surely there is some aerodynamic penalty.  Second, the bike was loaded more or less how it will be in the 600K 10 days from now -- and not much lighter than it will be in the 1200K in July.

My perhaps naive conclusion is that the weight made a real difference in the amount of work required to maintain basic speed, in light of the constant rolling hills.  I further conclude that it is easy to understand why holding 18 mph in a randonneuring event is so doggone hard to do.

Sunday, May 19, 2013

The Bike Leg

Today was the Columbia Triathlon, a notorious and uber-brutal early season race run just north of DC. The race has a massive international following, draws a great pro field, and is a must-race for a large share of the local talent.  An imperfect but telling comparison:  with a middling performance last year in the Nation's Triathlon, perennially the largest one-day triathlon in the country, I was 6th in the usually competitive 35-39 age group.  Today in Columbia, with what I would assess as a much stronger (if not as fast) race, I was 17th in what is normally a less competitive 40-44 age group.  141st overall.

I won't bore our hordes of readers with more about triathlon.  Let's talk cycling.  I'm a second- to third-rate talent in a triathlon bike leg.  Last year in Columbia the just slightly more than 40K bike leg took me somewhere in the range of 1:14-1:15.  (The exact split is unknowable because the reported bike split -- 1:16:54 -- includes the time I took in the first transition, due to a timing error.)  This year was cooler, but also wet and slick.  I got off the bike after 1:08:56, just shy of 22 mph for the course.  Here is data on my bike leg. 

What changed?  I'm probably a little fitter, with more miles, more hills (spring break alone was 425 miles worth!), and more high pressure work indoors on the trainer, than last year.  That's part of it.

I have a revamped ride.  In Boulder last summer, embarrassed to be the only person in town with a metal bicycle, I bought an updated carbon P2 frameset, had it built with components from the old P2SL, and made some other tweaks.  Included are shorter crank arms and cool 52-36 gearing.  New off-brand carbon wheels as well.

I'm riding what Jan Heine reports are the single best production tires, the Vittoria Open Corsa CX clinchers, and running a lower pressure per the recommendations in Bicycle Quarterly.

All that taken together may make a difference.  I doubt it, though.  The big difference was chucking any concerns about feeling good on the run and keeping the pressure up both up and down the hills.  By the last few miles I was praying for the end, whether that meant reaching transition or flatting so I could bow out honorably.  The result was a very rare (for me) bike leg that was more competitive than my overall result.

Tuesday, May 14, 2013

Ritchey Break-Away Frameset

Can I possibly communicate here how badly I want this frameset?
No, I don't think I can.  It comes with a carrying case that will fit the bike, broken down, with a total linear inch count of 66.  That is 4 inches greater than the airlines technically allow, but everything I have read suggests you can get away with it at check-in. 

The titanium version runs close to $3000.  The steel version runs $1000 even, if you look around, including the carbon fork.  Add saddle, seatpost, stem, bars, and wheels that I already have, plus a $500 S-105 groupset from Merlin Cycles (or less than $400 for Tiagra), and there's a less-than-$1500 travel bike with the same dimensions as my Gunnar.

In the alternative, I am facing a minimum round trip bike charge of $150 (Alaska) and possibly as much as $400 on US Airways (though of course I will not be buying that particular US Airways ticket!).  Oh yeah -- there's Frontier with normal $20 e/w baggage charges.  I'm ignoring that because it undermines my business case for a new bike.

Sunday, May 12, 2013

Bike Review: P2SL Road Build

I described in a below post my use of parts pulled from the Gunnar, together with a few newbies (WTB saddle, FSA compact drop bars and stem, Ultegra FD from the set I bought for the Gunnar, used here instead because it is a braze-on) to turn the P2SL frame hanging in my basement into a road bike.  My design was to create a semi-light, fairly aero, comfortable, and inexpensive bike for racing purposes.

It's something of a goofy-looking rig.  At their best, time trial frames are not pretty, when compared with the elegance of (metal) road frames.  When combined with drop bars the bike looks like an ugly mutt.  And I accented it with white and blue just to be sure nobody missed it.  (Picture in the sidebar on the right.)

As I first built it, the bike was _steep_!  In my first ride I felt like I was holding onto my knees.  I could find the aero position, but it wasn't comfortable.  This would not serve for a 24-hour race!  I turned the stem over, gaining perhaps 17 mm in stack.  I then rode a comfortable 68 miles on Sunday, up the Monon Trail from downtown Indianapolis to the northern suburb of Carmel, then north and west on farm roads through Westfield to Sheridan, and back.  

A few notes rom that ride:

It was a windy day.  Aero is an interesting concept for a bike.  Of course, I did not get a chance to compare the alternative, but despite teardrop tubing and Hed deep-dish wheels, plowing into the wind was just that -- plowing.  In other words, the bike may cut the air like a knife, but it needs me to move, and I seem to catch far more of the wind than the bike.

The torpedo bottle in the stubby aero-bars is less than ideal.  There just is not enough room to handle the bars with the bottle there.  This is different from full aerobars, on which I can grab the bars past the end of the bottle.  I may experiment with the positioning before removing it entirely.

The WTB Rocket V saddle is so far a winner.  (I had used one on Sam's Kestrel in California in March as well.)  I am sufficiently sold that I just ordered the top-end, the Rocket V SLT (leather covering) from Amazon for the Gunnar.  This saddle has a slightly wider seat than my prior favorite, the Selle San Marco Concor Light, and I'm pleased to find it comfortable even though I am still smarting from 260 miles one week ago.  

I was worried about a funny cable routing problem on the front derailleur.  No issues with that, so far as I can tell.  Tuning is good but for trouble accessing the smallest cog.  The compact crankset with 12-27 cassette is a workable setup (though a little out of place riding in Indianapolis!).

I am running Vittoria Open Corsa CX in 23c on the Hed wheels.  I dropped the pressure after reading the tire research in Bicycle Quarterly -- 100 rear, 90 front -- and I found the ride passably comfortable. I cannot profess the excitement about the ride that I felt based on the Grand Bois 28c with latex tubes (400K ride report).  The next step may be to go with latex tubes here as well.

In the final analysis, I am riding this bike because I had parts that were worth next to nothing if sold.  Given the maybe $500 opportunity cost, it is a winner.

Recommended Reading

If you haven't already, the Tim Bird ride report, "Boggle Hole and Back," Spring 2013 Bicycle Quarterly 47-60, is a real treat.  That's a ride report by a guy who knows how to report.

Whenever I read that sort of report I find myself surfing the web looking at the rando-specific rigs like those custom-built at Boulder Bicycles and Firefly, this cool new titanium maker to which D__ alerted me.  You know the type -- tires fatter than the tubing, bars level with the saddle, full integrated fenders, maybe even friction shifting on the down-tube, bar-ends or grips, and generator hubs.

Time for the views of the commentariat:  let us look for justifications!  I (we all, let's admit it) have different bikes for several purposes, up to and including "in case Sam comes to visit."  (This despite the reality that my best performances in biking and triathlon all have occurred on the Gunnar -- perhaps the least likely rig in my stable, per modern marketing literature.)  Does a world exist in which it makes sense to purchase, to build, or some combination thereof, a _pure_ rando rig? If so, what does one need and what can one leave off?

Saturday, May 11, 2013

Taking the Debate Public

D__ and I are heading to Alaska in July for the Alaska Randonneurs' Big Wild Ride 1200K. D__'s first crack at the distance. My third(!), with neither of the first two being successful. Sam should obviously be joining us, but no luck so far. (See the "Ride Announcement" post on April 8.)

The Big Wild Ride is a hard event to segment. There is no natural overnight in the first 600K. Nor is there in the latter 600K! Ordinarily a 6 pm start makes much sense -- the necessary night riding (and some is always necessary) is done on the first night of the ride. But on this one, that means one either commits to several night rides or rides 30 hours straight before sleeping.

I want to ride this loop in 72 hours or less if possible. If it is not possible, of course, I am ready to fall back on the planned overnights and just stay ahead of the sweeper car. My strong sense is that we will know how things are going by the time we have to make our first decision -- whether to sleep several hours in Delta Junction and proceed into the next night, or whether to leave Delta Junction in early afternoon and proceed 95 miles to Fairbanks.

D__, quite understandably, wants to finish.

My preference is to ride to Fairbanks in 30 hours or less. Possible? I say it is. In each of my two 600Ks on similar terrain (even sharing the same road for much of them) I've pedaled about 24 hours total. There are many problems with a 6 pm start, but one advantage is that we will encounter the night early; by 12 hours in, when things get really tiring, it will be full daylight again -- and it will stay basically that way until the following midnight.

The hard part comes in asking what to do next. From Fairbanks we ride to Nenana (50 miles or so), then Healy (another 50), where there is a planned overnight; then Cantwell (another 50 or so). If we stay in Cantwell, we will be 230-240 miles from Anchorage -- certainly possible in the remaining day and 1/2, and maybe even possible on the third day.

If we continue on, we encounter a staffed control at Hurricane Gulch, 35 miles away, and the next available lodging is 75 miles down the road at Mary's McKinley View Lodge. That puts us in easy striking distance from Anchorage on Day 3, perhaps even the 72-hour finish I had envisioned.

Either approach gives the same probability of finishing, because falling back to the back of the pack remains possible no matter what one attempts. In truth, I think this all becomes a game time decision about 10 minutes after the starting gun sounds. I may be napping in a volunteer's vehicle at 8 pm at the Mile 35 control, and that holds whether my plan is ambitious or pedestrian.

The issue seems to boil to this: physically, an "off" day on Day 2 seems easier. The rest-in-Cantwell option means riding 360 miles, 150 miles, 240 miles (the last leg possibly broken into two). The rest-at-Mary's option means riding 360, 225, 165.

On the other hand, psychologically, I like front-loading as much as possible. Waking on Day 3 with 400K in front of me sounds horrific. Waking on Day 3 with 250-275K in front of me is something I can stomach.

By my count, this blog has a readership of 3. What says our devoted following?

Tuesday, May 7, 2013

Gear Recommendations -- UPDATED

Update #3

The DiNotte went for 9 hours yesterday and has now gone for 3 this morning.  That's two full Alaskan summer nights at ~100-150 lumens on a single $7.15 set of batteries.  It's still going, but I'm getting weary of monitoring it!

Update #2

The DiNotte 7.5 hours in on medium.  I see no loss in brightness vis-a-vis the picture at 3.5 hours.  We are approaching a full night's riding in the summer, with a blinkie to handle dusk and dawn (for being seen rather than seeing).

DiNotte XML-AA on medium after 7.5 hours

Update #1

I am testing the Serfas on the second-to-low setting, which in a beam-against-the-wall test seems comparable to the XML-AA on medium.  That is consistent with my understanding that the Serfas puts out 500 lumens on high, then 250, then 150, then 75, and the DiNotte puts out 200/100/50, perhaps adjusted upward by modern LEDs (per Sam's comment below).  The coloration is different but the brightness appears comparable.

Serfas left, DiNotte right

Starting with a full charge the Serfas has gone 3.5 hours before the indicator light started blinking red. (The light is still on, so the current question is how long the indicator blinks before the light goes out.  *15 minutes in and the beam is fading visibly.  *20 minutes from my first observing the flashing and the unit is all but dead.)

That is better than I had imagined, in particular given the unit's sitting unused since Last Chance.  It also may be plenty for a backup use model.  If the batteries on the DiNotte go at 2 am, the Serfas can get you to dawn, at least in the summer.

The next test for the Serfas:  charge fully, wait 24 hours, then test.

In contrast with the Serfas, now more than 4 hours in, the DiNotte on medium with AA Lithium Ion is showing no signs of slowing down.

* * * * *

Stuff to Consider

Damon blogs about electronics (and other gear) he used on our 400K at Remembering Jaron.  My two cents:

1.  The Garmin 810/iPhone link was pretty cool.  I don't know whether it is much more than cool, as a cue sheet plus find-my-iphone does the same thing with a little more effort on the follower's part.  But the software and output that he describes was objectively cool.

2.  Damon's helmet-mount lamp from Exposure was quite nice.  For reasons Sam points out in his earlier analyses of ergonomically attractive lighting systems, I would need to see more to be sure.  For example, Damon's NiteRider lamp performed like my Serfas lamp (which I did not bring) -- i.e., it didn't perform just when we needed it to.  But the Joystick did perform, so maybe somebody has cracked that nut after all.

3.  The 6000 mAh Morphie Juicepack is an elegant solution to a vexing problem.  It even makes the Serfas light (which uses a USB cord charger) a possibility again.  Envision the Serfas and Garmin both bar-mounted, both plugged into the battery pack, for all-night riding.  (Pack would fit easily in a basic bento box or frame bag.)  Charge the Juicepack at a control.  Rinse and repeat.  There is also a ruggedized version for $75 on Amazon.  One caveat:  I had a hunch on my aborted attempt at the Last Chance 1200K that nonstop downpours contributed to the failure of my Serfas lights.  An insufficiently sealed unit will flop no matter how well the batteries perform.

QUERY:  is 6000 mAh necessary, or is the (presumably smaller and lighter) 4000 mAh version sufficient for the use model I describe?


Now back to my own gear: DiNotte lights.  Based on my experience and Sam's recommendations, steady output at 100-150 lumens -- low to medium on most lights -- would be sufficient for nearly all night-riding applications; the sole exception is bombing descents.  (I hear that people bomb descents at night, but I can't personally recall ever doing so myself.)  The 200L with AA battery packs is still available, but you have to dig around DiNotte's website to find it.  (Mine is the XML-AA, which is the same light with the updated XML look.)

DiNotte does not make any huge claims on battery life -- 2 hours on high (200 lumens), 4 on medium (50%, per DiNotte -- 100 lumens), and 8 on low (25%, per DiNotte -- 50 lumens?) -- with AA NiMh or NiCAD batteries.  DiNotte is very explicit (same link) that the XML-AA should not be used with the longer-lasting lithium ion batteries.  This is inconsistent with Sam's experience.  Is the DiNotte 200L sufficiently different from the XML-AA that the permissible batteries have changed?  (Also note that Sam's information on the 200L suggests different output levels -- 100, 140, and 200.)

The 400 lumen DiNotte XML-1 with the larger lithium ion battery pack, $199 from DiNotte, boasts 5 hours use at 400 lumen output, 10 hours at medium, or 200 lumens, and a whopping 20 at low (100 lumens).  On those numbers, for a 1200K in midsummer in Alaska you would need precisely 0 charges -- the light first goes on at 10:30 pm on low, goes to medium around midnight until 2 am, back to low until 4 am, and off until the next night at 10:30 (when any sane randonneur is down for the second night anyway).  This has a huge negative of an expensive sealed battery pack that if lost or degraded needs to be replaced.  The beauty of the DiNotte AA systems is the ease and low cost of changing batteries.

If, as Sam suggests, I can use lithium ion with the XML-AA, I'm set.  If not, I need something that will give more light for longer, at least for night riding south of the 60th parallel.

Monday, May 6, 2013

Leesburg 400K

Saturday, May 4, at 5 am is the start of the Leesburg 400K.  The route starts at the Leesburg Days Inn, an ultra-seedy motel off of Highway 15; heads north to Gettysburg after crossing the Potomac River once; after crossing two state lines (Maryland, then Pennyslvania); and after crossing one mountain range in Catoctin State Park.  The first control comes at mile 67 just before a brief tour of the Gettysburg battlefield.  11 of us start together.

(The map is here.  The Garmin cuts out 15 miles before finishing, which explains the start/finish disconnect.)

Daytime Riding

It is cold at the start! Garmin reports 46 degrees as the low, but I would guess closer to 40, and before the blood gets pumping it is hard to keep warm.  I have left the gloves and leg-warmers behind, so it is two hours before I can claim to be comfortable.  The hills are not bad on the northward leg, although the Catoctin Range crossing is several miles of constant upgrade with some short steeper sections, perhaps reaching at the worst a 12-15 percent grade.  A must repeat on this leg:  Route 77 descending the northern side of the Catoctin Mountains.  Nothing steep, but good quality road surfaces, nice shade, a pretty creek, winding turns, and few cars make for several miles of excellent pedaling and descending.

The ride divides quickly into D__, me, B__ (from Connecticut) and BB__ (from the DC Randonneurs) -- and everybody else.  For the rest of the day we have no idea how the rides are going for the rest of the crowd.  BB__ stays close enough to D, B, and me that we see him arriving as we leave the first two controls, but we then lose track of him after the Sharpsburg control at Mile 116.  B__, from Connecticut, is a very gracious chemical scientist working in the pharmaceutical industry, who made the jump from ultra-running to randonneuring when his hips started going bad.

Leg 2 heads south from Gettysburg to Sharpsburg, passing through the Antietam Battlefield just before reaching the Sharpsburg control at Mile 116.  On this leg we cross the range again in the opposite direction, with the primary climb's coming on Jack's Mountain Road.  Jack's Mountain Road is reputed to be a real climb, but perhaps because of its coming that soon in the ride my memories of it are not bad.  On the other side, after re-entering Maryland, the long straight descent on Rte. 491 is a hoot.  Perhaps 5 miles of good pavement and wide shoulder and consistent speeds above 35 mph.

D, B and I have a real lunch at the Sharpsburg Battle Market.  Like at Gettysburg, we rest perhaps too long -- but we are moving well and the controls are sparse, so it does not feel careless.

(Here is the ride data.  Somehow we average 17 miles per hour moving on this hilly route.  One does not need better proof that downtime at controls kills a fast ride!  The total climbing is, as always, subject to dispute.  The GPS file I use suggests 17,000 feet.  My data says 14,000 not including the last 15 miles, which might believably have included 2,000 more.  D's GPS says 18,000, which is consistent with 17,000 for the real ride plus extra for our detour.)

On the third leg we again cross the Potomac, this time going south from Maryland into West Virginia, somewhere not far west of Harper's Ferry.  We follow Rtes. 9 and 1 for 20 or so miles before re-entering Virginia near Winchester.  We take a quick break in Winchester for a soda -- 55 miles between controls is getting long at this point -- and D and I then lose B on the final 18 miles into the Strasburg control.  This stretch is fast.  Including the 10-15' break in Winchester we required about 3:15 for the 55 miles from Sharpsburg, MD to Strasburg, VA.

The Gunnar

Somewhere on leg three I notice how well the bike is performing.  A quick aside with regard to the Gunnar rebuild.

The new groupset is excellent.  It must be a combination of a recent tune job and new components (teeth not worn down, springs still wound tightly); the shifts are crisp and land right on target each time, even under load.  My only complaint is difficulty cross-chaining and hitting the 50-25 gearing, but with the ease of shifting chainrings it becomes a non-issue with a little experience.  I also found some difficulty shifting when my hands went numb later in the day from cold and fatigue, but I was able to find a work-around there too.

The tires and tubes (28C Grand Bois, Vittoria Latex) are either simply awesome or I am having an unexpected and unwarranted fitness moment.  I go with the former.  As evidence, despite my being no extraordinarily aerodynamic descender, I constantly am passing my companions on coasting descents, despite their carrying much heavier loads.  90 psi front and 95 psi rear made for fair comfort on all roads until the very end.

Conflating the tire quality is the new fork.  In theory the carbon fork should provide much of the comfort benefits that I attribute to the 28c tires with latex tubes.  The forks also lighten the bike by a full pound vis-a-vis the former steel forks.

And the fit is perfect.  One change from the picture on the blog: I have aero-bars mounted with small risers above the drop bars.  So the drop bars are set down an inch or more from being level with the saddle, but the fore-arm pads are approximately level with the saddle.  The aero position is positively comfortable, and -- if my ease of handling winds relative to my riding companions is any indication -- is still sufficiently aero to warrant the name.

Night Riding

After chili and an orange soda at the RBA's house, we leave Strasburg (Mile 171) at 6 pm.  B arrives shortly before we leave and tags along.  There is enough light that D and I ride without lights and still wearing sunglasses almost until we reach the final mediate control in Marshall.  We are slowing down, but still making good time; that 42 miles takes 2:45.  By Marshall (Mile 213) I feel dazed.

This stretch includes 10-15 miles of fast empty pavement along Highway 55.  It is a treat.  I sit in the aerobars and feel the tires float over the smooth tarmac as the light fades and traffic zips by on the parallel but separate Interstate 66.  We also enjoy a low concrete bridge over the Shenandoah River, which in wetter springs is famous for stopping riders in their tracks as water rushes over the bridge, forcing them to wade their bikes through 100 yards of ankle-deep water in the dark -- a misstep carries the penalties of a dunk and a swim downriver!  This spring is drier and we three encounter the bridge while it is still daylight.


We turned on the taillights not long after leaving the Strasburg control.  No problems there.  We add the headlights with 5 miles remaining to Marshall.  Shortly thereafter my DiNotte flutters, telling me the batteries are low.  Huh? After only 1.5 hours of use on the low setting this morning?  I have left it plugged in (though off) all day, and the instructions teach the unit drains the batteries even when not on.   I had assumed that meant over the course of several days, not several hours!  The lesson must be to unplug the lamp when not in use, which is not terribly inconvenient but does require attention.

A second lesson is to wear a head-lamp.  Reading the cue sheet in the dark is a challenge and I find myself stopping and lifting the bike to read street signs.  Too, a head-lamp is a convenient way to carry a backup light.  I had known this already, but the headlamp is easily left out of frenzied preparations.  I think on the 600K in four weeks I will wear the second DiNotte on the helmet.  If the lights are for the most part used singly, a third set of batteries (one set of back-ups) should be ample.

Heading Home

The last leg is from Marshall to Leesburg, via the posh hamlets of Middleburg and Purcelville.  We leave Marshall at around 9:05 with 38 miles to go.  I report to P__ that I am pretty sure we will make it by midnight; we have been riding far too strong to fall off to below 13 mph.  We see B again, having lost him on the prior leg; he rolls into the 7-11 as we are preparing to leave.

The importance of the lighting becomes clear after Marshall.  Having replaced the batteries with some from 7-11, I am now afraid to run the lamp on anything greater than low power.  That makes reading the road difficult and impossible at high speed, so bumps I would have avoided now add to the discomfort from 16 hours of riding.  It is also impossible on any but the best road surfaces to ride on the aero-bars (because of the sacrifice in control), so I lose the only remaining comfortable position.  Sitting in the saddle and resting on the handlebars is simply miserable.  And without a head-lamp I sacrifice my cue-sheet reading.

We have a route-finding SNAFU leaving the control, placing us several miles off-route before we discover the problem.  Even as we violate a fundamental tenant of randonneuring by taking the shortest path to re-joining the route (instead of retracing to our point of deviation) we add 8 miles to the ride, plus at least 15 minutes sorting out our predicament.

This part should be fun.  It is dark, quiet, not too cold; the cars are gone and the ones we do encounter are exceedingly polite; the climbing is not trivial but nothing I haven't done before; and the nutrition strategy has worked remarkably well on this ride, so my energy is up.  But by Purcelville, which we reach just before midnight, the miles and the stress of riding blind in the dark leave me wanting to be done; because of the detour, we have covered a very hilly 400K already.  I celebrate by sitting against propane tanks while D calls for backup.

So on the home stretch we violate a second randonnuering tenet.  C__, who is meeting us at the end with pizza, beer, and a ride back to DC, pulls in behind us and lights the road ahead for the remaining 5 miles.  So that's what it would have been like riding with two lamps on full power!  And that's what it would be like in a sanctioned ultra-cycling event!

Partly it is better roads but largely it is the better light; I can once again sit in the aero bars and we roll at a good pace to the end.  The computer has been out since well before Purcelville so I do not know our finishing pace, but the final _46_ miles from Marshall, including route-finding and my rest at Purcelville, took a not-shameful 3:40.  (My guess as to riding pace has been about on.)  We are back at the Days Inn at 12:45 am.  C, true to her word, offered beer and pizza.  I choose to take my carbohydrates purely in liquid form.

Random Thoughts

Packing light worked well.  I carried only a frame bag, basic tools, and three tubes.

I will pack less food the next time and buy more on the way.  I will fill the extra space with (1) slightly more warm clothes, and (2) two extra battery carriages loaded with charged batteries for the DiNotte lights.

No changes needed to bike fit.  Maybe a saddle change; I will see how the new WTB saddle performs and may experiment with it on the 600K.

Electronics:  D had an external battery charger that was billed as sufficient to fuel two iPads to full charge.  That could easily keep the Garmin running for a full 1200K.  No small sacrifice on size and weight, however, so I wonder if there is a smaller unit.  The Garmin may have to stay behind on longer rides.

Riding with others is interesting.  We genuinely enjoyed riding with B.  Later in the day our pace was faster, but he could catch up again at the controls.  Leaving Marshall we probably should have waited a minute or two longer for him to refuel and to join us (we already knew B did not dally at the controls).  If nothing else, a third eye on the cue sheet might have saved us 8 extra miles and in the dark our speed advantage would be minimized.  B did end up finishing before us, despite riding somewhat more slowly earlier in the day.  There is also the unsurprising phenomenon that after spending nearly 16 hours riding with somebody you sort of feel awkward leaving that person behind.

Thursday, May 2, 2013

Habanero Rebuild Complete!

It was about 4 weeks ago that Max and I were in a hotel room in Bishop, CA discussing some shifting issues I was having with my Habanero. After he pointed out a fantastic deal on a complete Ultegra 6700 groupset, I took the plunge and ordered. A few conversations later, and I also ordered some Neuvation R28SLW wide-rim wheels.

Fast forward a couple weeks and I'm back at home with boxes full of components, tires, tubes, bar tape,  and so forth. I gave the bike an initial weighing (10.7 kg, or 23.6 pounds), and then started disassembling and cleaning. The entire tale is below. The short version is that the only parts that stayed the same are the frame, seatpost, and seat.

Before Rebuild
After Rebuild

Replacing the Neuvation carbon wheels with the alloy R28SLWs didn't save much (about 100 grams), but was still a no-brainer move. I'm too heavy for carbon wheels, I don't race, and alloy wheels just handle better. The new wheels let me run a 700x32 Schwalbe Ultremo which may improve ride quality, but at the very least look comically awesome. I was able to sell the carbon wheels on craigslist for $400.

There is an impending 70g savings coming soon from replacing the Neuvation skewers with some Ti skewers. But Jenson USA is taking their time shipping.

The old drivetrain was a real mish-mash of components; Ultegra 6500 and 6600, 105 5600, FSA, and KMC. They're all compatible and worked well enough together, but theoretically replacing them all with 6700 will give me better shifting and will certainly look nicer. Oddly Merlin Cycles has the Dura-Ace 7901 chain for $4 less than Ultegra, so I did swap in the DA chain. All told the drivetrain changes saved about 250 grams,

Initially I planned on retaining my stem, bars, and fork, and headset. But my headset is notchy, which is annoying, so I splurged on a new Chris King headset. The 1" stem cap didn't fit my 1 1/8" stem, so I had to buy a new stem cap. While I was doing that, I vainly ordered a Ti crown race, which saves something like 6 grams over the stainless steel one. Dropping a Ti crown race on al Aluminum fork seemed criminal, so I picked up a Ritchey Carbon Comp fork (one of only 2 1" carbon forks I could find, the other being Nashbar). With a new silver headset, my old black stem started looking tacky, so I bought a new Ritchey Classic stem. But that has a 31.8mm clamp, so I had to buy new handlebars too.

While making gratuitous Ti upgrades, I decided to replace my aging and ugly aluminum bottle cages with King Cage Ti cages. That saved a remarkable 62 grams, and it's a huge aesthetic improvement too.

Savings on the frame come to 900 grams, but in fairness 540 of those are from leaving the aerobars off for now. Other than aerobars, the fork, stem, and bottle cages were the biggest contributors.

Somewhere North of a bunch of money later, the bike is now a svelte 8.53 kg, or about 18 pounds, 13 oz. Given that I didn't make an effort to go super-light (and the frame isn't that light to begin with), I'm pleased with the result. Adding the computer and a couple other odds and ends will probably bump it above 19 pounds.

But the goal here, as Max gently reminded me a few times, has never been to save weight. The goal is to give the bike that "new bike" smell. Get it back into a shape where it is shifting smoothly, rolling fast, and cornering like it is on rails.

Shakedown Ride
The part of any build that I enjoy most is the shakedown ride. That's where everything comes together (hopefully). In the case of the Habanero rebuild, the shakedown ride was about a mile, because as it turned out the 700x32 tires that fit perfectly in the garage when under-inflated, didn't quite fit when fully inflated. I've since replaced them with an Avocet Fasgrip 700x28 in front, and 700x25 in back.

With that swap, the bike is riding very nicely. Shifting is generally pretty good, though I'll be adjusting it a bit as the cables seat more fully. As Max and I have discussed several times, a major periodic overhaul seems to be money well spent. After 9 years of riding, I'd say the Habanero was due!

Wednesday, May 1, 2013

A funny line

I was googling the "Flatland 1200K," trying to find this ride we talked about several years ago in Holland.  I didn't find it, but I did find a ride report from the Gold Rush -- must have been 2009 -- from a Dutch cyclist.  He began by describing the ride up and over the dam:

The ride started out flat, with about the only hill being to get up onto a levee. Just like Holland. There was a slight wind and I found myself shuffling behind stronger riders to find a windshield. The levee reminded me of American Pie by Don McClean. "took my chevy to the levee" - now I understand what he mean't. All these years I thought a levee was a petrol(gas) station.