[Diy_efi] RE: Diy_efi digest, Vol 1 #413 - 12 msgs

[Shannen Durphey]

Didn't you once say you specialize in asking questions?  This is a great
one.  It's funny how we have an instinctive grasp of engine load, and
yet the definition can get very slippery when we try to put our fingers
on it.  

I should probably look around for an official definiton for engine
load.  I would say, with some thought, that load is resistance to
acceleration.  In this way, 100% load describes a situation in which the
engine is unable to accelerate, and zero load (which is physically
impossible) describes a situation where the engine can change speed
instantly.  Also, an engine under zero load is unable to produce either
torque or power, as those values are both defined in part by
acceleration.   This definition also allows load to be independent of
power and torque.  

In regards to the dyno, with load defined as resistance to acceleration,
we can apply an inertial load by using the tendency of mass to resist
acceleration, or we can apply a resistive load, which is a force, to
maintain a desired acceleration.  The resistive load can be measured
directly, as in a friction brake dyno, where power and torque values can
be worked out from the known force.  In the case of the electric brake
dyno, we can measure power required to maintain a desired acceleration
directly and calculate torque.  We generally don't take into account any
friction made at the contact patch between tires and rollers.  Instead
we calculate or measure the power required to produce acceleration at
the roller and assign that to the power produced by the wheels.  When
the dyno is producing an "8hp load" it's really creating a force equal
to the force applied by the wheels, but opposite in direction, and so is
creating a negative acceleration.  Load on the street would be a
combination of the inertia of the drivetrain, force of gravity, wind
resistance, and friction.

The true measure of load is change in rpm.  We are lucky to have a
natural ability to modulate engine power to meet load without even
thinking aboout it.  We can sense a change in acceleration and apply
more or less force to the road to meet that change in acceleration.  How
we apply more or less force is by modulating the throttle.  So we
associate load with throttle position.  If we grew up driving 2500 lb
vans with 10 horsepower engines, we might not have that association. 
The throttle would in that case be almost always fully on or fully off,
and we might only tend to think of situations where we need to apply
negative acceleration to slow down.  We also associate load with MAP, as
we know there's a fundamental air flow relationship between throttle
restriction and the pressure across the restriction.  When MAP is high,
it infers less restriction, which must mean we're attempting to overcome
a large load.

The issue between "large" and "small" loads shows up over time.  We live
in a world of finite distances.  To cross these distances, we tend to
assign an expected time.  Even without a watch, we have a sense of speed
that we use, which measures the time it takes to cross a percieved
distance.  With our grasp of speed and time, we modulate power to create
the acceleration we need to cross the percieved distance in the expected
time. : )

So where does this lead?  If we had a perception that was directly
related to the energy released by consuming fuel, in addition to the
velocity and acceleration that results from burning fuel in the engine,
then we might know without thinking that not all energy released in
combustion creates a force on the crank.  A fair amount of that energy
heats parts.  

Adam mentioned earlier that there's a fixed rate of heat transfer
through metals.  When an engine has been operating for a long enough
time, the temp of the cooling system + the temperatures of the heads and
pistons + the energy vented out the exhaust + the power output at the
crank has reached an equilibrium with the energy  produced by
combustion.  If we now desire to cover a distance in less time, produce
an acceleration, we consume additional fuel, convert additional energy,
to produce that acceleration.  If the duration of this event is short,
we don't make large changes to the equilibrium of the of the cooling
system/heads/wasted energy/crankshaft power system.  There isn't enough
time to significantly heat the cooling system and heads, and what isn't
used at the crank ends up in the exhaust.  As the time through which we
accelerate increases, so does the time which excess heat energy can heat
the cylinder heads and cooling system.  And the slower the acceleration,
the longer it takes for the exhaust and intake cycles, the less heat
energy is removed from the cylinder through these means.

Power production drops as density of the air entering the combustion
chamber drops.  Power maximized by adjusting the start of the burn to a
specific time will drop when the rate of burn changes.  If we go from
combustion to detonation, power production drops off rapidly, engine
damage is looming.  Increasing the heat level of the system will
generally decrease intake air density, decrease burn times, and
aggravate detonation.  It's important to fight these gremlins.

Decreasing air density suggests decreases in fuel.  Increasing burn rate
requires delaying the start of the burn.  But the real killer is
detonation.  No matter how carefully we adjust fuel and spark, if we end
up with detonation we lose power and eventually parts.  To prevent
detonation, we have some options:  1) reduce the time we spend
converting energy, that is, lighten the vehicle or let off the throttle.
2) alter the relationship between the engine's acceleration and our own:
change gearing, alter the load on the engine.  3) reduce the energy we
produce:  add an inert gas to the combustion process, for example  4)
try to keep the heat energy away from the cylinder heads.

Most tuning at the dyno centers around option 4.  We already work to
keep heat energy away from the heads during acceleration by providing a
mixture rich in fuel.  A/F mixtures richer than roughly 14.7:1 don't
react any more fuel than mixtures at 14.7:1.  But they absorb heat from
the combustion process and affect the burn rate.  So if we have a
situation where we're getting additional heating of the chamber, we can
add more fuel to absorb the heat, displace oxygen, and decrease burn
rate.  The trick is to add just enough fuel to prevent detonation and to
not add so much that power drops unacceptably. 

Gear ratio compensations _should_ be time compensations.  I've never
used an ecu with a gear ratio compensation function, so they might be
just that.  But the time compensation should be (imo) to first reduce
spark then add additional cooling media (fuel, water, alky, whatever) as
the amount of time under load increases.  For control systems which do
not have time compensations, the tuner must estimate if and when
cylinder heating will cause detonation.  And the tuner must add the
necessary amount of fuel to prevent detonation before it happens.  Which
means that if the tuner is wrong, or if the conditions under which he's
made his estimation are substantially different from the operating
conditions of the vehicle, his tune is less than the best.

<whew>

Shannen
William Shurvinton wrote:
> 
> I must admit in shame that there are a couple of concepts here that I am
> still having difficulty getting my head around, despite having followed it
> each time it is discussed. Whilst I understand that a truck engine in a
> truck has to work harder than the same engine in a hotrod I still have
> confusions about 'load'.
> 
> I can't help wondering whether is it because I am thinking about it from a
> MAP or TPS% perspective (NA). These are used to infer load, but are merely
> measures or inferences of the air entering the engine. So if you are
> comparing 2 applications on part throttle, for the same throttle opening,
> the light vehicle will need less fuel than the heavy one. Normally you
> modulate the throttle to match the load so you don't actually see this,
> except perhaps the drivability issues with a non-linear throttle reponse due
> to having too little 'load'. Or to put it another way TPS% is a measure of
> potential power, not actual or required.
> 
> The Dyno of course has difficulty replicating this.
> 
> It may click after a few more years tuning the car, or am I still
> approaching it from the wrong direction?
> 
> Bill
> ----- Original Message -----
> From: "Shannen Durphey" <shannen at grolen.com>
> 
> <snipped everywhere>
> 
> ."  Yet again the tune for
> > the long, heavy load isn't optimum for the short, light load.
> >
> > Gear and load corrections can be a good thing.  > Shannen
> 
> >
> > Dave Dahlgren wrote:
> > >
> . The better after
> > > market stuff has tables and corrections for what gear you are in. You
> can and
> > > should change timing fueling and boost according what gear you are in as
> it
> > > compensates for the relative load on the engine.
> > > Dave
> > >
> > > Shannen Durphey wrote:
> > >
> > > > Using a braking type dyno to tune a car for WOT can produce misleading
> > > > results.  A lightweight car with a high power to weight ratio would
> > > > typically never see the same load and duration that can be applied
> with a
> > > > brake type dyno
> 
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://www.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi