- mentally note the number and location of each new feature; and
- mentally note into which pocket you place each precious item, lest one of them apparently go missing over a weekend, leaving one with rather more grey hairs than one had before.
November 04, 2009
Arriving UFO
October 25, 2009
Seasons change
July 20, 2009
Magnum Opus
July 13, 2009
One little victory
- My P-38 is now sitting with 1,925 miles under its wheels, and a pair of stealthy black wheel discs on the back for (probably incremental) aerodynamic gains. With the new(er) rear derailleur, everything on the bike is working absolutely beautifully.
- My Speedmachine is sitting at 1,844 miles, although I think there might be a missing 700 or so because I reset the computer by mistake once. I didn't make a note of it, unfortunately.
- Annie, being somewhat older but underused these days, is still topping the chart with 2,770 miles since about 2004, which when you think about it isn't very much really.
- Speedy lies dormant in the shadow of my motorbike. It's still hugely entertaining to ride, but I also still have plans to sell it.
- Mavic XC717 rim, laced 3-cross to a Hope XC front disc hub with DT Revolution spokes;
- Mavic XC717 rim, laced 3-cross to a Hope XC rear disc hub with DT Revolution spokes.
July 22, 2007
Open secrets
So I already had a ton of parts all ready to go, a pair of wheels I'd built at least a month earlier, and...no headset tools. Being the happy DIYer I am, I thought I'd probably just get hold of some steel tube in the right size and bang the parts in myself, rather than paying the bike shop to do it for me. A headset isn't quite like a cartridge bottom bracket where you screw it in and everything lines up because it's all one piece. The two headset bearing cups have to be exactly lined up with each other, and the bearing piece that fits onto the fork crown also has to be exactly right. I started out not being able to find any metal tubes that were the just right match for the crown race bearing, until I had a brainwave and pulled out the main seatpost from the little Helios. It was almost the right size! But it wasn't exactly the right size and my bearing started to go squint as I hammered it on, and then it stuck. Now that was a slight problem, because although it was steel, any squintness might permanently put it out of round. So I hunted and hunted all over again, and eventually I got silly and tried the steerer tube from the forks of the Rockhopper, which were still sitting in the garage having been turned into a makeshift wheel truing stand at some point. Well wouldn't you know, the "Avenger" sized steerer was an exact fit over the "Standard" steerer of the new fork! So I sawed it off and filed it all nice and square and...it wouldn't slide all the way on. The tube was butted inside! Ok, so I'll file out the insides a bit, I thought. Well good quality Tange cro-moly tubing is tough stuff! I resorted to using an internal grindstone on the electric drill, which seemed promising until the grindstone broke. No biggie, there's a spare one, so I had another go and was making progress. Then that grindstone broke off too. I had one more spare so I carried on and tried a different movement with the drill. By the end of the afternoon I finally had a drift tool that would work. Of course, the tool was three inches too short but I was able to use the Helios' seatpost as a secondary drift. A bit of preparation with a hammer and screwdriver straightened the bearing out a bit, and then it was all systems go. It was the tightest interference fit crown race bearing I've ever seen. But it went on eventually, to my relief.
After that I wasn't going to take any chances with pressing the bearing cups into the frame though, and my new headset press made light work of the task. It cost at least as much as a shop would charge to fit a headset, and since Annie needs a new headset at some time as well, it made sense to buy the tool. "Quality is remembered long after the price is forgotten", said the uncompromising Henry Royce.
The rest of the assembly went very easily, for the most part. I did have to file off some of the paint from the frame's dropouts so that the axles would actually fit. I had some gear cable housing spare, plenty for the build, but when I came to trim it to the right lengths, it was like trying to cut through armour plating. My wire cutters couldn't do it, my pliers couldn't do it, my electrical wire cutters couldn't do it, and I blunted a chisel trying as well. Come on! So I spoke to the guys in Edinburgh Bicycle's workshop and asked what cutters they were using, and decided in the end to buy the same. Wow, talk about power! They'll last me a lifetime too.
The Lightning P-38 owes its design to their short wheelbased X-2 streamliner originally, back in the early 1980s, and is a more rider-friendly evolution that hasn't really changed in 15 years. It was one of the fastest bikes on the track at one time, especially in its F-40 guise, until lowracers came onto the scene and wiped the floor. But like the Windcheetah or the Brompton or the Moulton spaceframe, the P-38 was fundamentally right from the start, and good designs only evolve over time. It's not as aerodynamic as a lowracer or even some of the highracers like the Challenge Seiran, it's not full of carbon fibre or jawdroppingly light, but it's acceptably light, made of proper 4130 grade steel and it's very stiff. The riding postion is also more closed than most. For that reason, it's still regarded as one of the best recumbents for hillclimbing, and I have to say, I spend a lot more of my time going up than down. My Speedmachine is no slouch on the whole, but hauling 13lbs less metal up a hill should be nice.
There's no suspension on the bike, but in line with recent trends away from super narrow tyres, I've put bigger tyres on it than I'd usually use; it has a mesh seat rather than solid fibreglass, and the padded seat base is actually a cantilever and flexes when you sit on it. There are no disc brakes, because the bike was never designed for them (discs put huge torsional forces around dropout areas) but the v-brakes I've selected are well recommended by the people throwing themselves down mountains. The mesh seat should let my back breathe much better when I'm riding, since a) I practically create my own weather when I ride, and b) "sweaty back syndrome" is well recognised for recumbent riders. I kept an eye on the weight of all the components, but I've not gone too light where it matters: I know from experience that I need strong wheels, and that means touring rims, good tyres and more than a handful of spokes. With my basic LED lights fitted, I weighed the bike at 27lbs, compared with Speedy at 36lbs and the Speedmachine at 40lbs.
This does mean that Speedy is going to be leaving me. I've been its custodian for four of its 13 years and it's brought me into contact with a lot of cool, likeminded people. Like my friend Liz, they're people who are genuinely a bit offbeat at a basic level, rather than people who like to run against the grain as an occasional diversion from normality.
My remaining task is to be able to ride a bike again, but fingers crossed, I'm making some progress.
July 06, 2007
Stationary traveller
I haven't written a whole lot recently, so I haven't mentioned my new wheel truing stand that I picked up for a song on eBay. It's made by Ultimate and resplendent in anodised red aluminium and black cast iron. Ok, it's no Hozan, but it's solid and being one-sided it makes the wheel that bit more accessible. The trueness indicators have spring loaded tool steel tips, so you can spin the wheel and see where it graunches, without worrying about cutting into the sidewall. Having built all my previous wheels using the bike frame as the guide, it's light years more convenient having a proper stand. I also splashed out and bought a proper ergonomic spoke key; my three-way one is versatile but really uncomfortable to use. Like my Park PRS-15 workstand, having the right tools for the job really does make a difference. My Look CX6 carbon pedals had been sitting in a box and doing nothing since January, and after all those knee problems I have no intention of using Looks again. BentMikey bought them from me for his Hurricane upgrade project and he likes them a lot. I've actually been very happy with my el cheapo Shimano SPD pedals, which are on both the Speedmachine and Annie the Blue Bike, but they're a wee bit heavy and seem remarkably rust-prone, so I'll be looking for stainless hardware in future.
Meanwhile...
The big project begins and I'm excited! So far I have in front of me the following yummy parts:
- FSA Carbon Pro Team Issue chainset (in hard-to-get 175mm)
- RPM ISIS bottom bracket
- Shimano XTR front derailleur [1]
- Shimano Deore XT rear derailleur
- Shimano Deore XT 9spd cassette
- Sram PC-951 chain
- Shimano Dura Ace bar end gear levers
- Shimano Deore LX SPD pedals
- Mavic A719 rim, laced to a Hope Mono hub with DT Swiss Revolution spokes
- Sun CR-18 rim, laced to a Hope Mono lightweight hub with plain gauge stainless spokes
- Kore lightweight stainless QRs
- Avid Single Digit 7 v-brake
- Avid Ultimate v-brake
- Shimano Ultegra cartridge headset
The frame...well the frame's still top secret because it hasn't arrived yet. None of that aluminium nonsense though. Steel is real, man.
[1] From the days when XTR was burnished blue and aluminium, and looked like jewellery for bikes. I think it's one of the loveliest components they ever made.
May 23, 2007
Speedmachine front suspension disassembly
Alex noted that the HP Velotechnik boys are very good at designing bikes and not nearly so good at writing instructions, and I agree. I have the instruction manual for the bike and for the fork it provides no diagrams and no real clue as to what might need to be adjusted, and why. It goes so far as to say:
The adjustment of the fork play requires special tools and skills. Please have your bicycle mechanic perform this maintenance work. The following section refers to the expert.
Many years ago, I was a paid bicycle mechanic. I also have an engineering masters, and I still found the instructions difficult. I have therefore created some illustrations [1], so that one might have a better idea of what lies within before doing any unbolting of things.
There are three main issues you might have with the fork: it can squeak for unknown reasons (as did Alex's, and Dave Larrington's), the movement can become stiff and reluctant, and it can start to wobble as might happen with a loose headset. Mine was wobbling a little, enough to be disconcerting when powering uphill with the front of the bike moving around.
How does the fork work? How does it adjust?

The fork is essentially very simple, but made slightly more complicated by having externally adjustable damping. The fork itself has a long steerer and it slides up and down within a larger tube. Sandwiched between the two tubes are a plastic sliding bearing at the bottom and a teflon [2] sliding bearing at the top. The larger tube rotates inside the two headset bearings, and protrudes at the top where the handlebar stem is attached. So that you can actually steer, the steerer tube has a set of splines which engage with the inside of the plastic bearing, and the plastic bearing is splined to fit inside the end of the larger tube. This way everything turns together.
Friction damping is provided by two flat doughnut-shaped pieces of foam. The larger one is sandwiched between the two tubes, and the smaller is pressed against the inside of the steerer tube. The damping is made adjustable by the black control cap at the very top of the larger tube; as you tighten the cap down, it compresses the larger piece of foam.
The coil spring in the fork sits inside the steerer tube. At the bottom end it sits on a tube and is retained by a washer and a large circlip, and at the top by the base of a small tube which is connected directly to the black cap. As the fork compresses over a bump, the steerer tube slides upwards through the larger tube, and the spring is compressed because the black cap stops it moving upwards and the circlip stops it moving downwards out of the fork. The small tube ends in a flat section where the spring butts up against, and when the fork is at maximum extension, the flat section butts up against a circlip at the top of the inner steerer. Got all that?
If you need to adjust the fork, you can (in theory) do it without removing it from the frame. I couldn't tell what I was really adjusting when I was trying to adjust-out the slight wobble, so I removed the fork from the frame and took it completely apart to find out. Once you've removed the stem and the front brake, the fork can be tapped out of the frame by using a soft mallet on the top of the steerer. Mine was quite tight the first time.
Once you've hammered the fork out of the frame, have a think through the X-ray view here and the following pictures. You won't be able to take the fork apart further unless you remove the spring first so that you can slide the inner steerer into the outer to access the small tube. And you won't be able to hammer the fork back into the frame properly unless you remove the spring first, because the suspension will do its job of absorbing shocks, and do it rather well. So for disassembly, the first thing you should do is remove the spring, and this can be done whether the fork is still mounted in the frame or not. There is a slight preload on the spring: a couple of millimeters, according to the book, so be careful when you remove the circlip from the fork crown. This isn't the time for bodgery, you'll want proper internal circlip pliers.
What's inside, and taking things apart
With the fork removed from the frame, it'll look like this:

Peel the rubber boot down from the top and you should see the splines, the locknut, and the adjusting nut which bears directly on the bottom edge of the plastic sliding bearing:

For reference, here are the components which are sandwiched between the inner and outer steerer tubes:

You can see that if the outer steerer is held tightly in its big bearings, the inner steerer can wobble if either or both of the plastic splines bearing or the teflon upper sliding bearing are worn. The lower bearing is slightly tapered externally, and by turning the adjusting nut to force the bearing upwards into the outer steerer tube, it contracts fractionally onto the splines of the inner steerer, thus taking up any wear. The locknut is then snugged down onto the adjusting nut. I found it impossible to make this adjustment with the fork still in the frame, and very difficult to make the adjustment even when I could see everything. One might use a very time-consuming trial-and-error approach by adjusting the bearing pressure, reassembling the fork in the frame and checking for movement. I set the nut pressure just short of when the inner steerer became stiff to slide up and down inside the plastic bearing. Personally, I think Cannondale's Headshok needle bearing approach might take out some of this guesswork. I used Park Tool's HCW-5 lockring spanner.
That all said, the inner and outer steerers won't come apart until you unscrew the damping control cap from the outer steerer and you then unscrew the top bolt from the small tube that will be visible having unscrewed the cap. You'll need pliers on the small tube to hold it. The cap will then be free and you can slide the outer steerer up and off the inner:
Ok, not much to see there, but that's the inner steerer exposed. It will be a bit greasy and any grit that gets into the splines won't do them any good. Without the spring in place, the small tube can drop out of the bottom of the inner steerer and you'll find attached to it a chamfered teflon piece [2,3] and the second foam doughnut. Here is what's inside the inner steerer before anything is taken apart:
Reassembly
In true Haynes manual fashion, installation is the reverse of removal. More or less. With the spring out of the fork, reassemble the inner and outer steerers with their components, and with the outer steerer in place, check the plastic sliding bearing for the correct tightness. If it's ok, hold the small tube with the pliers and reattach the damping control cap and screw it a few turns into the outer steerer. Remember that it is one of several components that takes the full load of the suspension, so screw it back in until all the thread is hidden. There will still be a few turns of adjustment available and you may not need it screwed all the way in. Slide the fork assembly back into the frame and it'll wobble around until the lower bearing begins to seat itself. You will then need a drift such as a 1/2 inch diameter pipe, 18 inches long or so. I used a piece of lawnmower handle. Put the drift up inside the inner steer and carefully, but firmly, hammer the fork back into the frame. It'll go in anything from 1/4 inch to 1/2 inch more, and you can tell when it's fully seated as the sound of the hammer hitting the drift will raise in pitch and the blows will suddenly feel solid.
You can then put the spring, the spring cap, the tube, the washer and the circlip back into the bottom of the fork. You might want to use the drift again to preload the spring while you put the circlip in place. With the wheel in the fork and the bike on the ground again (you were using a workstand, weren't you?) you can press on the frame to make sure the fork is seated, and snug the stem as far onto the steerer as it'll go. The upper headset bearing cover should be vaguely stiff to turn. This is my main criticism of HP Velotechnik's design; the whole assembly is largely the same as any Aheadset setup, but without the starfangled nut by which you apply the correct preload before clamping the stem. No starfangled nut, no preload - other than pushing on the frame. With the fork in place, bounce the bike a few times and re-check the movement. If it still wobbles, for example, I recommend jumping up and down and swearing, before taking it all apart again. You might get away with knocking the fork halfway out of the frame, until the bearing locknut is visible, but on my bike the bearings were already out of their seats at that point, and the fork wobbled in the frame.
Where did that bit go again?
For comparison purposes (and just because I was on a roll with Adobe Illustrator that day) here are two cutaway views of the fork:
[1] I made some sketches as I disassembled things. I only made one measurement, so the illustrations aren't completely to scale. They're near enough, though.
[2] Well, I'd like to think it's teflon. It might actually be Nylon 66 or something.
[3] I didn't sketch everything, however. Some of the internal components such as the pressure cone I drew from memory and so might not be quite right. I didn't take everything apart again, just to check.


