Thoughts about chatter

For years now the teams have been talking about "controlled flex" in
frame design. Now a new Michelin has appeared that appears to cause
chatter from either the front or rear or both.

As an engineer I never liked the idea of controlled flex. It just feels
wrong. Mechanical chassis design feels like it ought to be about
stiffness to weight ratios and wheel movement should be about suspension
spring rates and damping with every effort to separate the two.

I wonder if what has happened is that the natural frequency of slip of
the new tyres has now coincided with the natural frequency of vibration
of the frame-swingarm with insufficient damping in the tyre
construction. This would be independent of suspension changes which
might explain why it's almost impossible to cure by tweaking spring and
damper rates. And the solution may be to stop messing with controlled
flex and to build stronger/stiffer frames with higher natural resonant
frequencies.

It might also explain why the Yamaha is less affected by chatter but
appears to be more snappy at the sliding limit. This would make sense if
their frame-swingarm were stiffer than the Hondas.

Just a theory. Maybe there's nothing in it.

 

You may already know that when the bike is leaned over and hits a bump
the wheel travel will have some component that is vertical to the road
surface. So until a 3D (or at least 2D) suspension is developed, flex
in the chassis or suspension is all that's available. I agree that it's
an ugly solution.

And a very interesting one. Another solution would be to build weaker
frames with a lower resonant freq and do a better job of providing the
teams with the ability to control the chassis flex and damping.
Paraphrasing some famous mathematician/physicist, if you correctly state
the problem, the solution is readily apparent. A lot of solutions fall
out of your statement of the problem.

 

It's not at all clear that this is what's going on. Try imagining a
motorcycle travelling in a straight line traversing a 45 deg slope. Now
put a bump in the road surface. Perhaps as a ridge at 90 deg to the
direction of travel. To what extent does the M/C move sideways? Now
imagine the same thing with an infinitely stiff chassis and wheels but
with soft perfectly damped suspension. Now try and plot the sideways
forces on the (rotating) wheels.

I'm not at all convinced that sideways flexibility of the contact
patches is needed to keep the normal force of the tyres constant against
the road.

But then received wisdom is that M/C rotate around the contact patch
during turn in and I don't agree with that either.

Whatever, I feel sure that there's a connection between tyre grip
characteristics, tyre flexibility and chassis stiffness/flexibility.

 

Theoretically, given an infinitely stiff chassis with zero friction, the
suspension will deflect in the plane of the suspension no matter what angle
the motorcycle is leaning. I'll agree there. Also, wouldn't the force
transmitted to the suspension increase to infinity as the motorcycle
approached a 90 degree lean angle?

Since we DON'T have an infinietly stiff chassis and no friction, though, one
must plan for the sideways deflection of the chassis. There will always be a
vertical component of force acting through the contact patch at an angle to
the chassis which will cause a sideways deflection no matter how stiff the
chassis. You could make a chassis so stiff as to make it essentially
immovable for the purposes of road racing, but then you'd have to deal with
the fact that stiction in the forks due to side loading will greatly reduce
the response of the suspension's ability to convert out-of-plane forces to
in-plane forces.

-Chris-

 

I thought that the 16.5" wheels gave mor damping in the tyre when leaned
over and that the higher side walls gave room for more flex of the contact
patch, giving a larger contact area, and more flex in bumps where the old
17" had a contact very close to the rim.
Anyway, interesting idea, although I don't believe in a frame as stiff as
possible. But a frame with a frame with rather dead material, giving som
flex but minimal return spring action. (sorry for my bad description but you
get the idea) I also think Suzuki tried a very stiff frame in the '80ies?
Totally unridable.
As mentioned in other posts, you will allways have the grafity causing a
force right down and stickyness in the suspension that need some kind of
suspension help.

Well at least a frame with a different resonant frequency and even better
with an alloy that make the flex "softer".

/MBE

 

I'm and eng (electrical) too and agree that the idea of tuned flex
'feels'
wrong. You might be right that the problems with chatter are related
to
the newer tire designs.

I have been playing with an idea in my mind to solve this problem.
Can you give me your opinion on it?

Imagine a swingarm that allows a small abount rotation (maybe 5 to 10
degrees)
in an axis along the length of the swingarm. The top of the rear
wheel would
be able to move left-right a bit and the bottom of the wheel (the
contact patch) would move the other way. The rotational movement
would have a fairly
high spring rate and be well damped. This decouples the road bumps
from the chassis when the bike is on its side and the normal
suspension is less effective. This decoupling would greatly reduce
and interaction between road bumps and chassis flex. If chatter is
the result of the natural frequencies of these two things interacting
this system should stop chatter completely.

As a side effect when going around a corner and hanging off, the real
wheel would have the tendency to lean less than the rest of the bike
and might increase the contact patch, increasing traction and the
possibility to lean a bit more.

I imagine that the rear end of a bike like this would feel very
'loose' on the back end. Riders arent used to suspension moving in
this direction. So the damping and spring rates should probably be
fairly high. I really have no idea if the rotation would make things
difficult for the chain since it would lose a bit of alignment when
the suspension rotates.

What do you think?

Doug Eleveld

 

Heron Suzuki did quite a lot of experimenting then with things like rim
brakes. One of their prototypes used bonded aluminium sheet on expanded
ally inner. I think what happened was that it didn't work with the tyres
and suspension of the time and didn't feel "right" to the rider. It was
also awkward to work on with the engines of the time making things like
spark plugs hard to get at. That point about what the rider "thought" is
significant. I think you'll find that modern frames and swingarms are
stiffer now than that bike was then. And tyres give much more grip while
we're using quite a bit less rake and trail than then.

Somebody said there are no M/C engineers, only experienced M/C
mechanics. And M/C development tends to go surprisingly slowly as
everyone concentrates on last year plus 5%. If you look back on race M/C
development it's amazing it took them all so long to get at least some
stiffness into frames, only to then switch to aluminium with out
changing the layout. It then took another 5 years to discover that ally
designs need big box sections. This is all 1st year engineering stuff or
high school physics.

Controlled flex frames have now become gospel whether it's actually true
or not. Now that tyres, wheels, brakes and suspension have moved on
maybe it's time to challenge that.

But then Ducati 999s go faster if you remove a bolt on frame member that
didn't add much stiffness anyway. Go figure.

 

Interesting. Can you elaborate?

-Chris-

 

It's a piece about 12" long that sits horizontally just below the rear
of the tank on each side. Part of Ducati's cunning plan with the 999R
homologation special last year was to add a bolted in extra frame
member that adds another triangle. Apparently the factory riders go
faster with it removed because "there's more flex in the frame, giving
more feedback". I guess they know what they're doing but it sounds like
a load of old bollocks to me.

You can see it in two of the bikes here
http://www.twistgrip.co.za/ducati_999r_fila.htm

 

Engineering is always a compromise...

That said, there is another solution: tire technology. I'd assume that the
fatter the tires, the more they can help absorb bumps when the bike's leaned
over.

But I agree with everything everybodu has said in this topic, and would not
be surprised if the suspicion turns out to be accurate. What's next, tunable
chassis flex where stiffness in critical chassis areas can be dialed in?

....pablo

 

Along with a poor chassis design, chatter may be a misunderstood fluid
dynamics problem within the forks.

 

This is certainly the case where the problem is not well understood...

Those would be the design parameters that the factory and suspension
engineers haven't quite figured out. Hence, chatter continue to exist.

 

It's not at all clear that this is what's going on. Try imagining a
motorcycle travelling in a straight line traversing a 45 deg slope. Now
put a bump in the road surface. Perhaps as a ridge at 90 deg to the
direction of travel. To what extent does the M/C move sideways? Now
imagine the same thing with an infinitely stiff chassis and wheels but
with soft perfectly damped suspension. Now try and plot the sideways
forces on the (rotating) wheels.[/QUOTE]

You're right it doesn't move sideways. The wheel must travel further in
it's single dimension of travel to absorb the full bump height as
measured normal to the road surface, or sqrt(2) x the bump height in the
45 deg example.

 

That's racing. You don't have to be perfect, just better than your
competition. You'd be surprised at how little "engineering" goes on in
racing. The space agencies use teams of engineers with PhDs. Most race
teams may not even have formally trained engineer, and maybe guys with
undergrad degrees. I can tell you that there is A LOT of math and
physics that engineers with only an undergraduate education are missing.
Vehicle dynamics simply cannot be solved using a simple three degree of
freedom model. Newtonian mechanics is just not the best way to solve
complicated dynamics problems, but that's what undergraduate engineers
are taught.

I was arguing with a guy on Slashdot who claimed to have done data work
on Schumacher's Benetton team. He said they don't even use the data
collected to help design the following year's car. The modelling
package powerful enough to do vehicle modelling sat dormant on their
PCs (http://tinyurl.com/68r4b). So the statement that there's no
motorcycle engineers really applies to all of racing. The "real
engineering" is being done at research facilities (govt, corporate,
universities). You may actually find better engineering work from some
random amatuer racer than you will on a full blown race team.

 

That said, there is another solution: tire technology. I'd assume that the

Isn't that why Colin Edwards was testing with a 16in front in Dec-Jan? The
smaller front wheel needs a tire with a taller profile to keep the same
rolling circumference. His theory (from his website forums) was that they'd
be able to mitigate some of the chatter by letting that taller sidewall soak
up some bumps when leaned over. There were some other problems though that
made it a bad choice and thus he went back to the stock 17in wheel. In
fact, at the beginning of the season they had grafted on his superbike forks
and wheel at first but it wasn't any better in testing.

--Tim

 

Edwards has been using the 16.5" since his superbike days. He has
mentioned that he was developing the GP tire being used now a few years
ago in superbike.

 

Edwards had been complaining of chatter and an unsteerable bike for ages.
He recently got some new goodies, his team made some other adjustments and
suddenly he is taking podiums. Interesting.

 

Followed by the team telling him he had to run a tire he didn't like
in the Brno race.

 

Yup, they've been treating him like the red headed stepchild for the whole
season.

 

Is Edwards the new Max Biaggi? Remember Max always bitching that he didn't
get the best equipment, thus "explaining" his lame performances? Just ride
the bike Colin, like Sete is doing........

Fred