Monday, April 06, 2020

Plan 2020 - Clincher Rear Wheel, aka "The DT Wheel"

One of the many critical things I need for training is a clincher fixed gear rear wheel. For the uninitiated, it's not just a matter of "converting a wheel". A track rear wheel is narrower, has a different hub, and is a specialized piece of equipment. With all the wheels I have, one thing I don't have is a knock-about fixed gear training wheel.

I do have a nice hub, so using that as a starting point, I built myself a nice training wheel.

Wheel building kit

Back in the day I built a lot of wheels, but wheels back then were simpler things. Nowadays the rims are taller, have a lot of nooks and crannies where you can lose a spoke nipple or something, and usually have lower spoke counts. Weight doesn't count as much as aero, and there's a whole new category of rims for disc brakes. With disc brakes the rim isn't the brake so the rim can be lighter and, most importantly (ha!), it can have decals on the side.

I decided to get a rim with decals on the side. I went with DT R460 rims just because they looked cool, they were clinchers, and they were relatively inexpensive.

Before my bike shop days ended I had gathered some wheel building supplies through a DT sponsored, in-shop course. Can you believe someone came to the shop and taught us how to build wheels?

Part of the kit included the pen-like gizmo in my kit above, a device designed to hold a spoke nipple so you could thread it onto a spoke deep inside an aero wheel.

I never used it until I build this wheel.

I also used spoke nipple washers (never used until this build). They're the cone shaped washers in the bag in the top right of the bin.

The crooked screwdriver is made to screw on nipples quickly. It's like a speed handle screwdriver.

The old wheel, before I tore it apart. Note cog on hub.

Built wheel, still to be tensioned.

Building the wheel took forever, probably about 2.5 -3 hours. Building it ended up being a calming experience, meditative experience. For the first time in a while I was lost in the process, focussed, and only realized how much time it was taking when I realized that I had to keep an eye out so I wouldn't be late picking up Junior (this was before the shelter at home thing). I left the wheel untensioned so there's a bit left to go, but otherwise it's all set.

Cog. I didn't remove it, built the wheel with it on the hub.

It's a bit funny, we used to joke about using cogs so small you could change a spoke without removing the freewheel. Well, this was my race wheel from 2009, and I just left the cog on through the entire tear down and rebuild process.

The other side of the hub.

It may not be pristine but for now it'll do.

Friday, April 03, 2020

Plan 2020 - CycleOps 300 Pro

One of the most critical things in the 500m time trial is the start. Actually, it's the most critical thing, and it makes sense. The goal with the 500m time trial is to cover the distance as fast as possible. You can do one of two things to do that. You can go faster or you can get to speed quicker.

The problem with going faster is that you're working against aerodynamic drag, which increases exponentially. It will take an incredible amount of power to increase speed meaningfully, and at what I presume is the goal speed, about 40 mph, you're covering about 60 feet per second. There isn't a lot of time to go faster.

With limitations to top speed, it makes sense that the rider should accelerate to top speed as fast as possible.

And that's what all the big racers have been doing for the last 20 years. The top speeds are about the same so they work to get up to speed faster. That means every iota of power has to go into the pedals, not into moving the bike around. So no real rocking of the bike

I first saw the unusual starting technique with Chris Hoy in 2002. I wondered about his technique since there was no rocking of the bike initially. It was more like he was on a spin bike and pushing down. Ends up that is exactly it - you basically lock your leg, push down as hard as you can (which means pulling up on the bars), and you end up putting down massive power and saving your quads for later. Here he is in 2002 (and incidentally he does the 500m in 31 seconds, which would win Nationals for me by a wide margin). And here is a crazy fast sea level kilo in 2019 (go to 21:00). That guy does a 30 second 500m. It seems to me that a 34 second 500m will put me in the ball park for Nationals.

Everyone has acknowledged that starts are critical. Therefore riders practice starts all the time, going to the track if possible. There's a unique resistance with standing starts that simply not cannot be replicated on a trainer - it's high resistance at the start due to the rider/bike inertia, but then it turns into a speed thing, turning the pedals fast with high power. You can't replicate that with the light flywheels on trainers, the 6 lbs or even 12 lbs ones. There are whole training days for starts, using different gears, doing start efforts (50-100m), etc.

For me getting to a track is not possible so I needed to find an alternative. Based on some intranet help from fellow trackies, the low buck way to replicate starts is to get a high inertia wheel. There's a bazillion dollar machine out there but a cheap and usable alternative is the CycleOps 300 Pro. New they were about $1500-2500. I saw them at Interbike in 2009, but I remember thinking, "why would I want one?". I liked the 400 for its ability to adjust power (so you could record a ride and then "replay" it at home, or set power test steps, etc).

Picture from the post linked above.
The 400 (above) is a smart trainer, i.e. it can adjust resistance.
The 300 is not a smart trainer but uses the same frame and wheel.

The 300 was initially sold as a fixed gear bike but later it went freewheel, I think for mass market appeal. When I contacted Saris (the manufacturer) about conversion kits to fixed gear I was told that the 45 lbs wheel still has the fixed gear threads, they just put a one speed freewheel on. I'd just need to remove the freewheel and install a fixed cog and a lockring.

With that settled I decided to look for a 300. They're available locally for about $500, I saw them for less outside the area.

The biggest problem was fit - I vaguely recalled being unimpressed with the bar drop when I sat on one in Vegas. When I looked at the local one I measured what the drop would be from saddle to bars. On my bike it's 14.5 cm. On a 300? About zero. When I measured where the "top tube" intersected the "head tube" it was about where I'd want the stem. With a huge amount of vertical adjustment, I figured that if I modified the frame so that the top of the head tube was as low as possible then I'd be good.

I bought the 300.

And got to work.

First off, the thing is really heavy. Shipping weight is 140 lbs, actual weight about 135 lbs. I struggled to get it out of the Suburban, and it noticeably weighed down the rear of the vehicle. In the process of moving it around I managed to break the plastic mount for the computer. On the good side I figured the weight would help anchor the bike, and the flywheel would really make a difference in replicating high inertia standing starts.

Stock head tube. Very tall.
Black insert in tube.

The black insert is a slippery plastic, sort of like cable housing liner. It's tough, resistant to rubbing forces, and flexible. It makes it possible to have a less than perfect head tube and still clamp the tube tightly. On a regular frame the seat post and stem have to be perfect to within 0.1 mm. With the sleeve, not so much - the tube is probably misshapen by a solid 2-3 mm from top to bottom, with significant distortion around the welds.

The clamp (with the big lever) looked to me like it can be removed. This was my big gamble. I figured they would make the frame modular, with a one piece cast piece for the clamp. No reason to do fancy brazing and lug work on this thing, so all the raised bits would probably be molded into the cast piece.

My request for help on Facebook.
The clamp assembly is circled in red.
The yellow vertical line is my lowering goal.

Note the spot weld bits on the clamp assemble. There are two on each side. When I removed the sleeve there were no weld marks at the top, only a lip to keep the clamp from sliding down too much. I gambled that the clamp assembly was slipped over the cap, a couple welding tacks put in place to hold it, and done. If they used more to hold it I'd be in trouble.

Starting cutting at work.

I didn't have a cutting wheel so I brought the 300 to work and borrowed one. When I started cutting I was afraid of ruining my $500 investment. Plus I'd never cut so much metal at once. I was timid, unsure, and eventually stopped.

Did the cut at home.
I started cutting the welds holding on the clamp assembly.

At home I was a bit less timid. I didn't have to clean up before close, I could work at my own pace (no work interruptions), etc. I generally work well like this although sometimes I end up veering down the wrong path. It's how I did the suspension and quarter panels of the Civic, built the inside of the trailer, fixed the Sentra, got the now-red frame repainted, and it would be how I approached modding the 300.

Cut head tube in front, the remaining frame in the rear.
I needed to cut out the rectangle for the stem clamp.

I tried to mirror the rectangle opening for the clamp wedge. I also notched the tube for the alignment bumps in the clamp assembly.

Notching the frame for the alignment notches.

One thing that I didn't anticipate was just how low I cut the head tube. I knew I cut it lower than the clamp height, and I planned on cutting down the clamp, but it was close. I underestimated how much the clamp overlapped the head tube - it went much further down than I thought it would.

Trial fitting the clamp assembly while holding the cut bit of the head tube.
Red thing is a flashlight to look inside the tubes.

At this point the clamp was about 5-8mm too tall - the lip that was supposed to sit on the top of the head tube (visible on the left side of the clamp) was 5-8mm away.

The pieces in my hand.

You can see how tall the clamp assembly is at the back, opposite the yellow lever. It's over 10mm tall, and I needed it much shorter.

The clamp assembly cut down.
I did smooth it out after.

I cut the back of the clamp assembly almost to the top - I'm just short of the cosmetic trim lip of the assembly.

You can also see the lip around the edge of the clamp assembly. There are notches cut out for the plastic sleeve. The springs spread the wedges in the clamp bit - there's another wedge that is not in the picture, and that one is the one that presses against the tube.

Close up of the finished product.
Note how low the head tube is now - the "stem" protrudes below the frame now.
I had to remove a stop that kept the "stem" from protruding down below.

I can't weld so I planned on using a metal epoxy to resecure the clamp assembly to the head tube. However, I actually did not do the last step in the process. I'll do it later but the metal epoxy I have was dried (I used it on the Honda quarter panels about 10 years ago!) and the clamp is holding fine right now without any adhesive.

The finished frame bit. 
The wheel still needs to be converted to a fixed gear.

The major mod is done now. I still have to remove the BMX freewheel off the wheel and install a fixed cog and lockring. I want to put different pedals on it also - the Keo Max pedals are too easy to slip out of, which is why I don't have them on my road bikes (I've unclipped multiple times during sprints). I may use my old SPD-Rs (the kind that don't unclip easily), or I'll use the one Look pedal that doesn't release accidentally, the original Look Keo Carbons that I still have in my pedals bin.

I gave it a test launch last night. I was in my PJs, no socks, and went to do some lifting before bed. I slipped on my shoes barefoot, adjusted the saddle, and did a Chris Hoy imitation. Junior has watched a bunch of starts with me and even he knows the drill.

Beep at 30 seconds.

Do what you need. Look around. Look down. Wiggle your hands on the bars. Think of the upcoming effort.

Beep at 10 seconds.

Focus. Hands still now. Prepare to send it.

Beep every second from 5 seconds.

At 2 seconds you lean a bit forward.

At 1 second you throw your butt way back, and start launching forward so that...

At "GO!" your pelvis is thrusting forward so it's over your front hub, your left (track tilts to left so you start with left so you don't take yourself out) leg straightens, and you push down as hard as you can while you pull with your arms as hard as you can. It's like doing a dead lift, max power, hold your breath, just get the bike going.

And I almost flipped over the bars as the rear of the 300 came up off the ground.


So I have some more mods to work on, to keep the 300 planted - apparently 135 lbs isn't distributed properly for standing starts.

And then I can work on some full power starts.