Fork Blades Bend at last   

Having done loads of sums and a lot more thinking I decided that a more traditional pipe bender was needed. Rummaging around the garage I found some nice 1" thick oak left over from the window ledges in our lounge extension. This combined with an angle iron fence post bits of old bicycle frame tubing some nuts and bolts and other scrap material and one of the rollers from tube bender Mk1 yielded the following:

Tube bender Mk2 initial version

The oak former was made in two parts using a 1/2" radius cove router in my milling machine.  I should have taken some photos but didn't.  The technique was to rough out the shape using a jig saw before mounting one piece on a rotary table.  The 11.5" radius arc was then cleaned up using a plane parallel router before changing to the cove router to add the quarter circle profile using the cove router - several passes where required as the miller speed is considerably less than the router speed in order to avoid excessive vibration and noise..
The second piece was then processed in the same way - but the other side up and the two parts glued and screwed together.
Knowing that a high force would be required the lever arm was extended using the bike tube and further extension was also possible giving a good 6ft of leverage.
Alas even 6ft failed to be enough to cold form the tube so heating had to be brought into play.

Bending under way

A little heat gets the bending going

Although heating the tube got the bending on the move, the oak former did not take too kindly to the very high temperatures needed and it was necessary to keep putting out the fire!!!!

Burnt remains

Bend nearly there

Fortunately the former survived long enough to make two bends, but only just as can be seen from the burnt remains

 The loss of the correct profile on the second bend resulted in a small kink but I hope to dress this out.

A word on the choice of bend geometry

Deciding on the fork geometry has been quite a problem.  Without an identity for the machine I have decided to use the study guide machine (see possible match photo) as a baseline.  This machine has a 30" front wheel and a 28" rear.  The geometry of the replacement forks is clearly all wrong - they have negative trail which makes the machine behave really bad. So some experimentation was called for. I made up some bracketry to attach to the forks so that I could move the wheel around an try different sizes.

After trying out the original 28" wheel in various locations and getting some improvement I took a trip to see Ray and borrowed a 30" wheel from one of his machines and tried that out.  It had to be tested up hill as the the brake could not be fitted and the sprocket is a free wheel.  Nevertheless, it was found that the 30" wheel worked well and even made the bike look better.

The final choice was to have a 30" wheel with around 3" rake on the forks.  This was about the minimum required to give clearance on the pedals when steering.

Applying some artistic thought then led to aiming for a bend radius of around 17" along the fork centre line.

So how to achieve this. From Bobs Blogg I learnt that when ovalising the fork blades some straightening would occur.  I made and estimate of how much by assuming that the inner and outer curved lengths would remain unchanged when the fork section was changed from round to oval.  Additionally I also expected some spring back after initial bending.  This was difficult to estimate but from various sources something like 5 - 20 percent seemed likely
So to finish with a 17" radius and bend angle of 31degrees I estimated that I needed to start with an 11.5" radius and 45degrees.  Just in case I then made sure that I left sufficient straight length on either side of the bend to be able to make any adjustments to get the desired rake etc on the finished article

Back to the Blades

Having at last managed to bend the 1" dia tubing the next step is to change the profile using the ovaliser.  I did consider changing the crank to make it easier to turn but having replaced the broken a gear wheel with the last 45 tooth wheel from my Myford decided that keepin gthe smaller crank would reduce the risk of another failure. - the theory worked but instaed of breaking a gearwheel I managed to inflict a nasty spain on my thumb - Oh well some you win some you loose

 As noted by Bob the curved blades feed through the profiler with little problem.


The final result was two blades which were virtually idendical but unfortunately have a slight twist as they go from straight to curved and back again.  I have yet to work on that but think I should be able to straighten them with a bit of heat a torque.

Checking the final radius I am pleased to report that the target 17" radius was closely acheived.


Once again the wisdom of doing some sums rather than just suck and see is proven