Monday 31 August 2015

New Bike #2

I like epicyclic gearing, that's probably no great surprise to you given the original content of this blog. I have and ride examples of most of the exotic stuff that Sturmey Archer made in their heyday. These are the ones that collectors pay silly money for on Ebay. As an aside it tickles me that the ASC trigger featured on the Sturmey Archer heritage web page is one of my reproductions rather than an original. That's one of my early ones, note how the escutcheon doesn't quite fit the trigger body and that the detail is not an engraving. The internals were a perfect match with a genuine ASC trigger though. I sold these with full disclosure but it seems that subsequent sales weren't quite so transparent...

Let me know if you need one, I have a few left over. 

Given this love of the tiny whirring gears, I was hugely excited to learn of a new 14 speed epicyclic hub gear about to hit the market. Subsequently I became an early adopter of the Rohloff Speedhub, buying #1045 new in 1999. Rohloff have made over 200,000 since then. It has always lived on my mountain bike and has covered many miles with no real issues. The only time it let me down was when it lost its oil on the descent of Mount Ventoux in France. Now that was exciting, I can tell you. The oil seeped out of the hub and down the spokes and onto the rim. V brakes don't work well when liberally oiled. and 25 km up a big hill is not a good place to have it happen. The very early hubs had faulty seals that were replaced under warranty. I don't really do much actual mountain biking since I can't afford to come off with my back issues. Where I live we have lots of gravel roads that would be ideal for a new type of bike, a cross between a road bike and a mountain bike. Naively, I thought I'd invented the genre but apparently it's already a thing and they're called Monster Cross bikes. Not a cyclocross bike, that's different and very traditional with fairly skinny wheels often with tubular tyres. I mean a bike with all the advances of recent mountain bike technology but with a road oriented riding position complete with drop handlebars. And the Rohloff I already have.

Paragon Machine Works make Rohloff specific dropouts.

I wanted disk brakes as the technology has settled down since the early versions. My 1996 Ibis tandem was made with early Hope cable disks which whilst adequate simply do not measure up to the modern equivalents. My lovely wife and I took this tandem to France to watch the '98 Tour, the year of the Festina scandal. We rode it up and down the Col du Tourmalet and as a result I think I can say with some authority that modern disks are an improvement. We still have the tandem but these days it's mostly myself and my giant son that set Strava KOMs on it. He's an awesome stoker, he weighs nothing yet is unnecessarily tall and has a huge power to weight ratio. I ask for more power on the uphills and the machine lunges forward. One day, real soon now, he'll be thrashing me on two wheels. But I digress. The first job was to send off the Rohloff to the local NZ agent to get it converted to a disk brake version, this is an easy swap that simply requires a new endplate to mount the disk on and a new external shifterbox. Chris emailed me a few days later to say that it has arrived but with damage despite being well wrapped. The result was a very stressful 6 month battle with the courier company to get them to admit liability and pay for the damage. I'm pleased to say that I was successful but the original #1045 hub shell was damaged beyond repair, it's something that still irks me. The upside is that the replacement hubshell is bright red (naturally) rather than the original silver.

Mmmm. Shiny red...

The Rohloff doesn't have a great choice of shifters for drop handlebars, there are a few twist type shifters available but most solutions look like the most awful bodge ups. The new Cinq5 shifters looked just the ticket so ordered a set from Pure Sports.

Right lever for lower gears, left lever for higher gears. shifts two at a time.

Nice clean installation.

In the meantime I hunted around and engaged the services of a local NZ frame builder. I know, I know, I could have built one myself but I enjoy and admire craftsmanship from other builders too. I needed to provide all my critical dimensions and the family and I had a good laugh getting the measurements one evening. Having had a fair few frames made for me over the years, I've always known that I have relatively long thighs. Being such a freak means that I generally can't get my seat far enough back on off the peg frames. The brief was simple, full width 29er wheels, disk brakes, Rohloff hub, single chainring, 80mm suspension forks and obviously it has to be red.

When the frame arrived, I was like a dog with diphallia.

I'd been collecting all the parts required over the previous months and spent a happy day assembling it as follows:

Frame: CCC Columbus Xcr 29er
Fork: DT Swiss XMM 15mm axle
Headset: Hope mix 'n' match
Stem: Thompson 110mm -17 degrees
Handlebars: Salsa Cowbell 46cm
Drop brake levers: Tektro R200A
Auxiliary brake levers: Tektro
Seatpost: BBB carbon
Saddle: Brooks Cambium C15 (the new skinny rubber one)
Bottom bracket: Hope stainless steel
Chainset: 170mm Middleburn RS8-X type with Rohloff spider
Chainring: 38T Middleburn Mono
Chain: SRAM PC 870
Brake calipers: Avid BB7 road SL
Brake rotor front: 180mm Avid HS1
Brake rotor rear: Rohloff 160mm
Front hub: Hope Evo Pro 2
Rear hub Rohloff 16T
Rear skewer: 90's Hope titanium
Gear shifters: Cinq5
Rims: Stans ZTR Crest 29er
Spokes: DT black stainless double butted
Tyres: Continental Race King 2.2 tubeless using Stan's NoTubes system.
Pedals: old Shimano spds

I would have willingly given one of my testicles for a bike like this when I was young.

So far I'm loving the tubeless tyres. OMG how good are they? The bike is not trying to be a mountain bike but is reasonably capable on pretty harsh terrain. I rode the Contact Epic on it this year and whilst not well suited to the second half of the course it was OK. For its intended purpose however it excels, the bike disappears beneath me and rides as well on gravel as on seal. It has opened up lots of local unsealed roads including some spectacular climbing within a short ride of my house. Just this last weekend, I rode a 50km fully offroad loop with 900m of climbing right from my front door. I'm such a lucky boy.

The little green tag on the Brooks gives me access to hundreds of kilometers of empty forestry
roads at the weekends. I highly recommend you join if you live in or near North Canterbury.


I'm afraid that the jury is out on the Cinq5 shifters, I really, really want to like them, but they've let me down on a couple of big rides since I've had them. I'll write a full report on them later.

[Edit - The Cinq5 Shift:R system has now been replaced by a Gebla Rohbox, Take a look at my long term comparison of both systems]


In other news, I've knackered my back again. I bent down to pick up an allen key from the workshop floor and popped one of my existing prolapsed disks again. I spent yesterday in hospital on morphine which was where I wrote this so apologies if it's a little garbled. I hope that normal service will be resumed shortly. My goodness me, morphine is a powerful drug...

Monday 17 August 2015

New Bike #1

I've been trying to clear a few projects lately. You know how it is when you have just too many things you have to do to make progress on the stuff you want to do. I also dislike having bikes in pieces. I'm increasingly finding that I'm forgetting details and mislaying or worse losing components. For these reasons, I've either been giving away or selling off some of the herd over the last year or so. I'm still not done but I'm making progress. The remainder is getting reassembled into complete bikes that I want to keep. The first example is this "BSA" from the late 1920s.

A BSA bicycle, yesterday.

The BSA is in quotes because although the frame is entirely made from BSA components, I have no idea who actually made it. The date is also a best estimate based upon components and a trawl of BSA catalogs. This approach is difficult here in NZ where local manufacturers had boxes of stock that was used until it ran out. In many cases long after the parts were manufactured. NZ was slow to accept new trends in bicycle fashion during the first few decades of the last century and lagged the rest of the developed world by a some way. When the first NZ & Australian riders took part in the Tour de France in 1928, the Europeans found it amusing that they were going to compete on technology that was 20 years out of date. The Australasian team upgraded as soon as possible to try and level the playing field. Even so the Kiwi in the team, Harry Watson, wrote home to complain about the reliability of the European bikes not being as good as locally made NZ ones. The bikes they took over to France would have been similar to this one.

I'm not sure why there was this lag. I've mentioned the likelihood of using up hoarded stocks of lugs etc. There is also the fact that NZ was very isolated from the rest of the world which would have meant an ignorance about emerging technology. But I suspect a large reason was shear bloody mindedness. "That bike was good enough for your granddad son, so it's good enough for you. Harden up". This continued right into the middle part of the century. The concept of a racing bike here was very different from in Europe. Steel rims were the norm although wood and rarely aluminium were used. Wide, heavy tyres were used as racing routinely took place on gravel, most roads being unpaved. Single fixed gear was normal and braking rudimentary or non existent. None of this detracts from the achievements of the riders though. These were genuine hard men, the equal to any in the world on two wheels. And of course since everybody was on the same equipment it didn't matter anyway. Genuine old racing bikes are not that common here. It seems that many were converted to roadster duties after retiring from racing. It's fun to spot these lurking beneath the modifications of time. Many were worn out and discarded, but even so very few seem to survive.

This one is something of a bitsa. Most components are BSA and are from the same period although the whole thing has been built up from spare parts. The frame is made entirely from BSA fittings. It has BSA lugs, brake and chainstay bridges, fork crown, steerer and dropouts, bottom bracket, rear forkends for the BSA cam system and a steering lock. I have no idea what the tubing is but is of the older 'D' profile chainstay, seatstay and fork profile. The steerer and crown are nickel plated and have numbers stamped which may or may not indicate 1928. The frame and forks are lightweight.

The wheels are 28 x 1 3/8" (ETRTO 642, 700A), a size that was obsolete in England by 1910-1915 but was still popular in the colonies right up to the sixties. Tyres are not interchangeable with the common 28 x 1 1/2" (ETRTO 635, 700B). New tyres are still available in a few specialist shops although I have a small stash of original Dunlops. These are the same tyres that were raced on back in the day, heavy block tread and all. The rims are wooden and I think French. The rear rim is the one that badly split a couple of years ago. I repaired the rim with modern adhesives and rebuilt the wheel, however the rim was unable to hold riding pressure in the tyre and it split again when sitting out in the sun.

The pressure was only about 40 psi at the time this happened. 

The new split was in a different place from the first one. I think that this rim will need to be for show only. The hubs are both BSA, the rear with the cams to adjust the tension. The rear hub also has my new cones made from through hardened 4130. I was going to run these as a trial to see how well they handle the real world before I start remaking all the worn parts on the facile. I'm going to have to learn how to make wooden rims in the near future if I want this bike to be ride-able on the road. More on that later.

The handlebars are early nickel plated racing bars. They are too narrow for me but I can't nip down the shop and exchange them so they will have to do. The saddle is a modern Brooks Pro that I bought new in 1995. I will change this when I find a suitable replacement. I'll put the original  gallows style seatpost back on at the same time. It doesn't work with this saddle, pitching it too far forward.

There are no brakes other than the fixed gear and no provision for them either. The frame is scruffy but has original paint and pinstripes so I've simply cleaned it up and oiled it. The chainset is an early nickel plated Durax, I've never seen a nickel one before and I didn't think the company was that old. Does anybody know when they started making these hollow octagonal fluted cranks? The bottom bracket cups and axle are Bayliss Wiley, the axle being a very elegant hollow example.

Tightly bound while the glue dries.


In other news, we've been tuning the trebuchet. The original design brief for these devices was to hurl the heaviest possible projectile a set distance. That distance being how far you had positioned yourself away from the target and out of range of troublesome archers etc. Our goal is subtly different, we want to throw a projectile as far as possible. Over the last few months the trebuchet has undergone a transformation. I've been studying the theory regarding the efficiency of these things (it's quite low by the way) and there are several things to pay attention to.

The weight needs to drop as vertically as possible to put as much energy as possible into the projectile. There are several ways to achieve this. The counterweight can be hung on a connecting rod as in our original design or the trebuchet can roll on wheels to allow the weight to rock the machine back and forth as it falls. It turns out that the length of the connecting rod is important, a longer rod allowing more correcting movement. In order to achieve this Kung Fu Pete came round and we moved the weight to hang horizontally.

Thomas, helping out.

At the same time I machined some better bushes for either end of the con rod. The whole machine was still being bodily moved with each throw, so it got some wheels added.

The wheels run on short tracks and the weight now drops more or less straight down.

As the weight reaches the bottom of the fall it has no option but to turn and swing horizontally. To get maximum energy into the projectile, you need to tune the release point to coincide with this direction change which will usually be when the throwing arm is vertical. There are two ways to do this, modify the angle of the sling hook and alter the length of the sling. These two variables affect each other so they need to be tuned together. A longer sling will release later as will a bigger angle to the release hook. It's a bit of a juggling act. Using a video camera and replaying in slow motion is an extremely useful tool to tune these variables. I imagine it's also what the medieval builders did.

The design of the throwing arm is worth some careful consideration. Ideally, it should be neutrally balanced so that the centre of gravity is coincident with the pivot point. This is hard to do and means that the weight side has to be bulky and the throwing side light and slender. I've consistently used a 1:4 ratio between the lengths of these with the current arm being 0.75m and 3m resp. The overall weight of the arm should be as light as possible to reduce the moment of inertia, again care needs to be taken that it doesn't get too weak. Since the tip of the arm on the throwing side is travelling extremely quickly, it makes sense to give it some useful aerodynamic qualities by making it wider in the vertical plane and narrower in the horizontal plane. I've also given it a rough aerofoil shape. I think I'm probably pushing my luck with the latest arm but it seems to happily throw 1 kg rocks without breaking or flexing too badly.

Swish it goes, very loudly.

The net result of all this work means that we've extended the range to just over 175 metres (~575 feet). There is a lot of variation between throws which is caused by having variable projectiles. Heavier stones describe a higher trajectory and  lighter stones a flatter one. I'm going to find a set of river boulders of equal size and weight and as round as possible. I can then fine tune the sling length and hook for maximum range using the video technique. Maximum range doesn't happen at a release angle of 45 degrees as you may think, it's more like 40 degrees. Air resistance having a large effect at the higher speeds means that releasing too high makes the projectile fall short. Mr. Middleton has badgered me into doing some accurate measurements in terms of range, height and duration. Watch this space.

We're pretty much at the end of the road in terms of more development of this design. That's probably not a bad thing though as there's now enough range to punch a hole through my nearest neighbour's roof if I get a foul shot. I really don't want that to happen. Loading and shooting this thing is also reasonably frightening, the forces and speed involved being potentially lethal. It has been an interesting exercise and it has certainly got people talking...

Tuesday 19 May 2015

Myford Super 7

I still don't have a proper workshop but I do have a new lathe. New being a relative term of course. The Myford Super 7 that I mentioned months ago has now been completely refurbished. It's a sad and somewhat ironic fact that now that the original Myford company is no longer in business, spare parts are much easier to get hold of. Sadly, I think that this is a temporary situation as the inventory will run dry in due course. I've replaced everything that was worn. Bushes, bearings, feed screws, you name it. If it was worn I got a new one. The lathe has been dismantled into its component pieces, cleaned, painted and reassembled. The only component that didn't get taken apart was the clutch assembly as it was fine and I didn't need to touch it. The bed and slides are in very good condition. The front vertical shear being the only part that required attention. I made a jig to run up and down the bed and scrape the front shear. It was located on the two rear shears and adjusted with its own jib strip. This was very successful and has removed most of the taper. I didn't want to take too much material off but the result is very satisfactory. I had trouble with the original 0.75hp motor but I found another one and got it professionally wired to the Dewhurst switch. I'm pretty confident that I could have done it myself but knowing when to not do stuff myself is a useful skill that I'm slowly developing. I've only made a few little bits and pieces on it so far but I'm looking forward to learning the new lathe on the next project. So far I'm loving the gearbox, the higher speeds, the extra rigidity, the long cross slide, the spindle lock, the really easy back gear engagement, the 360 degree top slide, the auto eject tail stock. I could go on but you get the idea.

A long bed Super 7, this morning.

When I sold the ML7 back to Mr. Middleton I included all the original spares it came with including the chucks as they all had matching numbers and my OCD meant that they must stay together. The chucks that came with this lathe weren't good enough for my standards so I bought replacements from a supplier in the UK. They are original Bernerd chucks with almost no use. I'm keeping the old 3 jaw as a roughing chuck but I've sold the old 4 jaw. None of the numbers matched anyway. I also bought a larger 9" face plate and I treated myself to a genuine Dicksons QC tool post. I've been using one of the modern copies that are sold on any number of engineering websites. I was never happy with it, the standard of accuracy just wasn't there. Of the 16 QC tool holders I had only 3 would lock down tightly, I was unable to use the remainder. The Genuine Dicksons model is a pleasure to use and all of the tool holders can now be used for the first time. The difference in quality between the two is like night and day.

I also discovered that the Myford cabinet that the lathe came on isn't. A genuine Myford, I mean. It's far too flimsy, almost useless and it would be impossible to level. So I've cut the top off and bolted it down to my work bench instead. I like having my lathe up quite high so I don't have to bend over too much and the raising blocks have made this easy to achieve.

I've been riding a lot but not a huge amount on the facile. I do take it out occasionally but it has been having some issues. I snapped the first spoke a few months ago and I had a really hard time getting the broken thread stub out of the flange. I was concerned at the time I made the spokes that stainless steel was not a good material for this particular spoke application. The spoke had fatigued and snapped flush with the flange. I tried to cut a slot in to the stub with my Dremel and use a screw driver to extract it but it was just too tight. In the end I drilled a 1.5mm hole down the centre of the 2.4mm spoke and then filed the remaining threads away. It too me two days and I needed to completely strip the drive down and remove the front wheel to do it.

I suspect that all remaining spokes are similarly fatigued so I'm reluctant to break another. 

During the strip down I noticed wear on the bearing surfaces, particularly those in the lever pivots and the live axle. In all cases the wear appears to have started as the result of an impact that has dented the bearing surface, this has then got progressively worse. This means that the nitrided 4140 is simply not the right approach. I'm going to have to replace quite a lot of components but I'll have to do some research first. I've made a few sets of BSA cones as my first trial of new techniques to harden bearing surfaces. They are machined from 4140 but have been through hardened instead of nitrided. I'll see how they perform before I invest too much time re manufacturing lots of parts.

A pair of BSA cones in hiding. Notice the cams for chain tension adjustment.

The hub will need replacing as it has 6 bearing surfaces. Given the spoke issue, I'm going to take the opportunity to rebuild both wheels as tangentially spoked. When I first stared this project I wasn't aware that Ellis & Co. made tangentially spoked models. I've since found an excellent description from 1888 and located two surviving faciles with tangential wheels, one is geared and the other ungeared. The spokes are in a most unusual arrangement with the front hub having just 6 or 7 holes in each flange depending on the wheel size and spoke count. Each hole housing 4 spokes. I'll provide full details in due course. I was hoping to be doing my big ride about now but I'll put it off until the bike is repaired. I'm reasonably fit at the moment which is a great irony.

So the first real project on the new lathe is going to be making a rim roller. The rims currently on the bike aren't the right profile and for various other reasons I want them off my bike. I've been collecting parts for the rim roller for a little while and I hope to be able to start it soon. Both my giant son and Kung Fu Pete have expressed a desire to own a racing penny farthing so I'm going to need to be able to make some really nice, light rims in various sections. Both of them being unnecessarily tall means that I can make some huge bikes, probably 60" or so. Kung Fu Pete has also taken on the project of getting rubber tyring in various sizes made locally. I am extremely grateful for this as I simply don't have time. He's had a sample made that shows promise. We're learning all about exciting things like the shore hardness of rubber and what extruding dies look like. This will all be the topic of future blog entries.


In other news, we've had both sets of parents here over the summer. My lovely wife's parents weren't too happy when I had a 9" face plate, a Dicksons QC tool post and a pair of chucks delivered to their address in the UK to bring over for me. What's 15 kg between family members for goodness sake? My parents only left a short time ago. I needed to go out to the garage to count the Super 7 when they left as my Dad has always wanted one. Being a long bed model, he wasn't able to fit it into his hand luggage though.

In more other news, I finally got around to making a decent size trebuchet as a garden feature for the kids to play with.

Garden furniture.

We're fortunate in that we live on a hill and have a good half a kilometer of our land between us and the public road. It's a work in progress and it still needs more tuning. Adjustments to the sling length, pouch shape, counterweight, connecting rod length and the angle of the release hook all effect the trajectory of the missile. We're probably at the upper limit of what the frame will support with a 100 kg counterweight but by tuning we should be able to hurl a 1 kg rock out to 200m (~650 feet). We can reliably throw to 150 m at the moment. As my lovely wife commented though, with the tuning and the addition of weight, it's gone from a fun toy to a genuinely lethal weapon. I threw three rocks down the field this morning before work just because I can. There's something about it that I find immensely satisfying.

My giant son made a video of it in action you can see it here.