Speaking of Spokes

This week I have been repeating myself. I try hard not to do it but as I get older I find that it's an undeniable truth. I think I'll start paying the kids to poke me when they catch me doing it in public.

But I digress and that's not what I mean anyway. This week I have been making the spokes which involves a huge amount of repetition, none of it is hard or arduous but it's easy for the mind to wander when doing something for the 100th time. A wandering mind and a large hammer is not necessarily a good combination I find.

The 13 gauge spokes are made from 2.4mm (0.094") wire. The threads are rolled on which raises the crest of the thread and increases the final diameter to 0.099" (#3 x 56 UNF).
The donor wire I have used is 308 stainless steel TIG welding rods. I'm somewhat reluctant to use stainless as it isn't an original material in use in the 1880's, However in my defence, I have scoured NZ for high tensile wire in the correct size and failed miserably. It is very easy to get 2.5mm wire but this is too large for the thread roller and the threads produced are then too big. That 0.1mm does make a big difference! If I do find a source in the future, I may revisit this. for reference, most modern stainless spokes are made from 304 stainless.

The first step is to put a head onto all the blank pieces of wire. I haven't done this for a few years and it took a bit of practise to get good at producing a decent shaped mushroom. The wire is clamped between the two halves of a die with a few mm extending. This is struck with the hammer to form a head into a slight countersink in the top surface of the die.

Now having previous experience of this I was able to make a suitable die to overcome some of the issues. The first one is that the wire has to gripped very tightly. I made my die from a length of angle iron I found in my magic cupboard. By clamping a sheet of paper between the two sides of the die before drilling the hole at size, the final hole will be a very tight fit when the paper is removed. The hole should be as long as you can drill it to increase the friction of the clamp. The second problem is that as you are hammering away, the die will get deformed. I've found I can get about 20 spokes headed in a mild steel die before it needs cleaning up. It is possible to make a die from tool steel and get it hardened to prevent this deformation, these do last considerably longer but can also shatter.

With no tool steel in the cupboard, I've solved this problem 

by making a throwaway die with a series of holes, 

when one gets too munted, start to use the next one along.

A few mm extend out of the die...

...which is then clouted with a big hammer.

I found I was unable to cold head the stainless steel, I needed to heat up the wire extending out of the die with a torch before hitting it with the hammer. This is not ideal as the resulting heads will be weaker. Time will tell.

The raw spoke heads out of the first die.

The heads at this point are all over the place, functional but very irregular and just a bit crap. Mr Middleton would have been fine with them but my OCD (according to my lovely wife) made me improve them.

I made another die from some tougher 4140 steel, this die is not split but has a slightly larger hole and a countersink (actually both sides were used since it still got munted from the hammering).

The second die.

I then poked the spoke through the hole and held the head up to the 

side of the bench grinder. By twiddling the spoke around in the hole 

I was able to make the heads round (rounder anyway). 

Then resting the die on top of the larger vice the heads are hammered down 

into the countersink, producing a far more even, consistent head. 

Next step is to calculate the length of the spokes using a bit of trig. I did this twice on separate evenings to check that I'd got it right before I did any cutting.

Finally roll the 13 Gauge threads onto the other end and we're all done.

I didn't think I had a 13G thread roller but my magic cupboard produced the goods once again.

The finished spokes.

The rims have finally been painted, but with the colder weather, the paint is not drying as well as I'd like. I'm itching to build the wheels up but I need to be patient.

In other news, my lovely wife is to have some fairly major surgery next week. Obviously, I am incapable of looking after her and the children unassisted so my Mother-in-Law is coming for a month to help out. We are all very excited about this as you can imagine. I have been given a very long list of "stuff which must be done". I suspect that if I haven't got the wheels built by then the paint will be fully cured by the time I do get around to it. We will see.

Le race is on Saturday. I realised last night that drinking beer and watching the TV does not constitute "training". I'll do 20km tomorrow night after work, that'll do.

Alchemy

I am in possession of a secret formula of such value that I hesitate to mention it in a public place such as this. The formula is handwritten and ancient and has been passed down through the generations. The origins of the author are lost in time although I believe that he was a chemist who lived far north of here. The formula would have been lost were it not for the sharp eyes of a friend of a friend of mine as he sorted through the dead chemist's notes.

The formula gives the keeper the ability to turn base metals into nickel plated base metals.

I would love to be able to say that I've been using this secret recipe for years to do all my nickel plating, but the truth is I was only given it a few months ago and I haven't tried it yet. I've been using a commercially available kit for over a decade with excellent results.

This past week I've had an interesting discussion with Tim, an English friend of mine. Tim and I share the view that modern, bright shiny nickel is not an appropriate finish on antique cycles. The nickel plate of old was more mellow yellow than the hard bright finishes available today. This is what originally led me to try my own nickel plating. Old nickel is often quite dull, more of a satin finish. I note with interest that Frosts now sell a dull nickel kit on the above link. If you live in England it may be worth giving this a try. Some of the dullness and colour of old nickel is partly due to the years of oxidisation and environmental pollution. I've experimented over the years with chemically accelerated ageing of fresh nickel. I own a copy of  "The Colouring, Bronzing and Patination of Metals" by Richard Hughes. This huge volume was my own starting point in this adventure. My most recent experiments have involved using the natural geothermal hot springs here in NZ. Some of them will turn any nickel coins in your pockets black. Geothermal hot springs are marked on the topographical maps with a red cross, further experiments will follow...

As an aside, it is an easy matter to 'turn' chrome plate into nickel plate. Chrome plating is always applied on top of nickel plate (which may or may not, in turn, be on top of copper plate). The chrome can be stripped from the nickel with no harm to the nickel by dissolving the chrome in hydrochloric acid (also known as spirit of salts or muriatic acid). I usually dilute my acid (add acid to water, not the other way around) and leave the components for about 30 minutes, then dunk into a strong solution of baking soda (an alkali) to neutralise the acid and prevent rusting of exposed threads etc. Rinse with fresh water and you are done. by the way, HCl is truly horrible stuff, do this at your own risk and use lots of PPE.

So this week I've been polishing and plating all the bits made so far.

Polished, ready for degreasing.

Polishing on a high speed rotary buffer is hot, dirty and dangerous. The potential to loose fingers is high as is the potential to have your exquisitely made components flung randomly around the workshop at high speed. You need to be careful, very careful. And alone with no distractions.

Cleanliness and preparation is everything when plating, the actual process is just chemistry.

Note the dull finish, I've spent ages trying to perfect this.

Plating adds a very thin layer to the base metal, usually not enough to effect clearance except where tight clearances are already designed in. Threads usually need a quick clean out after plating.

Soft soldering the flanges onto both hubs, 

note the three threaded wires installed to ensure 

that the flanges are timed correctly relative to each other. 

The temperatures used for soldering with lead are much lower than will effect the temper of the hard hub centres so this is safe to do.

The completed hubs.

The rear hub is familiar as being similar to a modern front hub. The front fork legs need to be sprung apart to install the front wheel when it is built, being 4130 this is possible without fear of bending or breaking. Assembly is then as described before. 

This is the first time that the front hub bearings have been assembled 

with the flanges in place and you can now just see the labyrinth seal 

formed by the recess in the flange and the inner bearing race.

In this photo I've installed a pair of threaded wires 

of the same length as the spokes to check that 

I have the spoke angle in the hubs correctly calculated.

Next week, I'll start work on the hollow rims.

Spoke Flanges part 2

This week I've been drilling and tapping some holes, quite a lot of holes actually. Which in itself is an interesting question, how many spokes does a geared facile have? There doesn't seem to be an easy answer. I've conducted a survey on my own original sources which includes original photos, catalogues, adverts and machines that I know have original equipment. I've disregarded restored bicycles where I either don't know or suspect that the equipment is reproduction. Front wheels can have 50, 56 or 60 spokes with 50 being the most common, rear wheels can have 24, 28 or 30 with 30 being the most common. Also bear in mind that wheel sizes varied, with the front wheel offered in 36", 38" and 40" throughout the 5 years of manufacture, The rear wheel size varied with age, earlier bikes having a smaller rear wheel of 22", later bikes 24". I don't yet know if the rear wheel size varied with front wheel size. The machine I am copying has 40" front and 24" rear wheels with 60 spokes front and 30 spokes rear, this is a pneumatic bike though and Ellis & Co. may have felt that the new fangled tyres required a  stronger wheel. I have decided on 50 front and 30 rear, which is a common combination, it also has the advantage that both wheels have the same spoke spacing at the rim. Aesthetic reasons and all that....

So the next question is what gauge are the spokes? Fortunately this is handily stamped on the flanges.

60 x 13 gauge spokes. 

The remains of the original spokes measured 0.093" x 56tpi they are also (fortunately) straight gauge with no butting. I have made butted spokes before but it adds a considerable amount of time to a project. So all I need to do is get some 13 gauge wire of a decent tensile strength and roll some 13 gauge threads on one end and head the other end. But what thread do I use for the tapped holes? 13 gauge spokes are a very good fit in #3x56 UNF threaded holes. For the record 14 gauge spokes are #2x56 UNC.

The first job is to make a mandrel with a keyway to securely hold the flanges whilst I drill and tap them.

I used a chunk of aluminium just because I happened to have it lying around. 

The important bit is to not take the mandrel out of the chuck after it has been turned, this ensures that the flanges will be true to the register and the holes will all be centred on the flange with no wandering. The dividing head is invaluable for the drilling and tapping operations.

First centre drill the flanges, then drill the through holes and finally tap the threads.

The second flange requires a little thought since the spokes holes must be timed correctly relative to the first flange, spoke holes falling between opposite spoke holes and all that. I've been indexing at 1/25 of a circle for the holes, so If I advance 1/50 and then continue at 1/25 this will all work out.

The rear wheel flanges were easier when I realised that I didn't need to make a mandrel at all since these flanges are not keyed and the holes are aligned at the time of assembly.

First bore through for the cones and then cut the mounting recess 

and the inner profile. I've tried to catch the shower of chips 

that comes off when machining gunmetal, it really is spectacular.

At this point I can remove the chuck with the work still 

mounted and remount on the Myford dividing head and drill...

 ...and tap the holes as for the rear wheel.

Then remount in the lathe and cut the outer profile and part off.

Next week I'll polish and nickel plate all of the hub components made so far at which point I can then think about rims.

In other news, today is the last day of summer. We haven't had much of a summer at all really and the mornings are decidedly cooler and definately darker. The commute this morning was quite chilly.

My daughter celebrated her birthday on Sunday, she invited a gaggle of school friends to come around and practice make up and do some shouting and other such stuff. My son and I decided to go out for the evening and watch one of the star wars films in 3D which was wasome. He wants me to make a podracer in my shed next. I bet they don't use 13 gauge spokes.