Recumbent seat build

Spring being here, and the option being perpetual grass-mowing, I have resurrected - and not for the first time, I am unkindly reminded - John's High Racer.

The seat frame, built back in May, looked all wrong and applying a ruler I discovered I had three inches clearance rather than two. Seats are hard to get for a recumbent and I don't mind having spares, so I thought I'd make him another one.

First was to cut the side members. I have acquired the use of an old steel-cutting bandsaw, and a splendid creature it is, saving a good deal of elbow pain. It has the option, with a lot of fiddling, to alter the vice angle so I can mitre tubes, but I am scared of altering vice angles in case I can't alter them back. It crossed my mind that - in an entirely hypothetical case that one happens to have a builder operating on the house - one could wait until the builder has gone home, and then hastily plug in his wood mitre-saw which has angles marked on it and is very accurate. - This is of course a hypothetical case and should not be taken to imply anything at all.

Anyway, what I have found is that the best angle of seat base to seat back is 57 degrees. This isn't a law of the Medes and Persians because people's backs differ, and 57 degrees for me may be 60 degrees for John. But it's a starting point and we have to start somewhere. If you want two tubes to meet at 57 degrees then they need to be cut at an angle of 61.5 degrees for a reason that I understood when I was thirteen but can't remember any more. And if you want two blocks of wood that will go in a vice and hold a bit of tube at 61.5 degrees then the bits of wood need to be cut at 28.5 degrees, which takes half a minute to set up in a hypothetical mitre saw and about five seconds to cut.

By a process we will not now discuss I possessed myself of two such blocks of wood and put a piece of tube in the bandsaw and found that when the vice was tightened, the blocks of wood slipped like wedges and didn't hold anything tight at all. A moment's thought, and I screwed a scrap of wood to the two of them with a single screw in each, so it would act as a pivot. Now tightening the vice resulted in the blocks jamming onto the tube, and everything was cut easily.

Linked angle blocks holding tube at exactly 28.5 degrees

Mitred tube welded at the corner isn't very strong, so I use a Lowing Joint which is a scrap of steel wended into a slot cut lengthwise into the ends of the tube.

Side member welded, and another ready to weld, with its internal Lowing Joint support

Now to the seat supports. After Mr. Knight's handlebar adventures I thought I'd show him how it's done and got a large spring with a ¾ inch hole down the middle, and whopped it into the vice and whopped the tube inside it and bent it smoothly into a curve and it took lots of force but about a minute to do and I don't know what all the fuss was about.

Bending...

Then I tried to get the tube to slide out of the spring, and it wouldn't. It was stuck. I thought it was stuck forever, but after about an hour of twisting I found that the bending process had moulded a series of indentations inside the curve that exactly matched the spring, so I tried to unscrew the tube out of the spring. This didn't work because as I unscrewed, so the spring tightened around the tube, gripping it harder. After about another hour I managed to screw the tube the other way and when finally the straight bit was inside the spring and the curved end of tube was outside, it slid out with only quite a lot of effort.

A slightly crinkled bend


Well I jolly didn't fancy my chances on the other end. I thought there'd be no hope at all of getting the spring off. So I did the Other Thing, which was heat it up red hot and bend it in the vice using a bit of solid poked inside the tube.

This resulted in something so imprecise that it needed re-heating and whacking with a hammer, and after a lot more whacking I had something which my children, when young, would have described as a Shape*. This, I thought, qualified me to be a professional outsourcing tube-bender, but I decided it might be easier in the long run to make some welded supports so that's what I did, and it took ages. But it wasn't as fraught as bending the stuff.

From the top: 
support bent to Mr. Knight's standard of excellence
welded replacement support
support prior to welding
two side members

Getting the ends of these supports to wrap neatly round the side members relied on crown-cuts again, it being easier to bend them to shape than accurately to file them to shape, and it gives lots of metal to weld, which my MIG welder likes. The side members were used as guides, laid on the floor exactly twelve inches apart.

Finally to welding them in place. Now while thinking about what one can accomplish easily with a mitre saw it occurred to me that if I cut a bit of wood exactly twelve inches long, and exactly square at the ends - which is easy to do with a mitre saw - I could use that as a spacer to make two wooden jigs with scraps of wood screwed exactly a foot apart.

And then, to jam the tubes in place, I'd need to cut two bits of wood exactly an inch-and-a-half shorter than a foot, and the two three-quarter-inch tubes would make up the difference.

However I'm not going to tell you exactly how I did this. But it worked perfectly.

So the next bit is the webbing, which I'm not looking forward to at all, because it takes forever to do.

*(After boiling a ping-pong ball to remove indentations, said ball became greatly flattened. Child: "Not a ball any more, is it?" Parent: "No. What is it?" Child: "It's a shape.")