Here is the second installment on frame making stuff:

What do you need to design a frame?

Computer aided drafting (CAD) programs, Finite Element Analysis (FEA) programs, and a masters in mechanical engineering are wonderful things to have, but aren't really needed to design a frame.

I drew my first 3 or 4 frames on a drawing board. A drafting machine makes it easier to repeat angles, but isn't required. For what I paid for my Mutoh drafting machine and drawing board 10 years ago you can now get a very nice CAD program (I presume you do have a PC, considering how you got this post). If drawing manually, plan on doing the drawing at 1/2 scale. This makes the drawing big enough for accuracy, but not so big as to be unwieldy.CAD is nice in that it provides very accurate drawings, and the ability to very quickly make multiple copies of the drawing with slight changes. For manual drawings, plan on drawing the fixed points and then doing the frame drawing on a tracing paper overlay.

Dimensions of parts: Accumulate as many components as possible before you start drawing. You need accurate dimensions on wheels, brakes, suspension components, the engine with carburettors, sprockets and chains. When using a drafting table, I took a side view photograph of the motor with carbs. If you put a 12" rule in the picture, you can then have the picture enlarged until the rule is exactly 6" long - 1/2 scale. Take the picture so that you can see the crank end, countershaft sprocket, and all motor mounts. If you must make a drawing of the motor, try to be as accurate as possible on the above points, and rough in the general outline of the motor.

Starting the drawing: Determine the basic dimensions of the chassis - wheelbase, rake and trail, wheel sizes and front fork dimensions (presuming you are going to use an old fashioned steering head style frame), front and rear suspension travel, engine width.

Wheelbase: longer = stable, shorter = manueverable. Do keep in mind that the shorter the wheelbase the greater effect the rider position has on the center of gravity, and the harder it is to move the c of g around. On many bikes, no matter how nose heavy they are, as soon as the rider is installed they become tail heavy. Also, if you aren't severely height challenged, a short wheelbase bike can be very difficult to fit on with any hope of avoiding cramps. Now draw the wheel assemblies in, remembering to include the rear wheel chain adjustment travel - I usually just put the wheel at the midpoint to start.

Rake and Trail: The trend is towards steeper rake and shorter trail. Current 250s are in the 20-23 degree rake range, and about 3 to 3.25" of trail. If you read Foale and Willoughby's book on chassis design, they did some experiments that indicate that even these numbers may be larger than needed. Pick what you like - I'm going to reserve my opinions on the "magic" numbers at this time. Once the rake and trail are determined, you can draw in the forks and triple clamp/steering stem, with the forks at full extension. This determines the position of the steering head. Next draw an outline of the front wheel where it would be at full bump. I prefer to keep the frame stationary, and let the wheels go up and down on the drawing

Engine ground clearance: modern slicks allow really good riders to exceed the theoretical maximum 45 degree lean angle (at which the centrifugal and gravitational forces are balanced - NOTE: this ignores the "grip" of tires). 55 degrees might be possible in a transient state. If you are cautious, as I am, you will want to make sure there is NO chance of the engine scraping the ground. Therefore I generally try to allow 50 degrees lean with both ends of the suspension bottomed. To determine this draw cross sections of the tires you will use, with the 50 degree line tangent to the tire. Position the engine inside the V formed by the lean boundaries (don't forget the engine placement side to side depends on the chain line). This will give you the minimum height of the engine at full bump. Measure the height of the countershaft centerline and then draw a C/L on the drawing at full ride height. Position the engine drawing on this C/L.

Engine fore/aft position: You now have the vertical position of the engine (horizontal was determined by the chain line). You should have an outline of the front wheel drawn at full bump. What I do is shove the engine as far forward as possible, leaving fender, fairing, and a bit of fork deflection/tire growth clearance with the front wheel. As mentioned earlier, installation of the rider usually makes the bike tail heavy, so a forward engine position is needed. This problem is worse with a light engine/bike than it is with a heavy engine/bike. If you are a demon rider, you may need to adjust engine position fore and aft or up and down to keep the rear wheel somewhere close to the ground under heavy braking. I think most of the rest of us will be better off with a heavy forward weight bias, so the rear end tends to step out first, and the front won't want to push as much. A low center of gravity will make the bike more manueverable.

Swing arm pivot position: The trend seems to be to have a degree of anti-squat built into the rear end. If you draw a line between the countershaft and wheel C/L, anti-squat will increase as the swing arm pivot rises above the line. Of course, the position of this line changes as the wheel goes up and down - darn. Also, as the pivot moves away from the C/L the change in chain tension increases - darn again. For the speeds I go, I usually just put the pivot on line when the rear wheel is at the midpoint of the suspension travel. If you run a rear sag of about 1/3 the suspension travel, you will usually have a bit of anti-squat in the normal ride zone. The LA area Computrak guys took a quick look at my Laverda and said the pivot position looked pretty good to them (disclaimer: I'm not associated with them, they didn't measure anything, don't blame them for any of this). Of course, when you look at the ATK tension eliminator and the Bimotas with concentric countershaft/swing arm pivots they seem to indicate that you don't want any antisquat. Go figure. The swing arm pivot should be as close to the countershaft sprocket as possible. Sometimes miscellaneous lumps of metal on the back of the engine case can be ground away to bring the swingarm closer.

Engines with integral swingarm pivots (Ducati, Rotax, Husky, etc) get the sprocket and pivot very close which is nice. One drawback with the dirt-oriented engines is that the pivot pin is often pretty puny. The Rotax has a 16mm pin, and the older Huskys had a 12mm pin. I like to use a 3/4" diameter hollow pivot pin, which saves weight and is stronger than a smaller solid bolt. When I built the Husky I put the pivot behind the motor. Unfortunately, the pivot mount was the only rear engine mount, and coming up with some way to grab the back of the motor while clearing the chain, etc was not an enjoyable experience. Also, I think that a full width swingarm pivot can easily be stronger than a forked pivot to go around the engine. It is also more difficult to keep the bearings in the forked pivot aligned when welding it all up.

Now that you have all of the vital components positioned, you play connect the dots with the frame members. Getting everything connected without putting a frame member through the middle of the motor or the rider, or having horrible offsets and bending moments in the tubes is done by making version after version of the frame until you finally stumble on the least egregious solution.

Next time - frame jigs and other stuff as I think of it.

Have fun,

Michael Moore

Euro Spares


Back to the top frame page
Back to the home page
© 1996-2005 Michael Moore, last update for this page 01 June 2005

For more information contact us by using