Trials Australia Forums

Thanks John I'm getting your line of thought now

You guy's have it all wrong!!!! If you had bought a nice Spanish bike that handles good and got rid of those imitation bikes from other countries then this whole thread wouldn't have been needed at all.
Great read though, lots of good meaty technical stuff.

I will also have to agree with Tri Cub on this one. It is a good read though.

There was a saying about 1970s trials bikes I remembered that fits here:
To do well there are things that the various bikes need - Spanish bikes need to be carefully maintained and Japanese bikes need to be carefully modified

.
It’s worth looking into this ‘loop factor’ in a little more depth. It’s a bit of a sidetrack but it does affect steering to some degree (under power).

The natural tendency of a vehicle under acceleration is to ‘squat down’ on the rear suspension due partly to the change in weight distribution since the vehicle’s C of G is considerably above the horizontal driving force. Thus it creates that ‘squat’.

We might assume that the drive force acts where the rubber hits the road, but it actually acts on a bike horizontally through the rear axle. And since the pull of the chain also has an effect, depending on the geometry of the swingarm & chain-sprocket set-up, the drive force can (& often does) produce a torque that extends the rear suspension. Since this is counter to the squat effect, it is understandably called anti-squat (or more correctly, anti-squat torque).

Apart from engine HP & grip on the road (eg sticky tyre, grippy surface, or otherwise) there are numerous variables that affect it: ie front & rear sprocket diameter, swingarm pivot position, swingarm angle & front sprocket position. The diag below ‘reveals all’.

When it all boils down in the geometry, it all depends on the dimensions 'a' and 'b'. They are the most critical. (In a sense the length and angle of the swingarm are irrelevant. What counts is the distance 'b'.)


You can see that the chain pull force ‘F1’ acting at a distance ‘a’ from the pivot produces an anti-clockwise torque (F1 a) compressing the suspension (ie squat torque) & will be greatest on full compression when ‘a’ is maximum. But the thrust force ‘F’ (ie drive force) acting at a distance ‘b’ from the pivot produces a clockwise torque (F b) extending the rear suspension (ie anti-squat torque) & will be greatest at full extension when ‘b’ is maximum.

Now it’s the overall resultant that is important. When F b > F1 a, the overall nett effect is anti-squat torque (Fb – F1a), but when F1 a > F b the overall nett effect is squat.

It is effectively an over-centre mechanism. At some point of suspension compression on most bikes the anti-squat torque (Fb) is going to be zero. That happens when b=0, ie when the swingarm is horizontal. But what happens to it on further compression? It becomes an additional squat torque added to that of the chain-pull squat torque.

Apart from ‘a’ & ‘b’ changing as the suspension is compressed, they are also dependant on the sprocket diameters & the height of the swingarm pivot relative to the front sprocket & rear axle. That’s why the pivot location is critical to squat/anti-squat (& why some factory GP bikes have adjustable pivot locations for diff gearing combinations, track/tyre/stickiness, suspension settings etc).

If you get the pivot too high you get too much anti-squat, thereby worsening the ‘loop factor’ by raising the C of G as the suspension extends under power, & also stiffening the rear suspension under power. But if you get it too low you get excessive squat & soften the rear suspension under power. (In fact on high powered bikes you tend to pull the tyre off the track & cause excessive wheelspin.)

Now as long as ‘b’ > 0 (ie the rear axle is below the swingarm pivot) there is going to be some anti-squat torque produced under power even though it reduces towards zero as the swingarm approaches horizontal.

Going back to the RL, it seems that its swingarm is not horizontal until the rear suspension is almost bottomed!


Add to that the RL’s ‘frisky’ engine & you can see why the high swingarm pivot potentially has such importance/effect. (You can also see that running longer shocks is likely to make the loop factor doubly worse, by both raising the C of G directly & increasing the anti-squat effect.).

While having some anti-squat torque still produced at/near full compression is not necessarily a problem in itself (since it’s the nett effect that matters), it probably does indicate that the RL has excessive anti-squat at laden ride height, raising the C of G & tendency to loop under power.

Apart from lowering the swingarm pivot, one could overcome it somewhat with shorter/softer shocks (which unfortunately also lowers ground clearance & slows the steering) or with a roller under the top chain-run to raise it a little further above the swingarm pivot, thus increasing ‘a’ (in Fig 5.15 above) & hence also the pro-squat torque, to overcome more of that anti-squat.


By contrast to the RL, look at this pic of a Greeves Pathfinder. The Puch engine has a very low countershaft sprocket, meaning the swingarm is almost horizontal, even unladen.


That’s going to produce very little anti-squat even unladen at full extension of the suspension. In fact I wouldn’t be surprised if there was noticeable squat under power even with only a very light rider on board (ie with minimal laden sag).

Love it John. That stuff about rear end geometry has fired me up thinking about swingarms again

Yeh, fascinating ain't it Dave, unravelling how & why things affect a bike's handling.

There's a bit more too...

Obviously apart from the tendency to loop, nett squat or anti-squat will affect the stiffness/softness of the rear suspension under power. That is, its action will differ under power to no power. If you have nett anti-squat at laden ride height, you will tend to ‘lock up’ the rear suspension under power. But if you have nett squat, you will soften up the rear suspension under power. Hardly ideal, either way!

It used to be thought that the swingarm should be level with the rider on board. That means zero anti-squat torque (‘b’ = 0 in Fig 5.15), so when any power is applied there would inevitably be squat & softening of the rear suspension.

Then it was thought that the rear axle, swingarm pivot & front sprocket centre should all be in line with the rider aboard (whether that meant horizontal swingarm or not). But that doesn’t achieve much of any real significance besides a taught chain.

Ideally, (if you want the rear suspension action to be independant of power applied) what you actually want is zero nett effect at laden ride height - ie squat & anti-squat torques equal & cancelling each other - so that the application of power gives neither squat nor anti-squat at that instant & suspension action is free of interference from power. ie So it does not vary under power or no power at that ride height.

That can be achieved quite simply because there’s a known relationship between F & F1, ie
F x rear wheel radius = F1 x rear sprocket radius.

This means that if the ratio of a:b (in Fig 5.15) equals the ratio of rear sprocket radius to rear wheel radius, then squat & anti-squat torques equal and cancel each other so there is zero net effect. If the geometry is arranged so that occurs at laden ride height there will be no nett squat or anti-squat at that instant, less tendency to loop, & the rear suspension action will be free under power. ie It won’t ‘lock up’ or ‘loosen up’ but have the same action/stiffness under power or no power. Quite simple really, if the factories had given it much thought.

However, if you want to be able to launch the bike up big rock steps as in modern sections you would want noticeable anti-squat built-in when you unload the rear suspension & apply power which would require diff swingarm geometry.

Unfortunately little/no attention was paid to this geometry in the era. I doubt many twin-shock bikes have been so designed, but as shown there is more than a little benefit in getting it right (or at least close). Foale notes, “In the past both squat & anti-squat was a largely ignored design feature… However current power levels &/or large suspension movements have demanded that attention be paid to this subject”. More responsive engines & stickier tyres have also paid their part.

Moving on to the KT250: – these are said to steer too slowly, whereas the TY250 is said to steer sweetly, but is it true? Why?

KT: 26.5deg rake, 79mm trail, 79mm axle offset, 1305mm w’base,
TY: 26.5deg rake, 90mm trail, 70mm axle offset,1295mm w’base,

The only surprise there is that rake is exactly the same & wheelbase is very close – the main things that affect quick steering. The KT will be marginally slower steering from the marginally longer wheelbase. Trail is quite different but doesn’t really affect the quickness of the steering.

But is there another factor here? Consider some more specs:

KT250: 96kg w lights, 310mm ground clear, 820mm seat ht, 385mm footpeg ht
TY250: 93kg w lights, 280mm ground clear, 760mm seat ht, 360mm footpeg ht

You can see the KT’s extra 3kg & that its C of G is going to be considerably higher than the TY’s, & that’s exactly how it feels - ie. Somewhat top heavy & ‘cumbersome’ compared to the TY. It would be more comparable if the TY had another 30mm ground clearance to match the KT’s.

But as is, are people confusing that cumbersome feeling with slower steering? Perhaps it should more accurately be said that the KT feels less responsive than the TY because of its weight & C of G. Riders sometimes jack up the KT’s rear end to make it ‘steer more quickly’, which raises C of G further. For better or worse? Or just different?

It might be more appropriate to lower the C of G by reducing weight up high, eg of tank, seat, airbox, exhaust, inner rear steel guard (all of which are quite heavy).

Two relevant observations here: i) Some of the Kaw factory trials bikes & the Gollner KTs seem to have the engine ‘rotated’ downwards at the front compared to the production KT (where the front of the engine & hence the crank, barrel & head is somewhat higher), thus lowering the C of G. ii) Before a recent trial I found & fitted some different pegs to my KT that are about ½” lower & a little more rearward and it made a noticeable difference to how it felt. (That was before I thought all this through. It was more ergonomics: I just thought I felt too high on a KT & the standard pegs are atrocious for lack of grip.) It'd be nice to lower them further and move them more rearward.

Hi.

I've really enjoyed reading the comments in this discussion - it's certainly given me lots to think about!

I'm just starting to build a trials Bantam, so I will try to incorporate some of the advice and dimensions above. Do any of you have any particular ideas for Bantams? I thought I might try to go for geometry something like the TLR250 - does that sound about right?

The current plan is to make some yokes and use a pair of GPZ305 forks (not anyone's first choice I know, but they are mint and 'in stock'!)

Cheers
Steve

What is a GPZ305?
You may want to read up on the classic rules before you get into it if you plan to ride in any MA events.