"Triumph twitch", heh, I like that. Bob's right. It is very predictable.
You feel the rear end start to walk around and you still have time to
react. But that's easy for us to say. we have both experienced it and
can feel it coming. If you haven't, it can be quite unsettling. It's not
like the dreaded Porsche 911 snap spin. Where your left wondering what
just happened as you exit the corner backwards. But you do get some
strange motions happening with the back end of the car. Worse if your
shocks are going or if your springs are a bit weak as well.
The scientific approach is also good advice, as well as different
drivers liking different set ups. Exactly why good communication between
a driver and the crew chief are essential for winning race teams.
One way to shortcut the scientific approach is to drive other peoples
modified cars. Making note of which modifications were made. Then sit
down with some good vehicle dynamics books (Carroll Smith books come
highly recommended), and learn yourself why the modifications perform
the way they do. Then, as Bob said, decide what you want your car to do
and go for it. It doesn't sound like much of a short cut, but it keeps
you from doing extra physical labor. Plus, you learn a bit more about
vehicle dynamics. Never a bad thing to do.
Again, as Bob said, it is easy to go to far and end up with a buckboard
ride. One of the things to decide is where will the car spend most of
its time. If it is to be like my TR6, a primarily street driven car,
with the occasional romp on a race track during open track days. Keep
that in mind when your setting it up. Just don't be upset when a more
prepared car walks away from you on the track. Remember, your there to
enjoy yourself and your car, not to impersonate Mario.
Shawn Loseke
1972 TR6
Fort Collins, CO
http://www.loseke.net/shawn
On Wednesday, September 4, 2002, at 09:40 AM, Robert M. Lang wrote:
> On Tue, 3 Sep 2002, Jim Hurley wrote:
>
>> Hello,
>>
>> On Tues. 3 Sep 2002 at 8:25 Shawn Loseke wrote:
>>
>> ....
>>
>>> A stock TR6 can be quite a handful when pushed to the edge. But it has
>>> been my experience that the handling foibles inherent in a stock set
>>> up
>>> can be easily cured without a major compromise in ride comfort.
>>
>>> Have others had similar experiences?
>>
>> Care to be more explicit? What do you mean by "a handful when
>> pushed to
>> the edge", what are the foibles, and how are they easily cured?
>
> Without going into a lot of detail, I would like to offer the following.
>
> Suspension movement is dynamic. The movement of the suspension relative
> to
> the body is predictable, but in the stock configuration, the body motion
> is considerable. Not nearly as considerable as a Cadillac, mind you, but
> when this is "tamed" the car will handle in a much more predictable
> manner.
>
> When you push a TR6 to the limits of adhesion, the car is actually quite
> predictable. What complicates the matter is that driver input can change
> at inopportune times as far as the suspension is concerned.
>
> For example: you enter a corner fast. Maybe too fast, but fast
> nonetheless. The front end will "push" a bit and when the driver detects
> this will either turn the wheel more (probably with no major effect on
> the
> car's motion - this is designed into most passenger cars to keep you
> from
> getting yourself into trouble!) or will lift the throttle. At this
> precise
> moment, the rear of the car will try to swing out, part of this motion
> is
> due to weight transfer from the front to the rear and part of it is due
> to
> camber/toe change which is the result of the "semi-trailing arm" rear
> suspension.
>
> Some people call this the "triumph twitch". I've driven a bunch of cars
> at
> the limit. This twitch is really not unique to Triumphs.
>
> At any rate, the question is: How do you fix it?
>
> Simple. Well, not really simple, but bear with me. If you increase the
> "roll stiffness", you will get less camber change. This helps keep the
> tire contact patch uniform. But you also get the potential added benefit
> of less rear suspension toe change and this also helps a lot because the
> car will tend to track in the same direction or very close to it.
>
> Believe me, it takes really small changes of toe and camber to make a
> huge
> difference in the feel of the car. So, if you tweak the suspansion to
> result in a degree or two of dynamic toe change you can go from "tail
> happy" to neutral.
>
> How do you get there? First off, you play with sway bars (assuming the
> rest of the suspension is okay). Go to a bigger bar in front first. This
> is a "safe" upgrade because all you are doing is increasing roll
> stiffnes.
> Most likely you will merely tune the car to "push" more. You then
> increase
> the camber a bit and try to optimize the contact patch. This will dial
> out
> some of the "push" because you are using the tires more efficiently.
> However you can only go so far before you start violating basic
> principles
> of physics.
>
> You then move to the rear of the car. Add a rear sway bar appropriate to
> the front bar. For example in the above case we may have installed a
> 7/8"
> Addco bar - cheap and effective. To "balance" things, we could install a
> 3/4" rear bar. Suddenly our "push" will be either neutral or it could
> even
> convert to making the car loose (oversteer). You then "tune" by making
> the
> effective "arm" of the rear bar longer (to soften) or shorter (to
> stiffen)
> the roll resitance.
>
> Still too much roll? Then you play with springs. And so forth.
>
> The idea is to increase one end and then bring that change into
> "balance"
> by playing with the other end.
>
> There are other factors to play with. Suspension deflection due to the
> bushing materials can result in camber and toe changes. This is why
> racers
> will toss the stock rubber pieces for harder materials. But there are
> tradeoffs. For example stiffer bushings can result in sheer forces
> beyond
> what the mounting points are designed for, so you get cracks in the
> frame. That's the tradeoff... better handling - more stress.
>
> You can play with "roll centers" by raising or lowering the car.
> There's a
> reason why most race cars sit fairly close to the ground... the roll
> center has been lowered. And most like a lot of other suspension aspect
> havebeen changed too, but that's a whole can of worms...
>
> Remember that the designers of the car had to make decisions about
> various aspects of the car's handling. We're probably talking about
> teams
> of engineers and "man years" of design, testing and development. Once
> you
> start messing with that, all bets are off and you have to re-duplicate
> those efforts. This isn't rocket science, but it's not "takin' out the
> garbage simple" either.
>
> You just need to decide what you want to do and then try to dial into
> that
> usage. It sounds rally easy, but you need to apply the scientific
> method -
> hypothisize, impliment, test, analyze, hypothesize...
>
> The situation is further complicated by the fact that the setup for me
> may
> be a whole lot different that the setup for you.
>
> And, of course, when you get rid of roll, you generally get rid of
> compliance - it is possible to tune your TR6 into a "buckboard". It'll
> corner awsomely, but you feel every pebble that you run over. That's the
> nature of the beast.
>
>> TIA,
>>
>>
>>
>> Jim Hurley "Life is too short for a backfire!"
>> '75 TR6
>> '83 F350 4 Door
>
> I hope this helps.
>
> rml
> ------------------------------------------------------------------------------
> Bob Lang Room N42-140Q | This space for rent
> Consultant MIT unix-vms-help |
> Voice:617-253-7438 FAX: 617-258-9535 |
> ------------------------------------------------------------------------------
|