| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 291.1 | Nagging questions. | DATABS::SOO | We need the machine that goes *ping*. | Fri Apr 17 1992 18:21 | 37 |
| Question one:
In road racing, why bother with ride-height adjustment? Don't you
always want to have the car as low as possible without hitting the
ground?
Perhaps the adjustment is indeed for getting the minimum ride-height
without the car hitting the ground, and it varies from track to track.
But isn't changing the spring a better way? I.e. a taller and softer
spring for rough track and a shorter and stiffer spring for smooth
track.
I guess changing the relative height of the two ends of a car changes
the balance of the car but how is this differ from changing the balance
by changing the relative stiffness of the anti-roll bars.
Question two:
How does toe setting affects the handling of a car? I suppose the toe
setting in the front-end affects a car only when it is going straight
and at initial turn-in, because in a turn, the Ackerman setting takes
effect. Now how does the toe setting for the rear-end affects a car at
straight-away, initial turn-in, steady-state turning, and when powering
out of a turn?
Question three:
Which way is Ackerman and which way is Reverse Ackerman? Am I correct
in saying that the front-end sticks better in a corner if it has a
slight toe-out in it. Why?
I hope no one is losing sleep over these. :-)
-=Chong-Liang=-
|
| 291.2 | | ALIEN::MCCULLEY | RSX Pro | Mon Apr 20 1992 23:40 | 95 |
| .1> Question one:
.1> In road racing, why bother with ride-height adjustment? Don't you
.1> always want to have the car as low as possible without hitting the
.1> ground?
Not necessarily. There are a lot of complex interactions with the
various parts of the geometry involved in suspending the chassis of the
car over the track surface. Height is but one of them. The most
important thing (more important then minimizing ride height, really) is
controlling the motion of various centers, the instantaneous roll
centers are defined by the geometrical relationship between the various
suspension components. It's enough to merit textbook treatment, more
than a quick notes posting.
.1> Perhaps the adjustment is indeed for getting the minimum ride-height
.1> without the car hitting the ground, and it varies from track to track.
.1> But isn't changing the spring a better way? I.e. a taller and softer
.1> spring for rough track and a shorter and stiffer spring for smooth
.1> track.
No, springs are used to control motion, not static parameters like
height.
In general, stiffer springs are better in terms of performance, up to
the point at which they are so stiff that they no longer absorb bumps
but instead start to bounce and skitter over the rougher surfaces, but
they are also more difficult to control and less forgiving. Softer
springs damp out more, bumps and transients and control inputs alike.
.1> I guess changing the relative height of the two ends of a car changes
.1> the balance of the car but how is this differ from changing the balance
.1> by changing the relative stiffness of the anti-roll bars.
It is greatly different. For one thing, changing the relative height
changes the relationship of the roll centers (and instantaneous
centers), which alters the effects of suspension motion. Changing the
relative stiffness of the anti-roll bars changes the lateral weight
transfer but really does not change how the suspension moves (just the
difficulty of moving it!).
For another thing, changing the rake can have significant aerodynamic
effects, for some classes.
.1> Question two:
.1> How does toe setting affects the handling of a car? I suppose the toe
.1> setting in the front-end affects a car only when it is going straight
.1> and at initial turn-in, because in a turn, the Ackerman setting takes
.1> effect. Now how does the toe setting for the rear-end affects a car at
.1> straight-away, initial turn-in, steady-state turning, and when powering
.1> out of a turn?
I don't think it's really all that important, compared with other
things, but I'll need to refresh my understanding by consulting a few
of my texts on the subject.
BTW, I also do not think that the "Ackerman setting" is as signficant
as you suggest, I believe that "Ackerman steering" refers to the
difference in toe between the inside and outside front wheels during
cornering, to reflect the differing radius required for the wheels to
scribe an arc around the same centerpoint. In fact, considering weight
transfer effects (thus different loadings on the inside and outside
wheels), different speeds and radius corners, body roll, etc. it is my
understanding that Ackerman steering is of little concern except in
low-speed handling. Since most "real racecars" (tm) steer as much by
applying power to the rears as by turning the fronts, Ackerman is even
less important to many of us.
.1> Question three:
.1> Which way is Ackerman and which way is Reverse Ackerman? Am I correct
.1> in saying that the front-end sticks better in a corner if it has a
.1> slight toe-out in it. Why?
Why are you correct/incorrect, or why does it behave that way?
danged if I know, why do I care?
Ackerman would be having the inside wheel turn slightly more than the
outside wheel, so that it follows the tighter radius required to form
an arc around the common turning center.
I don't believe that the front end will stick better or worse because
of toe-out, the grip is determined by a very complex interplay of which
toe is a relatively minor part.
.1> I hope no one is losing sleep over these. :-)
Actually, I'd like to be. More precisely, I'd like to be in such a
situation that they were of sufficient importance to me to merit lost
sleep!
--bruce mcculley
|
| 291.3 | good stuff from On_Track | ALIEN::MCCULLEY | RSX Pro | Mon Apr 20 1992 23:51 | 28 |
| An interesting article that contained a few tidbits about this topic
recently appeared in On_Track (Vol 12, #6, 4/3/92, p.58). Under the
title "The Delicate Touch" Jeremy Shaw wrote about a test day in the
Barber Saab cars. Here is a taste of what he had to report...
"Indeed my car had been fitted with what is referred to as 'the
baseline setup.' The double-adjustable Koni front shocks had been set
at 'five' for bump and 'eight' for rebound. ... The fact that there
was a little bit more rebound than bump reflected the desire to
transfer some weight to the front of the car under braking, in
readiness for turning into the corner. The rear dampers, meanwhile,
were set at 'eight' for bump, and 'four' for rebound. This time the
theory was that the read of the car would rise under braking, again
promoting a shift in weight balance toward the front, while then
restricting the return movement so that the car remained balanced when
the power was applied."
Later, "the first major change we made was to reduce the bump settings
in the front shocks, and increase the rebound. Effectively we were
increasing the rate of compression and stiffening rebound. Again, in
theory we were attempting to reduce the push and instead induce more
'bite' in the front tires by transferring weight more quickly to the
front, and then by stiffening the rebound, attempting to keep that
weight where it was needed for a fraction longer."
There's more, but I'm not going to try to digest it to type it in
because it isn't as concise as those passages. It seemed worthwhile
though!
|
| 291.4 | | DATABS::SOO | We need the machine that goes *ping*. | Tue Apr 21 1992 14:45 | 36 |
| >The most important thing is controlling the motion of various centers
Ah, yes, I had forgotten about that one.
>springs are used to control motion, not static parameters like height.
I know, what I am saying is you can get the height you want by using
the right springs, thus doing away with the the height adjusting
mechanism and save a few pounds.
>I don't believe that the front end will stick better or worse because
>of toe-out
I think it does matter. The way I understand it is that toe-out helps
but toe-in will "loosen up" the front end because some traction is lost
from the wheels "fighting each other", so to speak. It is more
important, like you implied, in places where the inside wheel still
maintain a good amount of traction, and that include low speed corners,
initial turn-ins, and for cars that need really stiff rear anti-roll
bar.
I have an article about an autocross team that tested their RX-7 Turbo
on a skid-pad for a set of toe settings that generates the highest
lateral g for that car. I am not saying that that proves anything but
that toe-setting may make a difference, and I am looking for a theory
for it, that's all.
>I don't think it's really all that important
Well, maybe so and therefore not many people talk about it, or maybe
not many people understand it and thus everyone deems it unimportant.
A good deal of suspension tuning (or engine tuning for that matter) is
still an art, and I thought it would be interesting to find out what
different people know, think or believe.
-=Chong-Liang=-
|
| 291.5 | | ALIEN::MCCULLEY | RSX Pro | Tue Apr 21 1992 18:53 | 31 |
| .4> >I don't think it's really all that important
.4> Well, maybe so and therefore not many people talk about it, or maybe
.4> not many people understand it and thus everyone deems it unimportant.
Thinking about it on the ride home last night I decided that maybe I'd
understated the importance of toe in one way. I think the questions
were phrased in a form that made it easy to infer static settings, and
I answered in that way.
In fact I think the critical factor for all suspension tuning is that
it is dynamic. It doesn't seem too hard to get some setting that is
optimal for any given situation, tire slip angles and loadings aren't
that complicated. What gets complicated is making the tradeoffs for
a dynamic and ever changing set of conditions in the real world of
drivers racing in traffic. That's where I tend to think that all
suspension setups are inevitably compromises, because there is no track
in the world with every corner identical. Thus necessarily any setup
is suboptimal in some way, and it seems to me that the specific parts
are not so critical as how they all go together, and even that is much
less important than how the driver uses the package.
That to me is the most interesting part (of course, I'm a driver! :-).
Considering things like weight transfers you can see why a setup that
is good for one driver may not be as good for another, as their
techniques vary slightly in how they handle the transitions of braking,
cornering and accelerating.
It all comes down to the fact that setup is not just a mechnical
formula but truly is a matter of "tuning" requiring some intuitive
sense of the aesthetics involved in operating machines at speed.
|
| 291.6 | | ALIEN::MCCULLEY | RSX Pro | Tue Apr 21 1992 19:18 | 78 |
| .4> >springs are used to control motion, not static parameters like height.
.4> I know, what I am saying is you can get the height you want by using
.4> the right springs, thus doing away with the the height adjusting
.4> mechanism and save a few pounds.
Ah, but you can't. You can get some given height under static
conditions, but you cannot do that and also get the desired
characteristics for dynamic control of bump and rebound. You also
cannot cannot manage ride height under dynamic conditions of body roll
and weight transfer. You also cannot guarantee that you will have the
same spring rates on each corner, since corner weights interact with
ride height settings. It is really best to decouple static ride height
adjustment from dynamic spring rates.
.4> >I don't believe that the front end will stick better or worse because
.4> >of toe-out
.4> I think it does matter. The way I understand it is that toe-out helps
.4> but toe-in will "loosen up" the front end because some traction is lost
.4> from the wheels "fighting each other", so to speak.
The wheels are still fighting each other, whether they are pushing in
opposite directions or pulling in opposite directions. Tires are
sensitive to the scalar magnitude of the slip angle not to the vector
quantity (in other words, they don't care whether the tire has toe in
or toe out, the tires just know that there is some small amount of
scrub being induced because the plane of rotation does not coincide
with the direction of travel). So static toe setting alone does not
seem to me to be particularly important or desirable. Minimal toe
would seem ideal, in the absence of other considerations.
There is some possible compromise in that because tires require some
amount of slip angle in order to develop their best grip, and setting
toe might be one contribution to establishing this, although I always
though camber was more commonly used for that purpose. However, some
designs do not easily accomodate camber changes, so toe might be a way
of tweaking this area.
Also, remember that there is a vast difference between static settings
and the real world of dynamic changes. One very very very important
concern is something called "bump steer" which refers to the change in
toe that may accompany vertical movement of the suspension. Again,
I tend to believe that the static settings are less important than the
instantaneous situation when the suspension is doing its work. Static
toe (and all other static settings) define the starting points from
which the dynamic variations derive that instantaneous configuration,
so they may be relevent in terms of a specific design and situation.
But even there I believe the real important issue is the design and how
it makes all the various factors interact.
.4> I have an article about an autocross team that tested their RX-7 Turbo
.4> on a skid-pad for a set of toe settings that generates the highest
.4> lateral g for that car. I am not saying that that proves anything but
.4> that toe-setting may make a difference, and I am looking for a theory
.4> for it, that's all.
Yes, it makes a difference *for a specific design and situation*.
(And I'm not sure that the RX-7 is a particularly good one, does it
have IRS yet?)
BTW, please understand that my context is that of a formula car racer.
I assume that there is essentially little or no constraint on suspension
design. My race car has multiple pickup points to show that in the
past someone has reworked portions of the suspension design, and I view
this as SOP. From that perspective almost any production design (short
of an F40, perhaps) will look crude and unsophisticated.
If you are talking about a model of suspension theory, I believe a
purebred "real racecar" (tm, slogan="real racecars don't have fenders")
is the closest realization in practice. If you are trying to apply
that theory to production street cars, please recognize that any
experimental deviations from theory are much more likely to represent
compromises in the design of those specific examples, not shortcomings
in the theoretical model.
--bruce
|
| 291.7 | | DATABS::SOO | We need the machine that goes *ping*. | Wed Apr 22 1992 19:02 | 80 |
| Re. Spring height and spring rate
>It is really best to decouple static ride height adjustment from
>dynamic spring rates.
Ah, but spring height has nothing to do with spring rate, in the sense
that you can have springs of any height with any rate. I understand
that it may be much more practical to have just a couple of springs with
different rate and adjust the ride height with some other means. I am
just saying that it is possible to combine the two and perhaps save a
couple of pounds. I suspect that the answer is yes, but no one (with
the possible exception of F1) does it that way. Or is it that no one
has looked into doing it that way.
Re. Toe settings and Ackerman steering
>The wheels are still fighting each other, whether they are pushing in
>opposite directions or pulling in opposite directions.
Correct. However, in a corner where the outside wheel is doing most of
the work, the inside wheel will still contribute to the cornering
ability or "feel" of the car. With toe-in, the inside tire is in
effect pushing the car towards the outside of the corner, whereas with
toe-out, the inside wheel is pulling the car towards the inside of the
corner. Moreover, toe-out is the setting which the front wheels don't
fight each other as much.
>there is a vast difference between static settings and the real world
>of dynamic changes.
My question was on how the "toe" of a car affect the traction or
handling of a car in a turn. The reason I used Ackerman steering and
not static toe setting is because the Ackerman steering is what the toe
is in a turn, so let's forget about static setting.
>Yes, it makes a difference *for a specific design and situation*.
Yes, you are right. I was just trying to say that, everything else
being equal, the "toe" (in a turn, which is affected by the static
setting) does affect the handling of a car (may be just this car, but
I doubt it).
>(And I'm not sure that the RX-7 is a particularly good one, does it
>have IRS yet?)
Yes, since '86. It is of a multi-link design with floating hubs that
introduce toe-in in high-g turns (some racers disable this bit). I do
Solo I type time trials with mine and may go IT next year. Trying to
tune the suspension of a street car is most frastrating, but the
discussiong here is not specific to this or any car.
>I assume that there is essentially little or no constraint on suspension
>design.
Same here, and I am ignoring the effect of bump steer, camber and such
by assuming that they can be (almost) optimally set.
>If you are talking about a model of suspension theory
Yes, and I am trying to concentrate on a couple of things that I do not
have good understanding of. I know how camber, caster, bump steer, roll
center, roll stiffness, weight transfer, polar moment, etc. affect the
handling of a car (I think), but I am not sure about toe, Ackerman,
spring rate and shock rate. So I am here trying to find the theory
behind these by isolating the effect these have.
In view of that, let me rephrase my question: If everything else has
been optimally set, and you are left with only the toe settings for the
front and rear axles, and you can set them differently for the straight,
initial turn in, constant state mid-turn, and exit of a turn, how would
you set them? Why? (Make any other assumption as necessary).
People say things like "the stiffer the springs and shocks are, the
better", but I am not convinced. I tend to find answers that can be
explained in pure physics. I am perfectly happy if I can be proved
wrong, but I need to be convinced. Discussions like these also allow
me to look at things in other ways, which may lead me down other paths
in my understanding of the science (or art).
-=Chong-Liang=-
|
| 291.8 | Toe-in for me please... | NWTIMA::BERRYDO | Shiny side UP | Wed Apr 22 1992 20:49 | 16 |
|
Toe adjustments have a grear impact on straight line stability. Toe-out
will cause the car to be *very* unstable. The more toe-in the more
stability. As more toe-in is dialed in, more drag is produced. Bump
steer or pitch change during acceleration or braking can have an effect
on stability.
When I first ran my Camaro on the drag strip, I had set the toe-in at
the factory recomended settings. Because I was accelerating for the
entire 1/4 mile the front end was pitched up, causing a toe-out
condition. My crew said that they could read the writing on the doors
from their spot on the starting line. My spots were a little closer to
home :)
Don
|
| 291.9 | | ALIEN::MCCULLEY | RSX Pro | Thu Apr 23 1992 13:51 | 11 |
| .8> My crew said that they could read the writing on the doors
.8> from their spot on the starting line. My spots were a little closer to
.8> home :)
from your description, it wouldn't surprise me if some of those spots
were close to the spot where the sun don't shine?
sounds like you were saying that toe-in reduces loading on the pucker
factor!
:-)
|
| 291.10 | Am I missing the obvious??? | JUNO::JUPP | | Thu Apr 23 1992 20:03 | 22 |
| Maybe I've been missing something all these years, But I always thought
that the Toe-in or out was to stop the front wheels from wobbleing.
This is to say that if the wheels are given a bias then they won't
wobble.
I also thought that whether it was designated toe-in or toe-out was
related as to whether the steering arms connected to the wheels in
front of, or behind the axle.
As far as ackermann goes, a friend of mine who does stress analysis on
many makes of cars (including Ferrari) said that ackermann as such died
in the late sixties, It is now considered much better to have the tyres
"scrubbing" through bends, as this improves traction/handling.
Concerning the Camaro lifting and altering the steering geometry by
such an amount as to change thing from toe-in to toe-out, why not limit
the travel by either adjusting the shocks or by simply putting a stop
in.
These words are from a complete amateur, before you tell me so....
Cheers Ian...
|
| 291.11 | | ALIEN::MCCULLEY | RSX Pro | Thu Apr 23 1992 20:07 | 48 |
| .7> Ah, but spring height has nothing to do with spring rate, in the sense
.7> that you can have springs of any height with any rate. I understand
.7> that it may be much more practical to have just a couple of springs with
.7> different rate and adjust the ride height with some other means. I am
.7> just saying that it is possible to combine the two and perhaps save a
.7> couple of pounds. I suspect that the answer is yes, but no one (with
.7> the possible exception of F1) does it that way. Or is it that no one
.7> has looked into doing it that way.
Maybe I'm confused. I thought you were suggesting using different
spring rates to adjust the ride height based on the amount the chassis
weight compressed each rate. That's really nfg, because it screws
up the dynamics. If you were suggesting using springs of different
physical dimensions, so that the spring rates are unaffected, that
gives you no win, because you need the ability to adjust the mounting
points to suit - and that's exactly the mechanism used to set ride
height independently of springs! BTW, I'm not sure just what the
problem is that you're trying to solve, there is no complicated
mechanism involved here and at least on my car there appears to be
absolutely no weight penalty at all. There is a threaded spring perch
that can be run slightly up or down the threaded portion of the shock
to set the height at which the springs perch. No complexity, no
weight, what's the problem?
.7> My question was on how the "toe" of a car affect the traction or
.7> handling of a car in a turn. The reason I used Ackerman steering and
.7> not static toe setting is because the Ackerman steering is what the toe
.7> is in a turn, so let's forget about static setting.
I still think you're confused on this one. For one thing, considering
Ackerman should tell you that (unless you are considering static
settings) there is no "toe of a car" there is only toe of each wheel.
Granted, it *should* be symmetric, but I don't know that it can be
guaranteed. Ditto for linear, and possibly for variations with speed.
.7> Same here, and I am ignoring the effect of bump steer, camber and such
.7> by assuming that they can be (almost) optimally set.
I don't think they can be ignored, I think they are (necessarily)
interrelated so that they cannot be entirely decoupled. That's why
suspension tuning is a non-trivial exercise!
.7> Discussions like these also allow
.7> me to look at things in other ways, which may lead me down other paths
.7> in my understanding of the science (or art).
Strong agreement on that from me - that's why we're here, right?
|
| 291.12 | More ups and downs of suspension design | NYTP05::JANKOWITZ | Twisty little passages all alike | Mon Apr 27 1992 09:30 | 105 |
| I've been too busy to get an answer in here but this is one of
my favorite topics. I just got my car back together and did the
test day at Lime Rock last Tuesday so I have a few more minutes
to myself now. BGC Systems is one the the companies (there are
about 3) which sells PC suspension and chassis design/analysis
software. They are also my sponsor so this is a subject I'm
always working with.
First, the only job of a race cars' suspension is to keep the
tire contact patch in contact with the road while providing
optimal grip! There are mainly two types of movement, bump
travel and roll travel. Bump travel occurs while traveling over
bumps as well as during acceleration and deceleration. Roll
travel takes place in turns.
As Bruce mentioned, there is a compromise. In all conventional
independent suspension designs, camber changes evenly and in the
same direction (positive or negative) on both sides under bump
loading and in opposite directions (negative on one side and
positive on the other) under chassis roll conditions. The reason
suspensions change camber under these conditions is to attempt
to maintain the best tire contact with the road under these
situations. By knowing the camber curves of your suspension
under bump and roll situations, you can set your car up to
provide the optimal tire/road contact for your driving
conditions.
Two important rules can be applied:
� choose springs to limit bump travel
� choose swaybars to limit roll travel (although springs add a
small bit here too)
In most cases, we would like the center of gravity to be as low
as possible. The lower the C.G., the less weight transfer occurs
under acceleration, deceleration and cornering. We GENERALLY
would like to lower the car as much as possible without
bottoming too hard!
Between this and the camber curves we can decide on a set of
springs to limit the camber change during bump travel as well as
to limit bottoming. Since these are the deciding factors for
choosing springs, we would not want to change spring rates to
adjust ride height because we would also be affecting our camber
angles during bump travel.
Next, we can look at camber curves for roll travel to decide
what swaybar rates should be chosen to limit our camber changes
during cornering as well as limiting the weight transfer.
A number of other factors influence how much travel your car
will experience for certain spring/swaybar rates. These include
C.G. height, static corner weights, roll axis location (an
imaginary line drawn lengthwise through the center of the car
about which the chassis rolls. This can be determined manually for
static conditions but requires a computer for dynamic conditions)
and aerodynamic forces.
Together, all of these factors also affect the weight transfer
while driving. To get a neutral balance, you must also work with
all of these settings to control the amount of weight transfer
which occurs while driving.
It isn't a simple task! If you would like a better explanation
of this you can buy the Carrol Smith set of books or the BGC
Systems Chassis/Suspension Analysis Package which comes with
software to help determine what settings to use as well as a
manual which explains this better than I can here.
Another question was about toe settings. Toe settings affect a
cars stability. A slight amount of toe in in the front will
provide straight line and braking stability while a slight
amount of toe out in the front will improve stability at turn
in. As far as I know, you NEVER want toe out in the rear. For my
race car I usually set about 1/4 degree of toe out in front and
0 to 1/4 degree toe in in the rear.
Bump steer is a change in toe with bump travel. From what I have
been reading for the last few years, bump steer is undesirable.
I would assume that the intent of bump steer was to increase
toe-in under braking. People now seem to feel that they do not
want the toe to be changing while they are driving as this is
affecting their tires grip dynamically and would have to be
compensated for while driving.
That leads into anti-dive which is also frequently considered
undesirable. The idea of anti-dive was to prevent the front end
of a car from diving down under braking. The way this is accomplished
is by angling the suspension pickup points, putting the front
pickup point lower than the rear. As the front end drops, this
angle becomes greater and essentially binds the front suspension.
When the suspension binds, it no longer performs its other
functions and in effect becomes a solid suspension.
As far as using different springs to adjust ride height, Bruce
mentioned that there is no weight penalty. Since the suspension
height is also the adjustment used to set corner weights, it
would be impossible to use different length springs to make
these changes. The first reason is that you'd need springs in
increments of 1/10s of an inch all of the same rate. The second
reason is that spring rates are only good to a couple of %
accuracy so changing to a spring 1/10 inch different in length
wouldn't guarantee that you'd actually be changing the height in
the correct direction or the correct amount.
Glenn
|
| 291.13 | | DATABS::SOO | We need the machine that goes *ping*. | Mon Apr 27 1992 17:31 | 32 |
| Your explanation is perfectly clear, Glenn.
>Bruce mentioned that there is no weight penalty.
Actually, when I asked this question, I was thinking of a production
car, where adding the height adjustment mechanism can add a couple of
pounds of unsprung weight per wheel. If the ride height is rarely
adjusted, or if it is only adjusted to one of the three settings, then
having a non-adjustable setup and, say, three sets of springs for the
three tracks you drive at, may be feasible. There may be some other
factors that I do not know, like do you need to change the ride height
for different track conditions?
>I would assume that the intent of bump steer was to increase toe-in
>under braking.
I always thought that bump steer is a problem and not an intention.
The knuckle end of the steering tie-rod follows a complex curve when
the wheel moves vertically, but the tie-rod, being a simple rod, forces
the knuckle end to move in a perfect circle, and ends up steering the
wheel.
>the only job of a race cars' suspension is to keep the tire contact
>patch in contact with the road while providing optimal grip!
I think you meant "to allow the largest possible contact patch under
most, if not all, condition". I have one question regarding this: how
do you determine that you do get all the contact patch you want,
especially during cornering when the tire distorts? Is pyrometer the
only gauge of how close you are?
-=Chong=-
|
| 291.14 | more sticky stuff | NYTP05::JANKOWITZ | Twisty little passages all alike | Tue Apr 28 1992 13:14 | 62 |
| >> Actually, when I asked this question, I was thinking of a production
>> car
Which car? Bilstein has shocks with grooves and a circlip. The spring
perch sits on the circlip which you can put in any of the grooves.
The ride height is often adjusted for each track to control the
amount of bottoming. Usually, this is from 1/4 inch to 1 inch. If the
car isn't bottoming at a track, it's probably too high. You don't want
the car to smash into the ground either as this can upset the car in a
turn.
>> There may be some other
>> factors that I do not know, like do you need to change the ride height
>> for different track conditions?
Usually, you want to soften the car in wet conditions. This causes the
weight transfer to occur more gradually which can make the car easier to
drive. If you go to softer springs for a wet track, you would also
want to raise the car.
> I always thought that bump steer is a problem and not an intention.
I don't think it's too difficult to design a suspension without bump
steer. A simple method is to put the steering rack at the same height
as the control arm (upper or lower) with the tie rods being the same
width as the conrtol arm.
steering-arm
tie-rod | rack
| | |
v v v
---- ----
| |======o-----o[[[]]]o-----o======| |
| | | |
| | /--o o--\ | |
| |/ \| |
| |\ /| |
| | \--o o--/ | |
| | | |
---- ----
^^^^
a-arm
> I think you meant "to allow the largest possible contact patch under
> most, if not all, condition". I have one question regarding this: how
> do you determine that you do get all the contact patch you want,
> especially during cornering when the tire distorts? Is pyrometer the
> only gauge of how close you are?
I was also trying to allow room for weight transfer. Just having the
maximum contact patch on the ground isn't enough if the weight isn't
distributed well. A tire produces more grip with increased weight. So,
transferring weight to one tire increases its' grip, however the one(s)
that give up weight lose grip. These amounts aren't the same!
The tire companies have information about how much each tire distorts
and how good the adhesion is under varying loads. I'm not sure how to
get this or if they would even give it out. Without having that, the
pyrometer and a g-analyst are the only methods.
Glenn
|
| 291.15 | This AWD car has no steering! | DATABS::SOO | We need the machine that goes *ping*. | Tue Apr 28 1992 17:30 | 65 |
| >Which car?
Cars with struct type suspensions generally have to weld or glue on a
section of cylinder with thread on the outside, and a large nut/spinner
which the spring sits on.
>A simple method is to put the steering rack at the same height as the
>control arm (upper or lower) with the tie rods being the same width as
>the conrtol arm.
Assuming the drawing is a top view of the front end.
But then this end of the tie rod is going to fall outside of the rim,
if not the tire, if the rod is on the same plane as the A-arm. The
problem is just that you can't place the tie rod on the same plane as
either of the A-arms--there is no room inside the wheel.
|
|
V
---- ----
| |======o-----o[[[]]]o-----o======| |
| | | |
| | /--o o--\ | |
| |/ \| |
| |\ /| |
| | \--o o--/ | |
| | | |
---- ----
^^^^
a-arm
One solution is to use cables (assuming they don't stretch) to steer the
car:
(Top view, from side of car)
A-arm
vvvvvvv
| |
O O-----O O
\ \\ // /
\ \\ // /
cable->\ \V/ /<-cable
\ V /
-O-
Or a rod that turns, like a half-shaft, to steer the car. A side effect
is a saving in weight, since we have just done away with the entire
steering rack assmbly. Do you think I should apply for a patent here.
:-)
>I was also trying to allow room for weight transfer. Just having the
>maximum contact patch on the ground isn't enough if the weight isn't
>distributed well.
I agree with you if you mean weight transferred longitudinally, but
weight transferred from one tire to the other on the same axle may not
be desirable. I believe it is better to have weight more evenly
distributed between the two tires on the same axle, that is why the
end with less roll stiffness sticks better. Of cause, now the
subject of toe comes into play. It is important that the two tires
don't fight each other--this is more important when the weight is more
evenly distributed between the two tires.
-=Chong-Liang=-
|
| 291.16 | | ALIEN::MCCULLEY | RSX Pro | Tue Apr 28 1992 18:20 | 47 |
| .12> >I would assume that the intent of bump steer was to increase toe-in
.12> >under braking.
.13> I always thought that bump steer is a problem and not an intention.
.13> The knuckle end of the steering tie-rod follows a complex curve when
.13> the wheel moves vertically, but the tie-rod, being a simple rod, forces
.13> the knuckle end to move in a perfect circle, and ends up steering the
.13> wheel.
I too always thought bump steer was a side-effect rather than an
intention.
I know that it is not limited to situations involving steering tie-rods
and such, because one of the places where it is particularly
undesirable is in the rear suspension. (lends a whole new meaning to
the phrase "rear steer" eh?)
I don't understand the statement that "The knuckle end of the steering
tie-rod follows a complex curve when the wheel moves vertically"
because I believe that the wheel and everything attached to it is
constrained to move in a simple arc (ignored realworld imperfections in
rigidity and suchlike). The only thing I can infer is that there is an
assumption of steering input moving the knuckle, which I think is
completely irrelevent to bump steer effects.
To understand bump steer, consider the simple case of the rear wheel of
a formula car using a trailing link design. There are two such links,
with the forward ends attached near the cockpit and the trailing ends
attached to the rear hub carrier. There are also a wishbone and a
compression link attaching the hub carrier to the rear subframe
adjacent to the halfshaft (parallel with, and above and below the plane
of the shaft). This gives four points of location to the hub. Now, if
the wishbone and compression link are parallel and equal in length,
they scribe arcs that keep the hub carrier in the same relationship to
the track surface, changing only the track as it moves, and if they are
not parallel and equal length, the angular relationship of the plane of
the hub and the track surface will change, causing a camber change.
Likewise, if the relationship of the trailing links is assymmetric
and/or poorly chosen relative to the other attachments locating the hub
carrier, the motion of those links can cause a change in the angular
relationship of the plane of the hub with the longitudinal axis of the
car as the links accomodate vertical motion of the hub. This is a
change in toe of the rear wheel, and produces bump steer.
Forget steering knuckles, forget complex curves. There are only simple
arcs, and complex interactions.
|
| 291.17 | | NYTP05::JANKOWITZ | Twisty little passages all alike | Wed Apr 29 1992 09:30 | 38 |
| > Cars with struct type suspensions generally have to weld or glue on a
> section of cylinder with thread on the outside, and a large nut/spinner
> which the spring sits on.
Hmm, McPherson struts don't work like regular springs anyway since
they are pre-loaded. Just backing off the spring perch won't
necessarily change the ride height. It will change the pre-load on the
springs. Different length springs wouldn't help either. It seems that
the only way to lower the car would be to change the length of the
strut itself.
> Assuming the drawing is a top view of the front end.
> But then this end of the tie rod is going to fall outside of the rim,
> if not the tire, if the rod is on the same plane as the A-arm. The
> problem is just that you can't place the tie rod on the same plane as
> either of the A-arms--there is no room inside the wheel.
Yup, it's a top view and my drawing isn't to scale but my race car has
the mounting points right at the outside edge of the rim for both the
steering arm and the a-arm(s) and it has no bump steer. My 914 mounts
the a-arm just outside the rim also but I believe that it uses a
different method to negate bump steer.
> I agree with you if you mean weight transferred longitudinally, but
> weight transferred from one tire to the other on the same axle may not
> be desirable.
Without having the information relating load to adhesion, it's hard to
find the point where you get maximum grip. Either way though, weight
transfer is a fact of life unless you can get the C.G. at ground
level. I do agree that having equal weights on both sides is better
than having all the weight transferred.
An interesting thought though, in ski racing you WANT to transfer all
of your weight to the outside in a turn!
Glenn
|
| 291.18 | Non-preloaded McPherson struts | ASDG::ZETTERLUND | | Wed Apr 29 1992 11:23 | 16 |
| re .17
> Hmm, McPherson struts don't work like regular springs anyway since
> they are pre-loaded.
I believe that stock McPherson struts always have preloaded springs.
However, if you cut the springs or if you modify the strut with
adustable spring perches it is easy to get a non-preloaded strut.
On my BMW 1600-2 I have replaced the normal spring perch with a
threaded perch. I replaced the 5" OD springs with 2.5" ID springs;
the springs are 8" long. When the suspension is at full droop, the
spring is about 2" shorter than the distance between the lower and
upper spring perches. (The spring is fastened to the upper perch.)
BTW, Chong, I think that this is still legal for COM ST# classes.
Bjorn.
|
| 291.19 | | NYTP04::JANKOWITZ | Twisty little passages all alike | Thu Apr 30 1992 10:17 | 17 |
| > On my BMW 1600-2 I have replaced the normal spring perch with a
> threaded perch. I replaced the 5" OD springs with 2.5" ID springs;
> the springs are 8" long.
Does it work well?
I've actually gone the other way with my race car. I've put in
pre-load in the front suspension in order to assist turn in (less
initial weight transfer under breaking/turn in). I then use the pull
rods to adjust ride height in the front (not something most street
cars could do). Doing this, I can still use my desired spring rates to
limit bump travel. Unfortunately, my suspension geometry is limiting
the amount of pre-load I can put in. The next trick is to limit droop
by restricting how far the shock extends. This requires internal shock
modifications which I don't think I can do myself on my Bilstiens.
Glenn
|
| 291.20 | | DATABS::SOO | We need the machine that goes *ping*. | Thu Apr 30 1992 15:39 | 27 |
| ALRIGHT! We got up to four people into this discussion now.
Re .16
>I don't understand the statement that "The knuckle end of the steering
>tie-rod follows a complex curve when the wheel moves vertically"
If the upper and lower control arms aren't the same length or if their
pivoting axis aren't parallel to each other, and the knuckle falls
somewhere between the upper and lower control arm attachment points, it
doesn't follow a simple arc but a curve defined by the interaction of
the two control arms.
Re .17
>race car has the mounting points right at the outside edge of the rim
>for both the steering arm and the a-arm(s) and it has no bump steer.
You are right. I just verified that on a Truesports March.
Re .18
>Chong, I think that this is still legal for COM ST# classes.
Correct, but I am still in SSA.
-=Chong=-
|
| 291.21 | | ASDG::ZETTERLUND | | Thu Apr 30 1992 16:03 | 17 |
| re: .19
> Does it work well?
It works very well. The main reason for using the 2.5 ID springs is to
be able to lean the strut in for more negative camber. Some IT racers
bend the struts at the base instead! Before I converted to the race
spings, I had stock springs cut 2.5 coils and attached to the top spring
perch. Making sure the the springs fit into the lower spring perches
when lowering the car after changing tires was a real pain. It's also
nice to be able to change the ride height. I've been running the car
so low (the lower the better) that I may have compromised the
suspension geometry. I'm going to try to run the car at a height that
gives me maximum negative camber at full roll. When the rear shocks
wear out I plan to convert the rear suspension to coil-overs, too.
Bjorn.
|
| 291.22 | how do you do conversion? | NYTP05::JANKOWITZ | Twisty little passages all alike | Wed May 06 1992 09:43 | 8 |
| .18> I have replaced the normal spring perch with a threaded perch.
Bjorn,
Just curious, how does this work? Is it just a two piece perch with
threads which fits where the old one went, no gluing or welding?
Glenn
|
| 291.23 | Front strut coil-over conversion | ASDG::ZETTERLUND | | Wed May 06 1992 12:16 | 18 |
| Glenn,
The parts are standard circle track parts. I bought them from AFCO.
The lower spring perch consists of a threaded collar and T-shaped
treaded perch with a wear plate (big washer). I welded a support ring
to the strut below the treaded collar. I have silicone sealant to take
up the slight clearance between the strut OD and the threaded collar
ID. The upper spring perch is the standard cone-shaped type. I made
an adjustable camber plate that uses a spherical bearing to take the
load. There are IT-legal ways to get the same effect, but the coil
over set-up is very elegant. I've had it on the car since the 1988
season.
BTW. I removed the old spring perch completely and ground down the old
weld bead to leave a smooth outer surface on the strut before welding
on the support ring.
Bjorn.
|
| 291.24 | From a rally preparation viewpoint | ESBS01::RUTTER | Rut The Nut | Mon May 18 1992 15:42 | 20 |
| >> the springs are 8" long. When the suspension is at full droop, the
>> spring is about 2" shorter than the distance between the lower and
>> upper spring perches. (The spring is fastened to the upper perch.)
>> ...
>> perch. Making sure the the springs fit into the lower spring perches
>> when lowering the car after changing tires was a real pain. It's also
This sounds like a risky setup to me !
I can see that you wouldn't normally expect full droop of suspension
on a racetrack, but the consequences of this happening could be
quite serious indeed - if the spring didn't relocate in the perch.
I thought a standard technique was to have a limiting strap (or cable)
attached to the suspension to prevent it moving outside the range of
the springs. Have you considered this ? Not easy to determine where
it would be fitted at the upper end of a strut-type suspension, but
if a suitable mount could be found, it would avoid dislocated springs.
J.R.
|
| 291.25 | | ASDG::ZETTERLUND | | Tue May 19 1992 09:43 | 10 |
| re: .24
> This sounds like a risky setup to me !
It works just fine. The springs are not seated only when the car is
jacked up. The 2.5 ID springs have ground tops and bottoms; aligning
them at the bottom spring perches is not a problem. They are also
restrained by the strut tube so, in effect, they are really captive.
Bjorn.
|
| 291.26 | Can't the springs unseat anytime the wheel is unweighted? | MLTVAX::FISHER | Kill your television | Tue May 19 1992 17:57 | 15 |
| > It works just fine. The springs are not seated only when the car is
> jacked up.
Actually anytime a wheel is at full droop, right? I've seen a lot of photos
of cars coming over crests with both inside wheels lifted, could that cause
one of your springs to become unseated? I think that's what JR was talking
about. The spring would meet up with the spring again certainly, but perhaps
not in the right place, probably slightly altering the "ride height" of that
wheel, which I'd think would redistribute your corner weights a bit.
Something I have seen done to avoid this to make a slot in the upper and
lower spring seats and put metal hose clamps through the slots and around
the base coils to hold the spring in place while unweighted.
Carl
|
| 291.27 | | ASDG::ZETTERLUND | | Tue May 19 1992 18:32 | 12 |
| re: .26
The springs are attached to the upper spring perches just as you
described (and as I described in a previous note). I don't have the
springs attached at the bottom because I don't want the "metal hose
clamps" or equivalent to support the weight of the front suspension
at droop. The springs don't have much play around the strut tube and
can't come to rest anywhere but squarely on the lower spring perches.
Besides my car doesn't have enough power to droop the suspension
very much anywhere I drive, including Climbing Turn at LRP :<).
Bjorn.
|
| 291.28 | | NYTP05::JANKOWITZ | Twisty little passages all alike | Thu May 21 1992 09:19 | 15 |
| > Can't the springs unseat anytime the wheel is unweighted?
That's an interesting point. When my brother was racing in IMSA, we
would have to set the springs every time we took the car off the jack.
I can't have that problem on my front suspension because I've
preloaded the front springs. My rears have always set when I take the
car off the jack but there's nothing to insure it. Two problems could
occur if the springs weren't set.
- The corner weights would be off while driving
- If the spring sets in a turn, it could upset the car causing an
accident.
Didn't someone attribute an accident at Indy once on a spring setting
while driving?
|
| 291.29 | Further comments on max. suspension travel | ESBS01::RUTTER | Rut The Nut | Wed May 27 1992 06:34 | 37 |
| >>one of your springs to become unseated? I think that's what JR was talking
Yes, this is what I considered a potential problem.
>>Something I have seen done to avoid this to make a slot in the upper and
>>lower spring seats and put metal hose clamps through the slots and around
>>the base coils to hold the spring in place while unweighted.
That sounds the right solution. Now you mention it, I do recall
reading that in a [rally] preparation book some time ago.
I know the considerations of racing and rallying do differ, but I
still wonder at the risk involved. Also, with strut suspension,
the spring won't be 'going anywhere', but can still be out of position
when suspension closes up again - whereupon it may shift again.
I don't see any harm in having both ends of the spring attached
to the seat using some form of clamp. When the suspension is at
full droop (whether the car is jacked up, airborne, or leaning lots)
the clamp will only have to manage the 'unsprung weight' of the
suspension/wheel/etc. If it failed in this job, you would still
be no worse off than not having the clamp in the first place, but if
it is working as intended, you have a bit more security in the setup.
Just another consideration in suspension set-up. Really, the suspension
should not be allowed to extend further than the spring travel anyway,
if it does I would think things aren't completely right anyway.
Also, what does limit your suspension travel on droop ?
On compression, bump rubbers are quite standard, but in the opposite
direction they often do not exist. This can mean that your damper rod
is performing this duty, which it is not intended to do. Some other
form of limiting device should be used, a job which could be
fulfilled by fixing the spring to the upper and lower mounts...
J.R.
|
| 291.30 | | ASDG::ZETTERLUND | | Wed May 27 1992 09:51 | 14 |
|
re: .29
> Also, what does limit your suspension travel on droop ?
The answer is:
> ... your damper rod is performing this duty,
> which it is not intended to do.
Quite to the contrary.
Bjorn.
|
| 291.31 | | ESBS01::RUTTER | Rut The Nut | Wed May 27 1992 10:10 | 26 |
| >> > ... your damper rod is performing this duty,
>> > which it is not intended to do.
>> Quite to the contrary.
Perhaps we should agree to differ here.
I think that the function of a damper rod *should* only be to operate
the valving (or whatever friction device is used) of the [misnamed]
shock absorber during suspension travel.
If the damper rod/valving is used as the limiting factor in your
suspension travel, then the innards of the damper will have a
coming-together that they were not expecting. Again, due to the
force [hopefuly] not being too strong (not true if the suspension
is forcing the wheel down into a hole at speed), this may actually
result in damage to the damper. On compression, dampers often have
a 'bump stop' type of rubber to avoid the collision of parts that
will arise when the limit of travel is reached. Even with this,
your suspension should try and avoid depending on this part, as the
shock body is not intended to take the pressures of bottoming-out.
In theory, suspension components should only be required to perform
a single duty. To place multiple requirements on a component will
usually mean that it is not working in it's best fashion. Note 'theory' !
J.R.
|
| 291.32 | | NYTP05::JANKOWITZ | Twisty little passages all alike | Wed May 27 1992 10:17 | 23 |
| >> I don't see any harm in having both ends of the spring attached
>> to the seat using some form of clamp.
In my car, the top perch is threaded and the bottom just sits there.
If you attach the spring to the perch, the perch will become unseated
instead of the spring. I don't know which would be more desireable.
>> Really, the suspension
>> should not be allowed to extend further than the spring travel anyway,
>> if it does I would think things aren't completely right anyway.
Why would you say that? In my race car, I can adjust the ride height
using the pull rods on the front suspension. In the rear I just have
rocker suspension and the only way to adjust the ride height is by
moving the spring perch. If the spring was against the perch then
adjusting the ride height would be putting different amounts of
preload on the rear springs, which is considered undesireable.
>> Also, what does limit your suspension travel on droop ?
Yes, the damper rod is performing this job. That is part of its job!
One of the Indy car tricks to suspension tuning involves using spacers
inside the shock to limit droop.
|
| 291.33 | Still agree to differ on some points | ESBS01::RUTTER | Rut The Nut | Thu May 28 1992 06:22 | 44 |
| >>If you attach the spring to the perch, the perch will become unseated
>>instead of the spring. I don't know which would be more desireable.
No gain in either case I suppose.
>>Why would you say that? In my race car, I can adjust the ride height
>>using the pull rods on the front suspension. In the rear I just have
>>rocker suspension and the only way to adjust the ride height is by
>>moving the spring perch. If the spring was against the perch then
>>adjusting the ride height would be putting different amounts of
>>preload on the rear springs, which is considered undesireable.
Main difference here is that I was considering a more simple,
strut-type, suspension. In your situation, I would say that
the *ideal* way of adjusting ride height at the rear would
involve adjusting both upper and lower spring perches. Of course,
this extra mechanism would probably outweigh any gain in the
operation of springs/dampers...
>>>> Also, what does limit your suspension travel on droop ?
>>
>>Yes, the damper rod is performing this job. That is part of its job!
>>One of the Indy car tricks to suspension tuning involves using spacers
>>inside the shock to limit droop.
With regard Indy setups (of which I know nowt), the dampers at least
in this case are 'expecting' to be used as travel limiters.
My reading/knowledge is mostly related to rally-type preparation,
where suspension travel is going to be much more severe than is
usually experienced under racetrack conditions. If you then allow a
damper to 'top-out', there is a risk that you will damage it.
My comments as regard suspension component functionality would actually
be more relevant to race set-ups, but only 'in theory', as the real
thing may benefit from certain 'multi-functionality' even if this
goes against the supposed ideals.
Of these comments, I think that if you can avoid springs from coming
loose form their perches (without the perches then being loose !)
then that is worth doing. If the suspension travel is limited by
the dampers, that is not so important (when racing).
J.R.
|
| 291.34 | What makes a car track straight? | MLTVAX::FISHER | Kill your television | Mon Dec 07 1992 16:28 | 28 |
| I have a question about steering. Is there some feature of steering racks
or front end geometry which determines what "straight ahead" is?
When I was recently working on a friend's car replacing a tie rod end (I
made an effort to reproduce the adjustment), a test drive showed a very sharp
pull to one side. We couldn't think of any reason why anything except
the toe-in would have changed, so we adjusted the tie rod ends on both
sides and were able to eliminate the pull and also get the steering wheel
straight (without removing it).
I still don't understand how this worked. I had always thought that a
vehicle moving in a straight line would divide its toe-in equally between
the two front wheels, so any imbalance in the adjustments to the tie rod
ends would show up only as the steering wheel not being centered right,
but the car would still track straight, though maybe not very well.
So there must be some other factor which makes the front end want to align
in a particular direction, like maybe the steering rack itself has some
centering force to it.
The vehicle in question is a 1977 Saab 99 with manual rack-and-pinion
steering. Camber at each front wheel appeared to be equal and was not
adjusted in any way.
Can anyone explain this to me?
thanks,
Carl
|
| 291.35 | Tires same | LEDS::LEWICKE | That Hideous Strength----Polyester | Mon Dec 07 1992 16:40 | 7 |
| Carl,
Did it have identical tires on both sides? I know that when I've
had toe in screwed up one tire will track and the other will scrub.
(particularly enjoyable on standing water) If one tire on the car was
"dominant" it would always track and the other tire would always scrub.
John
|
| 291.36 | Tires the same | MLTVAX::FISHER | Kill your television | Tue Dec 08 1992 16:39 | 6 |
| Yes, the tires were the same on all 4 corners. My friend could not say
for certain that it did not have the pull before we changed the tie rod
end, but we certainly did eliminate it solely through adjustment of the
tie rod lengths.
Carl
|
| 291.37 | Q and A time | COMET::COSTA | Slap it, flip it, rub it down. | Wed Jun 23 1993 02:04 | 20 |
|
It is very important to have the tie rods as close as possible to
being the same length. Differences of as little as an 1/8 of an inch
can create undesirable scrub and produce a dominate tire. It is also
important to check track width at the front of the front tires against
the back of the front tires. This will give you your total toe. Toe out
will make a car more stable, toe in will make it more responsive, but
we are still talking about very small differences here. Usually 1/16 to
no more than 1/4 of an inch.
Now, I have some questions on front suspension systems. Coil spring
versus torsion bar specifically. Would a torsion bar system be superior
to a coil system, or vice versa? I would think that since a torsion bar
lies flat, you would have a lower center of gravity compared to an
upright coil. Would a torsion bar be lighter than a coil and have less
unsprung weight? Is the progressive rate of a torsion bar more or less
desirable then the linear rate of a coil?
TC
|
| 291.38 | Springs and things | AUSSIE::COLE | This is lucky Phil. | Wed Jun 23 1993 02:30 | 53 |
| re -.1
> It is very important to have the tie rods as close as possible to
> being the same length. Differences of as little as an 1/8 of an inch
> can create undesirable scrub and produce a dominate tire. It is also
> important to check track width at the front of the front tires against
> the back of the front tires. This will give you your total toe. Toe out
> will make a car more stable, toe in will make it more responsive, but
> we are still talking about very small differences here. Usually 1/16 to
> no more than 1/4 of an inch.
I think it is the other way round. Toe out forces the inside wheel
to more of its share of generating steering force. For a track car,
I would adjust toe for the best handling compromise and then measure
for future reference (like after you hit something and need to put
it back for the next race).
> Now, I have some questions on front suspension systems. Coil spring
> versus torsion bar specifically. Would a torsion bar system be superior
> to a coil system, or vice versa? I would think that since a torsion bar
> lies flat, you would have a lower center of gravity compared to an
> upright coil. Would a torsion bar be lighter than a coil and have less
> sprung weight? Is the progressive rate of a torsion bar more or less
> desirable then the linear rate of a coil?
Spring wise, a coil spring in nothing more than a torsion spring
wrapped up to make it more compact. There is no progressive rate
unless the geometry of the situation (push rods, bell cranks and
so on) makes it so. Also, you have to fit the damper in there too.
The only decent dampers would appear to be telescopic, so you are
stuck with a coil shaped thing anyway.
Modern open wheelers tend to put the spings above the drivers legs,
lying almost horizontally, with the dampers mounted co-axially within
the spring. The whole mess is operated by pushrod and bell crank.
Springs don't have that much mass anyway, relative to
the mass of the car, so C of G in probably a tenth order consideration.
The mass of the torsion bar would be exactly the same as the equivalent
coil spring, since elasticity is a bulk property (spring rate will
depend on the material and volume of the spring).
If a coil acts directly on the suspension, then some of its mass will
count as sprung weight, but if you use a linkage, then all the spring
can be sprung.
Both torsion and coil systems achieve rising rate by having geometry
which varies with wheel travel. In addition coils can get rising rate
by winding them with a varying pitch. In this case, the rate rises
as the coils bind progressively, so it is a somewhat crude method.
That's enough opinions for now.
Phil
|
| 291.39 | | COMET::COSTA | Slap it, flip it, rub it down. | Wed Jun 23 1993 20:57 | 21 |
|
My understanding of generating additional force from the inside tire
on a turn was to work with rod end angles and perfecting bump steer
conditions, rather than going to additional toe. Too much toe would
scrub the tires too much on the straight costing you speed and
overworking the tires.
Now back to springs.
For purpose of my comparison, I was thinking along the line of stock
front stub race cars. Chevy versus Dodge, actually. I'm sure once you
have a governing body that allows a great deal of modification, then
there are plenty of ways to work around the shortcomings of any system.
I'm stuck with working on stock stubs that must use stock components
and are 20 years old. I also believe that the Chrysler set ups are rear
steer as compared to front steer in the GM.
TC
|
| 291.40 | Toe in or out? that's the Questions | VERSA::ROADES | | Wed Jun 30 1993 19:09 | 12 |
| re: toe
we who run stock cars on oval tracks (dirt and asphalt) use 1/16 to 1/4
inch toe out. (the larger number for dirt were scrub in the straight
line is less important.) The toe out stables the car and provides a
tighter turn for the inside wheel in the corners..."ackerman". This is
for oval racing. Wonder why guys who turn left and right what toe out?
shouldn't we really want to same? Whould not toe in or out stablize
the car? What is the diff??
jeff
|
| 291.41 | | COMET::COSTA | Time to rebuild | Wed Jun 30 1993 19:57 | 11 |
|
The tighter turning of the left front wheel should be accomplished
thru geometry changes rather than just toe. If you run static toe out,
when you turn and unload the left tire, the bump steer can make the
tire toe in, or return to neutral, which is undesirable, espescially if
the need for steering corrections becomes needed. I seem to recall
seeing that an oval car needs the left front to steer and additional
3-5 degrees more than the right front.
TC
|
| 291.42 | MOre on toe | AUSSIE::COLE | This is lucky Phil. | Thu Jul 01 1993 00:48 | 17 |
| re: re: toe
Us guys who turn both ways want toe for the same reasons as those who
turn left. On road circuits biassing the camber and brakes to the left
turns is not such a good idea, however :-). Here in Sydney, there are
both CW and CCW circuits, and it would be real pain to change the
setup all the time, particularly at my level, running a modified road
car.
Toe in or out would be different, because the change in steering force
due to lateral weight transfer would be in opposite directions. Scrub
radius would also have a lot to do with this.
I just adjust to until I like the way the car handles. If I'm feeling
really scientific, I might them measure it.
PHil
|
| 291.43 | | AUSSIE::COLE | This is lucky Phil. | Thu Jul 01 1993 00:55 | 11 |
| re .41
> The tighter turning of the left front wheel should be accomplished
>thru geometry changes rather than just toe. If you run static toe out,
Depends whether your regs allows this. Best thing is to move the rack
or whatever around to increase ackerman angle. This can be hard if the
rack is in front of the wheels and you find it necessary to put the
steering joints in the middle of the front brake rotors.
PHil
|