Sunday, 28 November 2010

By all that's Holy....

It's all very well to blithely say things along the lines of "drill some speed holes in it.", but that ignores the time and effort that goes into working out the positioning of the holes, and then measuring and marking out whatever scheme of spacing you arrive at.

Drilling some holes in an existing part, or a part that has to fabricated, but has a size and shape that is going to be defined by the job that it does, such as the new rear upper engine mounting plate that I've concocted, imposes some limitations, and provides a place to start if nothing else. Since I'd decided that I wanted speed holes in the gussets that I was going to add to the frame, and the size of the gussets in two of the places where I felt a little stiffness could be added wasn't critical. That being the case it made as much sense to size the gusset to suit the speed holes, as to take any other approach.

Because the main loop of the BMW frame is more or less a rectangle, it will be prone to lozenging. Since the front down tubes are attached to the top if the headstock and the top tube to the bottom of the headstock, what will tend to happen is that under braking when the bottom of the headstock wants to move backwards, the only thing stopping the top tube moving relative to the down tubes are the gussets from the down tubes to the headstock. Much like the Norton Featherbed frame which is said to have inspired BMW's engineers in the design of the Airhead frame, racers tend to adjust the steering head angle by the simple expedient of crashing the motorcycle gently into a wall, which is enough to "tweak" the frame and reduce the headstock angle, and in turn quicken the steering. Good isn't it?

I feel it's obvious that adding a gusset between the top tube and the down tubes will stop most of that from happening, but to my mind it also makes sense to box in the bottom of the factory headstock gussets as well, since that would stop them from flexing, which is what they'd have to do to allow the sort of motion I've been talking about.

The other area where I feel the frame is lacking is the swing arm pivot area. The anecdotal evidence would suggest that a beefier sub frame that was welded to the main loop would improve the handling. For that to be so, then the problem must lie not with the sub frame, but with the rigidity of the swing arm pivot area of the main loop. The tubular bracing should address some of that, but adding some gusseting to the area should not only enhance the stiffness, but provide yet another place to drill some speed holes.

The gusset that will run from the brace bar to the lower rail presented a slightly different design problem. The rear most edge of it had to be the shape that it had to be, but the leading edge wasn't quite so constrained, with the overall extent of the thing being more severely defined than for the headstock gussets. In essence, I traced the shape of the rear of the gusset onto some card, fiddled around with some hole layouts, and allowed the sequence of holes to define the leading edge of the gusset.

In the last picture, you can see three pairs of gussets. To the left are the swing arm pivot gussets I just described, the pair in the centre are the down tube/top tube gussets, and the last pair are to tie the cross tube beneath the swing arm to the lower frame rails. The point of the last pair being that they should contribute to reducing any tendency towards fore and aft movement in the swing arm pivot points.Especially if one side is heading fore, and the other aft....

That of course leaves the boxing plate for the bottom of the standard headstock gussets, which I haven't made yet, and may lead you to enquire just what it is I've been doing all week? Well, just making the headstock/top tube gussets consumed a day of my time, and the other day of this week that I could devote to things BMW was consumed in making the swing arm pivot area gussets. 

Aside from the entire thing taking for ever and a day to carry out, I'm quite pleased with the look of the gussetry I've made, and at the risk of belabouring the point, you are likely to spend more time looking at a project once it's completed than the amount of time that you are going to spend working on it, so rushing through things makes very little sense.

Talking of looking at motorcycles, I wonder if you noticed the curious thing about the headstock/top tube gussets? More on that later perhaps....

Sunday, 21 November 2010

Wisdom, the ultimate, and the infinite.

 If you've been on board this particular bus since the month of July, you may remember that I cut the original BMW brake pedal up and welded a chunk of steel to it so that it could act as an idler in the linkage between the brake pedal and the brake proper, serving to correct the pedal ratio and place the final operating rod in a position where it's arc of movement wouldn't conflict with the arc of movement of the brake and thus ensuring that the effect of suspension movement on the brake actuating mechanism was minimised to the point where it became negligible.

If you don't recollect the occasion, this picture should refresh your memory. While it all appears to work, with the wisdom conferred by hindsight I think it's fair to say that it's not the ultimate in infinite beauty. As you should be able to see in the picture to the right, taking at least one leaf out of Phil Irving's book (of which I own several, and so should you) I thought I'd add some lightness, though I settled for making it look a little more elegant rather than "simplicate" it, and in truth, I only added some lightness where it showed.

In my time I've seen seen quite a lot of motorcycle parts that have had lightness added through the ever popular medium of holes which have also had unsuspected quantities of weakness added that led in turn to them acquiring substantial degrees of brokeness. Drilling "speed holes" is largely a waste of time, in most cases you can easily engineer something of equal strength to, and far lighter than, the original, but it has to be said that drilling unnecessary holes in motorcycle parts practically screams "racing motorcycle".

Lest you think that all this is the lead up to another pathetic excuse for not progressing the work on the frame, I give you another picture that ought to dispel any such notion. Careful study of the afore said will show that it wasn't just the rear brake mechanism that caught a dose of lightness, as the crankcase is looking positively anorexic too. The other item in the illustration which is adorned with holes, is the new engine mount I've added. The thinking behind it is not entirely uninteresting I feel. It's fairly obvious that the frame has had some attempt at triangulating it in addition to the main tubes I bent up for it, but it's equally clear that the brace tubes aren't perfectly triangulated and therefore don't appear to offer the ultimate in stiffness.  Oddly, they probably do offer the ultimate in stiffness, what they don't offer is infinite stiffness.

Possibly you feel an explanation is in order? Then I'll attempt one. Considering the frame as a separate entity, it's subjected to assorted forces which are transmitted to it via the forks and the swing arm. If for a moment we simply assume the forks and the swing arm to be infinitely stiff, then the limit of the forces that can be applied to the frame is the limit of the forces that can be applied to the forks, and the swing arm. And those forces are applied by the tyres.

Which ought to make it obvious that the unrelenting pursuit of stiffness to resist forces much greater than those which the tyres can handle, is not only a waste of time, but adds weight. So the considered wisdom should be that the ultimate in stiffness is sufficient stiffness to resist all the forces likely to be applied this side of the ditch, while saving the weight that infinite stiffness would require...

With that in mind, I happily compromised the geometric purity of my additions to allow the fuel tank to fit over them without requiring modification.

I'm reasonably sure that the solution I've arrived at will require more force to deflect it from the true and righteous path than the skinny tyres are capable of generating. Sadly, BMW saw fit to mount the engine off of the lower rails, and my reinforcement of the existing structure does nothing for them, but at the same time makes it impossible to remove the engine. That whole pot of woes is further stirred by the engine being repositioned so that it is impossible to change the oil filter without removing the engine from the vicinity of the right hand lower frame rail.

Using the same logic as most other BMW modifiers, I'd elected to make the lower rail on that side removable which would not only facilitate removing the engine, but allow for removing the oil filter without taking the engine out of the frame as well. Sadly, it doesn't take Werner Von Braun, or any other rocket scientist, to realise that as the engine is held in the frame by mounting it to the lower rails, removing one of them is going to make for a fairly floppy engine. Not only that, but the engine experiences a torque reaction around the drive shaft, which causes the top of the engine to want to move to the right of the frame under hard acceleration.

Hence my top engine mount, which I freely admit to having copied in principle from somewhere German, although I forget where. As well as feeding the torque reaction into the additional bracing , the upper engine mount serves the purpose of supporting the engine while the right hand rail is removed for routine oil filter changes. Not to mention providing somewhere else to drill speed holes.

With everything as welded as it could get with the engine in the frame, I set about removing the engine from the frame. Somewhat unusually, I had to cut the rail off the frame to allow the engine to be removed. With the offending (lower right hand) chunk of frame rail removed, I was slightly alarmed to discover that it was what we in the trade like to refer to as "stupidly thick" in the wall department. So much so, that the Suzuki Bandit frame splices I hoped to employ wouldn't fit into the bore of the BMW frame tubes. After a quick visit to the lathe, the Suzuki splices slipped into the bore of the BMW tubing with only the lightest of hammer blows. As a job, perfectly doable with a file, should you wish to adopt a similar strategy at any point. Since the frame splices were something on the order of 65mm long when installed, I took the trouble to mark the frame with two lines around 65mm apart, before hacking the rail out of the frame, ensuring that the cut I made was between the marks that denoted the extent of the frame splice. Accordingly, once the chunk of rail was cut out, I trimmed the cut ends back to those marks, and reassembled everything with the inclusion of the frame splices.

Since I rambled on about artificial, self imposed dead lines in my last diatribe, it was at that point I decided to call it a day and head off to the wonderland that is Tesco in the run up to Christmas and lay in some provisions for the weekend, leaving the final fettling of the frame rail and the last of the welding for next week.


Sunday, 14 November 2010

Rush hour.

I'm writing this on Sunday evening and the smell of roasting crown of duck is wafting in from the kitchen. The reason that the smell of roasting duck is wafting around is that I didn't make any progress on the BMW. In TV land, when motorcycles are being built, there is always a deadline, and somehow that idea, along with "themes", seems to have attached itself to the subconcious of the viewing public like some variety of psychological leech.

I'm a little suspicious of self imposed dead lines for leisure activities. I'm working on the BMW because I quite enjoy the process of working on the BMW, so why should I be in a hurry to finish working on it? I can understand that some people may build motorcycles because they want other people to admire their handiwork, and that they're building motorcycles because they like showing off. Some people even seem to confuse having their handiwork admired with being admired themselves, I've noticed. In that case I suppose the need to have the motorcycle finished for this, that, or the other show, gathering or event where people congregate in large enough numbers to admire the finished product makes a certain sort of sense, and the inevitable, "we nearly didn't make the dead line..." anecdote is all part of that too.

I find enough stress in the course of a day to keep me well supplied in that department, so I'm not going to start applying it to my main leisure activity, which is building the BMW. As I've mentioned before, much of what I want to do is indistinguishable from what I have to do to an outside observer, but the main difference is that it's doing what I have to do that keeps a roof over what I want to do.

So, my point is, that at six thirty on Saturday evening, having done most of what I had to do, I decided that the BMW can wait until next week before I get on with the frame, and as a consequence of that decision, I made it to the supermarket in time to snag the last crown of duck they had on the shelf.

Quite what the moral of that is, I'm not sure, but I'll give it a little thought as I enjoy my duck, with roast new potatoes and garlic, washed down with what promises to be a half reasonable claret.

Tuesday, 9 November 2010

...and just coming into the frame....

I've made a start on the frame at last. Having mentally sorted through various options, I've decided on welded in bracing with a drop out lower rail for the purposes of engine removal, and more importantly, oil filter replacement. The problem with that is that the engine is only attached to the frame with two studs, and those would have to be loosened to remove the frame rail, which would leave the engine hanging on two studs through the left hand rail. I rather suspect that wouldn't do the studs much good. To remedy that I thought I'd utilise an idea I saw somewhere else and add a mounting plate from the top two gearbox mounting studs to the frame bracing rails.
As I've mentioned before, I want the motorcycle to look like a BMW that's off to the races, and not like a racing motorcycle that happens to have a BMW engine in it. With that in mind, I bent up a couple of tubes that would run under the fuel tank and then drop down to meet up with the swinging arm pivot once they were clear of the carburettors and their air filters. Bending the tube didn't present much of a problem as my principle tube bender is a serious industrial piece of kit intended for series production work

I used one and a quarter inch 14 gauge CFS 3 tube for the braces because the front down tubes are that diameter. However, the brace tubes also meet up with the tubes that run from the swinging arm pivot up to the frame's top tube, and those tubes are one and an eighth of an inch in diameter. Not only that, but they're ovalised where they meet the swinging arm bosses, narrowing their width even further. This was all going to make the joint with the brace tube quite interesting, even without the three quarters of an inch tube that runs just above the swinging arm bosses getting in on the act.

Which is why, although I have a serious tube bender, I don't have a tube notcher. It took me about half an hour to cut the mitres by hand using a hacksaw and a selection of files. My theory is that if you cut tube joints by hand you tend to develop a "feel" for the job and even oddball examples like this one become much less trouble, and cutting simple ninety degree mitres takes so little time that setting up a tube notcher would probably take longer.

With the brace tubes shaped, I cut and mitred a cross brace that will eventually support the top mountings for the torque plate from some more one and a quarter inch tube, and after rummaging around a little I found some offcuts of one and an eighth inch tube to tie the tops of the brace tubes to the rear of the engine cradle.

The idea is that there'll be another inch and a quarter cross tube between the brace tubes at (or close to) the points where they're tied to the rear of the engine cradle. From either end of that cross tube another pair of tube will run up to meet the top tube, at (or very near) the point where the brace tube from the front of the frame meets it.

Aside from that, I want a pair of gussets tying each of the down tubes to the top tube where they cross, a cross tube under the frame where the original footrests mounted, box in the front of the factory headstock gussets, and a gusset arranged so that there's something other than the swinging arm boss tying the top and bottom of the frame together.

While I haven't perhaps cut as much tube as I would have liked, I have gone from a vague idea of what I was after to a fairly concrete plan, albeit one that exists only in my head. It should all be fine, as long as I can remember it all....