The history behind it
To date I have not attempted to form such a complex structure as a single sided swingarm. The key to it's success is acheiving the right balance of lateral stiffness to weight ratio. The most logical approach is to increase the section size as much as possible as this has a higher influence to resist deflection than sheet thickness itself.
Of as much importance is the internal structural webbing to prevent the larger flatter sections deflecing under stress.
Overall the swingarm needs to form through restrictions around the chain and wheel so there are areas of compromise. To help this situation I took the challenge of construting the swingarm
using the least possible amount of panels; allowing each setion to then be formed 3 dimensionally and gain stiffness.
The old one
Back into 2012 when the project first started I welded a bracing structure on top of the swingarm which was a token gesture to stiffness. I cut a standard swingarm open to examine the casting thicknesses and sections throughout. I was
quite surprised to see how much material there was particularly around the shock linkage mount and just how little there was towards the hub clamp. I could see some good scope for improvement. The ‘jaws’ as I call them that grasp the hub are equally poorly supported and could see the source of
the flex. An improvement would be to make one side of the jaw short and rigid whilst reducing the length of the opposing jaw as much as possible in an effort to reduce both lateral and torsional flex.
The new one
Most of the panels were quite easy to roll on my beautiful set of 1950's sheet rollers.
Each panel was mocked up in cardboard, plastercene and occasionally bodyfiller to begin with before being cut out of sheet 5083. The formed sections were made using an array of different press tools I constructed for the job.
The steel press tools are the best but most expensive and time consuming to make. With a dowel in each corner to guide them together they worked first time.
For the larger top and bottom formed panels I experimented with large concrete moulds which failed spectaculary the first few times; I used my uncle's Manitou as the press which suprisingly wasn't upto it; a 16 tonne machine lifting itself off the ground with my futile 2 foot square concrete press staring back at it without even flinching.
I resorted to some crude agricutural engineering that day; a bucket of spuds dropped at speed with said Manitou. The Manitou just took off quicker, and then the concrete started to crack.
The second revision was a much better success; 6 M16 threaded bars welded to spreader plates only just managed to form the bottom panel with me working overtime with a gas torch on the plate and a 2 foot breaker bar on the clamping nuts. Hot forming was the only way to go with such large complex forms. If it were a production part for a motorcycle you would see a press tool the size of your house to do these parts.
The new swingarm is constructed with 5083 sheet throughout purely because it avoids the
need for post weld heat treatment. Pivot mounts and hub clamp were machined from rolled solid. Cast blocks are easy to obtain but rolled offers slightly greater grain alignment but are harder to come by. Sheet sections are 3mm thick in the H111
condition, rolled and pressed using my old flypress and rollers and then welded with 5356 rods (uniformly pre heated to around 150C). The old metalwork equipment is great; solid cast iron. I am adamant in years to come equipment like this will be priceless.
To finish off the fabrication I made a cushless drive sprocket carrier along with a new large diameter eccentric hub from 7075 aluminium. I would have prefered to use a 7000 series aluminium sheet for the construction of the rear arm;
such as 7020. It offers greater tensile strength per relative thickness of section used in comparison to 5083, however 7020 is hard to come by and expensive in particular sizes. It's also likely to require heat treatment and post machining processes
to compensate for the distortion. Mine will be slightly heavier as a result due to the use of thicker section sizes. It will be favourable to the bikes weight distribution at least, which is currently somewhat biased the front: Cue the carbon front fork project!