rec.autos.simulators

I once plucked that coefficient from an Arrows F1 car, which was 0.74.
Seems street cars have much more engine braking, like 1.5 (that's just
an educated guess though). Figured that out once I redid my drivetrain
to include the full analog clutch and the revving of the engine was
funny.

Loved the demo btw; looks great and drives pretty well (although I
only tested with the keyboard). Too bad you're going to have time off
to move.

Ruud van Gaal
Free car sim: http://www.racesimcentral.net/
Pencil art  : http://www.racesimcentral.net/

(probably wraps, so paste it together in your browser).

Second, I may indeed be wrong, but here's my view: an F1 car runs at
much higher revs than a street car. Having MORE engine braking than a
street car (at the same rpm) would give the racecar a disadvantage
since you have to pull the engine through all that torque trying to
brake the engine. I'd say an F1 car engine would be built more
efficient in that area than a street car (although other goals might
contradict this goal).

There are 2 parameters that decide the speed of revving up and down
(in neutral); the engine torque (which includes engine braking and
positive torque when the throttle pedal is pressed), and the engine
inertia.
A big flywheel gives more inertia, and that's the parameter I know
little about. Haven't got too many clues what a street car engine has
for inertia vs. an F1 car.
Still, I'd think an F1 car will use a much lower inertia (less
weighing flywheel) to be more responsive. Although ofcourse a higher
inertia gives you a more stable accelerating/braking car.

Did you drive it at LeMans back then? Would be way cool. :)

...

I think that's more to do with the low INERTIA of the components,
rather than the high braking coefficient (which just gives you the
torque with which the engine is trying to slow down, not the actual
speed at which this finally happens, since that is related to inertia
as well).
F1 engines must be powerful to push the cars through the aero drag,
even with high wing settings. So a massive engine braking contradicts
that goal, since at high RPM the engine braking would really fight the
engine from generating any torque to push the car forward through the
air. So I'd think striving for less engine braking would be beneficial
for an F1 car; at high speeds you depend on the aero downforce and
brakes to get good braking, instead of using the engine as a brake.

All this is just how I see it; perhaps someone with more knowledge
(Doug?) can shed a more practical light on this.

Right, but I think that's to do with the inertia of the Toyota engine
being far higher than the F1's. As the speed of engine rpm change is
like:
  engineAcceleration=engineTorque/engineInertia

I think the product at the right side here is higher on an F1 car just
because the inertia is much lower, although the engine braking torque
is a little lower.

Nope, they're completely separate. I define the engine braking
coefficient as being a measure of compression braking, and any
friction in the engine. I don't try to include any other implicit
effects. Couldn't do that even, because that would make it harder to
later implement strong winds.

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

hm.  I always thought engine output and engine braking were completely
separate.  engine braking is, like you say, just mass, while power
output decreases with increasing mass.

actually, a heavy flywheel does, but let's not be picky.  anyway;
I think there's a lot of magic around the flywheel.  I seem to recall
something about fitting a heavier flywheel might increase max power,
since the ratio of rotating mass to, uh, "mass that moves up and
down" [1] (i.e the pistons and con-rods) gets bigger.  or smaller. [2]

not that any of this matters, as I'll make a helpful comment in a
little while.

you're probably right [3], but [4] I'm pretty sure F1 cars have
back-torque limiters, which would mean you can pretty much just make
something up, or even make it user configurable, if you're a really
nice guy.

pah!  stability schmability! :)

oh ok.  you knew that.

is it bedtime yet?

[1] forgive me, I didn't get any sleep last night. :)
[2] see [1]
[3] I mean, who has ever heard of a heavy part for an F1 car.
[4] this is the helpful comment.
--
Gunnar
    #31 SUCKS#015 Tupperware MC#002 DoD#0x1B DoDRT#003 DoD:CT#4,8 Kibo: 2
                                     FJ?SE!

--
Gunnar
    #31 SUCKS#015 Tupperware MC#002 DoD#0x1B DoDRT#003 DoD:CT#4,8 Kibo: 2
                                silence is FOO!

The torque being sent out by the engine then is just linearly
interpolated from the lower to upper curve. I know, this isn't really
linear in real life, but for F1 cars, anything's possible (as it's fly
by wire), and linear seems quite ok (your controller isn't linear
anyway).

Ok. :) I indeed meant 'heavy'.

...

Hm, interesting, never heard of them. I'll have to google that.

...

Dinnertime here, lol. Must... get... food...

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

Jonny

Actually, no.  The mass is irrelevant to the steady-state torque
produced by the engine, be that driving or braking torque.  What
increased mass *does* do is make the car effectively heavier, so
it requires a greater braking torque to achieve the same deceleration.

There's an interesting concept called 'equivalent mass', which
includes the rotating parts in the complete vehicle inertia.  An
odd side effect is that the car is actually heavier in lower gears
- something I've noticed with my model race cars, which accelerate
better than I'd expect when I fit silly tall gearing for a big
outdoor track.

Because applying torque to the drivetrain is the *only* way the
engine can affect anything...

No, because then the overall torque is positive.

Jonny

Jonny

The one with*out* the exposed wheels and big wings.

The one which *doesn't* generate more than its own weight in
downforce at 100mph.

Hint: there's a thing called a 'lift:drag ratio'.  Basically
it means that if you want downforce, it'll cost you drag.  Racing
cars before the 70s were optimised for low drag.  Then some bloke
called Chapman discovered downforce, and things changed a little.

Second hint: a Celica has perhaps 100 hp, and a top speed of
perhaps 120 mph.  The F1 has 800 hp and a top end of 210 mph.

Aero power requirements rise with the cube of speed.

(120^3 / 100^3)^0.33 = 1.2 (~= mph / hp)

(210^3 / 800^3)^0.33 = 0.26.  IOW, the Celica is getting far
more maximum speed per horsepower than the F1 car - which suggests
it has just a tiny bit less drag.

Jonny

P = k1 * drag * speed^3
drag = k2 * P / speed^3

For the Celica, that gives drag of 100 / 120^3 = 0.000 057 8 [units]
For the F1, it's 800 / 210^3 = 0.000 086 4 [units].

Not a huge difference - the F1 is smaller but has a higher Cd
- but still the same way round.

Jonny
proving that you *can* get a decent degree without being able
to get the right answer first time...

Too bad btw that the US uses lbs, which is dependent on gravity, which
is a force, so you might swap weight for force thinking it's the same
thing. Mass has no direct relation to force on the other hand (in the
SI system).

We're probably talking about the same things here, though in a
different context or perspective, I think.
If we take:

  engineAcceleration=engineTorque/engineInertia

you can indeed see that higher (flywheel+engine parts) mass, or rather
inertia means engineInertia goes up, so engineAcceleration goes down.
Which is lower engine braking.
No argument there.

The way I use 'engine braking' in my sim is the torque the engine
generates (at the gearbox input) when no throttle is applied at a
certain rpm. If you're running 10,000 rpm and close the throttle,
surely the engine generates a negative torque to slow the drivetrain
down; I mean, pressing the clutch will let the car coast on while the
engine slows down faster.
So, in that perspective, at closed-throttle at 10,000 rpm, the engine
is generating a lot of negative torque which slows down the
drivetrain. This is what I call 'engine braking'; the amount of torque
that the engine generates at closed-throttle under a certain rpm.
Seems like this can't be far from the truth (it's working properly in
my sim).
The flywheel just constitutes more inertia, and I add the flywheel
inertia to the engine inertia to get a total inertia (at one end of
the clutch).

You'd say it does, otherwise the engine would accelerate?! If fuel is
burning, and no friction/braking occurs, the engine would accelerate
(if you applied the clutch, otherwise it would be dependent on the
wheels etc).

Idling is a balance between engine friction+braking and applying fuel.
It must be that we're using a different term of 'engine braking'. If
nothing would counter the engine from rotating, you would have an
accelerating engine. The fuel gives the power to rotate the engine
uptill the engine fights back just hard enough so that rpm neither
goes down, or up.

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

  In defense of Jonny, the coefficient of drag of modern F-1 cars is usually at
or above 1..  Several times higher than a Toyota Celica.  The frontal area is
lower, however, but if an F-1 car was optimized for low drag rather than
downforce, it would certainly not have a drag coefficient at least twice as
high as any road going car, would it?

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

My little car sim screenshots:
http://performancesimulations.com/scnshot4.htm

Ok, so then it seems both the engine inertia is lower, AND the engine
is fighting (giving negative torque) more than in a streetcar.

The lower gearing is a secondary effect though and has nothing to do
with the engine wanting to brake itself. The thing is I'm not looking
for a net effect, since if some mysterious power was rotating the
wheels, the engine wouldn't brake anyway, and you'd think engine
braking torque was 0, which it isn't; it's just counterbalanced.

So forget everything behind the engine and flywheel. Uptill there it's
an autonomous system and should be viewed as such to look at 'engine
braking'. Keeping the car in neutral or with the clutch applied, for
example.

Oh, yes, right. :)

Ok, then probably, with your experience, given a 0.74 coefficient for
an F1 car, a street car might end up more like half of that (instead
of being more).

That's true.

Must have been awesome. :)

Ouch! ;-)

Not at all; inertia is a fixed property of an object, just like mass.
Nothing 'causes' inertia; it's there because the thing has
molecules/mass.
Like F=m*a, T=I*w where T=torque, I=inertia, w=angular acceleration.
Mass is a constant property of an object, just like inertia (rather
the inertia matrix).

Right, but the inertia of the object remains fixed; it's the torque
that changes (the engine pushing or pulling rotationally), and the
angular acceleration is a result of that (w=T/I).

Surely the engine is pushing the drivetrain in the reverse direction
of which it (the engine/drivetrain) is rotating in. Otherwise it
wouldn't decelerate. So the engine is torqueing the drivetrain, no
doubt in my mind.

Power is just a measurement of energy and is just a nice statistic,
drag and inertia are inputs, yes.

Didn't know about the braking improvements, but that surely does make
sense. It seems though the downforce and brakes are most important,
since the downforce generates implicitly the maximum longitudinal
braking force you can apply to the car without locking the wheels. I'd
say the brakes of an F1 car are powerful enough at any speed to lock
the wheels, so in that sense, engine braking is of no importance for
the total amount of braking you can apply to the tarmac; the limit is
easily reached by the brakes alone, so engine braking just gets you
there faster.
But I may be completely off there, and F1 brakes may be impossible to
lock at high speeds (but I don't think so).

Hm, it's getting late. Have to rethink that. Indeed, less
flywheel/engine inertia gives better deceleration. In my perspective,
it's probably that the brakes can overcome the flywheel inertia as
part of their job, so at a certain point you don't need a lighter
flywheel.
But I guess then maybe the wheel brakes aren't capable of locking the
drivetrain at 300 km/h? Otherwise they could brake both the wheels and
the engine, so the flywheel wanting to spin on wouldn't make that much
of a difference.

Right, that and the engine parts themselves.

It's in there; my engine inertia includes both flywheel and engine
parts themselves (they are summed). The rest of the drivetrain
ofcourse also has inertia and can be tweaked in every which way.

The inertia just doesn't come into play until there's a torque from
the engine and you want to know how fast all the parts in the
drivetrain are going to accelerate (which is done in 2 parts in my
sim; preclutch and postclutch; they can rotate at different speeds,
which I need because I have an analog clutch simulated).

Interesting chat, haven't written this much stuff in a post for some
time. :)

Ruud van Gaal
Free car sim: http://www.racer.nl/
Pencil art  : http://www.marketgraph.nl/gallery/

 Hehe..  I'll be watching you more closely from now on :0)

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

My little car sim screenshots:
http://performancesimulations.com/scnshot4.htm