rec.autos.simulators

I started racing GPL, like many others, using 85/30
ramp angles in the differential.  I was a little
bit different, though, in that I used six clutches
because of the stabilizing effect they had on my
tendencies to trail brake.  Well, after reading
several things stating that ramp angles of 45/85
would be much closer to those that are actually
used (or had been used), I started trying to make
some setups using the 45/85 angles.  Needless to
say I had (and am still having to some extent)
trouble adapting.  I still use six clutches and I
have become accustomed to the throttle sensitivity
required in mid-corner situations.  Just keeps my
concentration levels up. :)  My main trouble is
adapting to the coast side angle of 85.  I simply
cannot come to grips with the lack of corner entry
grip provided by the rear end with this ramp angle,
even still using six clutches.  I've backed off a
bit and now use a 60 angle with six clutches, but
even with this, I find myself countersteering ALL
THE WAY INTO THE CORNER from the moment I turn in
to the moment I hit the apex.

My question is this:  Is this effect in any way
realistic?  I'll concede that some of this effect
may be my setup, but it isn't entirely that.  If
these cars are SO apt to spinning on corner entry
with these settings, like GPL appears to be, I have
a MUCH larger respect for the ability it takes to
be a quick race driver.  I've been using the 45/60
ramp angles now for at least two months and still
find myself spinning out on corner entry about 5-10%
of the time.  I know one suggestion will be to move
my brake bias back to 58% or 59% from 56%, but that
compromizes my straight line braking to the point
that I can't outbrake anybody anymore.  I've had some
luck using Nunnini's 40%/60% spring and 60%/40% ARB
theories, but it's still a bit of a handfull.

Well, I've rambled on long enough.  Discussions
or suggestions on setups would be appreciated.

Thanks,

-Chris-

Sent via Deja.com http://www.racesimcentral.net/
Before you buy.

I use only 1 clutch as the transition from power on and power off is
much smoother and more consistent and to qoute Carrol Smith "the most
common misconception with these units is that the amount of preload
(clutches) is a primary determining factor of the amount of lockup. It
can be, but things are a lot better if it is not. The spreading force of
the bevel gears in an order of magnitude greater than any conceivable
preload".

What seems to work pretty well for me is to make smooth transitions from
trailing throttle to heavy braking...for instance going down the front
straight at Kyalami as I approach my braking point I begin to roll off
the throttle smoothly as I equally apply heavy but smooth pressure to
the brake (I have the bias set to 51%). Having the coast side set to 85
means that there will be lest likelyhood of locking up the rear wheels
under braking not to mention that it does not cause a lot of mid corner
understeer as you might get with a lower coast side ramp angle.

Let me know if you'd like to talk shop, I'd be glad to exchange thoughts
and experiences with anyone that's been using this kind of setup perhaps
we could all learn something. Also, I'd be happy to swap setups/replays
etc.

Also, I'd strongly recommend buying yourself a copy of "Drive to Win" by
Carrol Smith as there is plenty of info on differentials.

Thomas JS Brown.

I'm much more comfortable with it now than previously, but most
tracks I'm still slower than my 85/30 times.  I have, however,
used it to good effect at Monza, where I can now take the Eagle
or the Lotus to middle or high 1:27's after the tires are good
and warm.  Not consistent at anything but low 1:28's though.

I agree.  I've found myself becoming smoother at my throttle/brake
transitions.  Especially the point at whick I let off the gas.  I'm
also now very careful not to turn in abruptly (such as Curva Grande)
while still on the brakes, or I'll swap ends.

Here's where I'm gonna say <huh?>  If I'm correct, the numerically
larger the ramp angle is and the fewer clutches that are present,
the less lock-up force is supplied.  This all leads me to believe
that the inside rear wheel is *MORE* likely to lock up, rather than
less, thus, the trailing throttle oversteer since most of the engine
drag torque is given to the inside rear.  Also, if the inside rear is
locked because of the brakes, there's less diff pressure to split
the ground dragging torque present on the inside rear.

Yes.  I like the mid-corner charactarisics much better.

Um, too late.  I think I just did. :)

I may just do that.  It would go nicely on the shelf next to
the Millikin's "Race Car Vehicle Dynamics" that I got on a whim
back when I was in school.  Engineering text book store and
all, I just couldn't help myself. :)

Sent via Deja.com http://www.deja.com/
Before you buy.

"Since we want minimum lock up under braking, especially
trail braking,
 typically we use an 80 or 85 degree ramp on the coast side
with a 45 degree
 ramp on the power or drive side."

"The original 90 degree cost ramps tended to chip and have
largely been replaced
 by 85 degree units. For faster corners with more download
some teams use 60/30 ramps.
 I have never been happy with 45/45 (too much corner entry
and mid-phase understeer),
 but I know some very successful people who are - in the UK.
I don't know anyone
 who likes them in slow corners or on bumpy courses.
Personally, I cannot conceive
 of using any ramp of less than 80 degrees on the coast
side.
 Currently many successful teams are using 80/80 ramps with
little or no preload, depending
 on the spreading forces of the bevel gears to limit the
slip. This is a very benign setup
 if there are no really slow hairpins (or not much available
torque)."

--
Richard 'Ricardo' Nunnini
http://website.lineone.net/~richardn
 - mirror in case the ISP chokes  -
http://nunnini.homepage.com/

-Chris-

PS:  I hope you'll forgive me.  I've been indulging too
much in the fine drink this evening already and seem to
have waxed poetic.  A rare occasion for me, to be sure.
Pray I don't drink further or you all will be smitten
with verbiage the likes of which none have known before. :)

-C-

PPS:

<VVBG>       *  hic  *

- Me, who else? -

Once again, from "Drive to Win":

"To my mind, minimum preload should be used only to ensure that the
diff's transitional characteristics from power off to power on remain
consistent and that the unit will limit slip when one wheel is completely
unloaded. One of the problems with the clutch packs - especially on road
cars and with sedans - is that, when a driving tire is completely
unloaded, the diff functions on preload only - the ramps don't come into
it." - Carrol Smith

Like I mentioned in my previous post, I have my brake balance set to 51%
and the front tires are the only ones that lock up under heavy braking (
the least loaded one), the trailing throttle oversteer seems not to be
due to a rear wheel locking up but rather weight shifting forward to the
front tires (I pretty much run around 3.5 to 4 inches ride height) which
is easily corrected by squeezing on a bit more power and/or maintaining
constant throttle and locking up the front tires if you've really stuffed
it up. This of course will ruin a good lap, but it will keep you heading
in the right direction and more often than not on the track rather than
the grass or the kitty litter.

Thomas

In a nutshell, I wonder if a 1967 grand prix car differential would be setup
radically differently from say, a heavier sports car with much more momentum
to overcome the engine braking, or an F2 car with a smaller engine and
therefore less engine braking.  Basically, I'm thinking it would take more
force to get the inside rear wheel to lock up on a heavier or less powerful
car.  Might explain why 85/45 feels neutral in GPL, even though it defies
conventional wisdom.

    -Aub

May your poetic verbiage wax well
David
:)

buy.

Color me stumped.  My intuition says that if the coast ramp is set
at 85 (i.e. nearly no lock up) that the inside rear wheel will lock
up due to severe braking.

I see several things happening when a driver brakes for a corner:

1)  When the brakes are applied, the rears unload evenly and the
fronts load evenly proportional to the amount the rears unload.

2)  When the driver begins turning in, the inside rear unloads
further and that load is correspondingly transferred to the outside
front tire.

3)  Now, the inside rear tire has the least amount of grip availabe
because it is supporting the smallest amount of static weight
of the car of any of the four tires.

4)  We shall assume that the coast angle (85) and the clutch preload
(1) are both set to minimum for the following theories.  It is assumed
that these settings on the differential will make it behave almost
as if it was coasting as an open differential.

5)  In the following theories, I'm lumping the mechanical braking
torque in with the engine drag torque, so that the case of rear
wheel lock-up can be considered a bit more simply.

These thing I belive are correct.  Following, I have several theories,
but I'm not sure which is correct.

Theory 1:  Under these conditions, as the inside rear wheel approaches
the threshhold of locking up, it will receive 94.5% of the engine drag
torque because it has the least amount of grip available to it.
(Similar to the way an open differenial will transmit 100% of the
drive torque to the spinning wheel when it slips, but modifed for the
85 degree ramp angle.)  When this happens, the outside rear won't
receive enough engine drag torque to lock up, and the car is prevented
from spinning, since the outside rear wheel is responsible for
(probably) about 80% of the total grip of the rear wheels.

Theory 2:  Same as Theory 1, except that the outside rear WILL lock
up because it now receives a disproportionatly large share of the
engine drag torque.  The car will then spin because the inside rear
locks, robbing the car of about (probably) 15-20%) of the grip at the
rear, thus unbalancing the car from the position in which the driver
has placed it.

Keep in mind that I've lumped mechanical braking torque and engine
drag torque in together, since the rear wheels don't differentiate
between the two.

I don't know which theory is correct, but I do know that the car
will display an increase in it's tendency to oversteer under trailing-
throttle and trail-braking situations.  I hate to question a prophet,
but the above quoted section from Carol Shelby goes against what my
intuition tells me, which is that the locking of the inside rear
wheel, due to minimal differential locking pressure caused by 85
degree ramp angles, causes the trailing throttle oversteer.

I'm certainly open to opposing view points and would love to hear
discussions on this topic in further detail.

-Chris-

Sent via Deja.com http://www.deja.com/
Before you buy.

Make that OUTSIDE rear.  Damn, I hate it when I***up.

Sent via Deja.com http://www.racesimcentral.net/
Before you buy.

Not in my experience. The least loaded front tire is the only one that even
begins to lock (both fronts if I've gone too deep).I don't know which theory
is correct, but I do know that the car

Carrol Smith, not to be confused with Carrol Shelby of Cobra fame.

No offense, but if I have to make a choice between your intuition and Mr.
Smith's experience as an engineer with the likes of, Adamowicz, Amon,
Andretti, Bond, Cobb, Foyt, Gurney, Hulme, Ickx, Matich, McLaren, Millen
(S.), Moffat, Ongais, Posey, Rahal, Redman, Ruby, Revson and Vasser.....well
the choice seems pretty cut and dried.

        However, this discussion often assumes Papyrus has done a
brilliant job of modeling the real thing, bevel gear spreading and
all.  I don't think so.  So observations about how the game responds
may have validity even tho they disagree with the Great Man.

                bob

Could you be more specific? In what way does the differential not correspond to
the real thing in your opinion?

It wasn't an observation per se but rather a hunch that Mr. Cavin had about how
the rear wheels may lock up under heavy braking due to the diff being more open
on the coast side. My personal observations/experiences within the game tell me
otherwise.

Thomas

My original hunch went something like this:

Things I'm assuming are fact (please correct me if I'm wrong):

--A 30 coast ramp provides maximum locking force while an 85
coast ramp provides minimum locking force.

--Six clutches provide maximum locking force (for a given ramp
angle) while one clutch provides minimum locking force.

--With the above two assumptions, a 30 coast ramp with six
clutches would provide the most locking force available, while
an 85 coast ramp with one clutch would provide the least
locking force available.

--The greater the coast locking force, the less likely the
inside rear wheel will be to lock since the two rear wheels
are kept rotating (by the high differential locking force)
at close to the same speed.

--The lesser the coast locking force, the more likely the inside
rear will be to lock since the the two rear wheels are allowed
(by the low differential locking force) to rotate at very different
speeds.

My own personal observations:

--I originally used a 30 coast ramp with six clutches.  This
combination gave me stability under trail-braking and trailing
throttle situations.  The car exhibits little to no trailing-
throttle oversteer with this combination.

--I have since tried an 85 coast ramp with one clutch.  This
combination gives me much less stability under trail-braking
and trailing-throttle conditions.  The car exhibits much trailing-
throttle oversteer with this combination.

The combination of the things I'm assuming as fact with my
personal observations:

--I observed the greatest amount of trailing-throttle oversteer
when the differential was set with the minimum amount of coast
locking force.

--I observed the least amount of trailing-throttle oversteer
when the differential was set with the maximum amount of coast
locking force.

Combination of observation, assumption of fact and theory:

--Since I observed the greatest amount of trailing-throttle oversteer
when the differential was set with the minimum amount of coast
locking force, this oversteer condition is caused by the inside
rear wheel locking due to a lack of differential locking force.

--Since I observed the least amount of trailing-throttle oversteer
when the differential was set with the maximum amount of coast
locking force, this stable condition is casued by the inside rear
wheel not being allowed to lock due to the differential locking
force.

These are only my observations and theories based on how GPL behaves.
I have no experience with a rear-wheel drive car, road or racing.
So far, I can't see any flaws in my logic, but I invite others to
do so.

Cheers,

-Chris-

Sent via Deja.com http://www.deja.com/
Before you buy.

Thats correct

Again, correct....however keep in mind that the more preload, the less
the ramps come into play. The diff will lock on preload only.

And therefore creating a lot of mid corner understeer since the unit acts
more like a spool then an open diff.

Only if your brake bias is set too far rearward, otherwise under heavy
braking the least loaded front wheel will lock while the rear wheels are
free to turn.

But bags of mid corner understeer.... Right?

<snip>

The only way your argument makes any sense is if you've got your brake
bias set too far rearward. Think about it for a second, you're claiming
that less locking is more likely to lock a wheel....

However try setting the brake bias more rearward and you'll find that
BOTH rear wheels will lock due to the fact that the diff acts more like a
spool.

In my experience with GPL the trailing throttle oversteer you describe is
not a product of a locked rear wheel but wieght transfer. Ask any race
driver what happens if you suddenly lift off the throttle (even while
going dead straight).

Driving with a 45/85 is much more demanding due to the fact that you must
balance the car on the throttle and brake (and how smoothly they are used
in conjunction), the transition from power on and power off is much more
volitile and less forgiving but with practice allows more control (in my
opinion) through the various phases of cornering.

<snipped from Ricardo's post>

This is what allows me to set the brake bias at 51% and not lock a rear
wheel, if you question the validity of my assertion I will gladly provide
you with a copy of my setup and a replay.

Thomas