The first version was a disaster. As I described in the first article the base was a Halfords carrier. This consisted of an aluminium extrusion with an pressed steel assembly at one end This is seen at the top of the image, Figure 1.
Figure 1 |
Figure 2 |
Figure3 |
This clamp arrangement only pulls the extrusion onto the roof bar and does not provide and torsional resistance (probably by design).
Figure 3a (of a standard design) hopefully shows this as well:
Figure 3a |
My modification involved removing the arms, and at the other end of the rack mounting the Saris fork clamp, also seen clearly in Figure 1. The fork clamp is mounted to the rack using a piece of aluminium plate which is bolted to a piece of aluminium box section, which finally is bolted to the rack. I chose aluminium over steel because it is much softer and hence easier to work and because it would not need painting or any other corrosion resistance treatment which steel would.
My pieces of aluminium, and their assembly, was fine. I managed to drill all the holes in the right places and bolt it together (using stainless steel screws and nyloc nuts). It resulted in a neat job with no movement. The problem was however that the bike could easily put twist into the aluminium extrusion, as this had no torsional stiffness. With the original design the twisting force transmitted by the arms from the bike "wobbling" is constrained by the wider steel mounting assembly. I had not considered this torsional loading in my original design.
Given that I now realised the torsion problem from the bike to the extrusion, I had 2 options to resolve this:
1 - to move the fork clamp the the other end of the rack so that the torsion was constrained by the mounting plate, or
2 - beef up the fork clamp mounting so that it would be bolted or at least clamped to the roof bar itself
I decided on option 2, partially because I came up with a modification to the mark 1 version rather than starting completely again.
The finished Mark 2 version ended up looking like this, Figures 4 and 5:
Figure 4 |
Figure5 |
Here is a picture of it in use:
I am fully confidant of it now - I don't bother using any rope any more - it holds the bike as well as the Thule model. The run to university in September 2013 saw a high speed run (we were late!) in high winds for 170 miles on the M25, M4 & M5 with absolutely no problems.
Further Improvements
There is still an issue with this design. I now have forced the spacing of the roof bars to match the spacing of my carrier design. Most cycle carriers have one end of the rack fixed with the other end floating and I have removed this capability. I think the fix needs to be to mount a slightly modified design at the of the rack. This would leave the original clamp for the opposite end but as it has no need to constrain any torsion this won't be an issue, and it would allow adjustable roof bar spacing.
The final improvement is that some security is required - both for the bike to the carrier and the carrier to the car. On the Thule there is a for locking the clamp that holds the rack to the roof bars and the fork clamp is also secured with a lock. For my design the rack security to the roof bars is by the fact that 8 thumb bolts need to be undone, so not secure but time consuming. There is no way to lock the bike to the carrier. However several months later and purely by chance I saw this product in Halfords:
This is lockable using a padlock though the holes. Its more expensive but a viable alternative to the Saris model I have used .
No comments:
Post a Comment