Page 8 - Fuels & Lubes International 2014 - Q2

FUELS & LUBES INTERNATIONAL Volume 20 Issue 2 8
wherein an equal amount of time is
spent at three different temperatures
(20°, 33° and 75°C), the fuel
economy benefit was 5.5% at 20°C,
3.9% at 33°C, and 3.2% at 75°C.
The difference is due to the fact that
oil viscosity drops as temperature
rises. Therefore, friction within the
engine drops as it warms up.
As can be seen by this example,
the maximum that can be achieved
varies as operating conditions
change. “The ideal results indicated
by the model cannot actually be
achieved, but they provide a target
of low friction lubes and hardware
to aim at,” Mainwaring said.
“There has been some significant
progress,” he added. “In prototype
lubricants that Shell has managed to
put together, we’re close to achieving
an actual six percent fuel economy
improvement for this engine on
this test cycle relative to the starting
point lubricant. That’s a pretty sub-
stantial benefit.”
Such work runs into two chal-
lenges. For one thing, ideal lubricants
don’t exist and real world lubricants
cannot eliminate friction entirely.
Another hurdle is that lubricants also
protect parts from wear, an impor-
tant task that ensures engines don’t
fail prematurely. A lower viscosity
lubricant will result in a thinner pro-
tective film, potentially bringing an
engine closer to the edge of failure.
“Shell can model the lube and
bring specialized knowledge to
the task,” Mainwaring said. “But
to really get the greatest possible
benefit, what’s needed is co-
engineering of the lube and the
engine. Getting involved as early
as possible in the design cycle will
allow more trade-offs to be made
between lubes and engines.”
There are signs that OEMs are
considering such an approach that
co-engineers lubricants and propul-
sion systems. More extensive, more
sophisticated and more complete
modeling of components before an
engine is built is possible because
of ever more powerful computers.
“Over the last 10 years
compute capacity has grown over
a thousand-fold. We see this trend
continuing,” said Radu Theyyunni,
engineering group manager for
systems design and performance
analysis at General Motors.
That computing power is put to
use with the analysis of components,
sub-systems and systems in the
engine. The resulting simulations
include information on the size of
the parts and their relationship to
one another. In addition to such
dimensional and geometrical infor-
mation, other elements of the model
involve the properties of the different
materials and fluids in the engine.
Such data, and improving upon it,
are vital.
“Having accurate material
properties is key for future growth
and fidelity,” Theyyunni said.
At present, the accuracy for
engine performance expectations is
about 2-3%. That is, the measured
parameters of the finished engine
should agree with the simulated ones
by better than 97%.
“Lube analysis is among the most
complex parts of the model because
lube properties vary significantly
with temperature, aeration and age,”
Theyyunni said. GM’s lubricant mod-
eling is growing more complex and
sophisticated, with such factors as
vent system design, oil slosh, friction
and bearing performance being more
accurately simulated. However, more
must be done, and that may require
outside expertise.
“Increased cooperation with
the lube makers and formulators
would be very helpful and essential
in further growing the models,”
Theyyunni said.
He added that hitting the
upcoming 54+ mpg goals may
require it. Further friction reduction
will almost certainly be part of any
solution to that challenge.
Such co-engineering can take
many forms. For instance, one
Small Block V6 truck
engine used in the
Chevrolet Silverado.
Photo courtesy of GM.
Image of oil slosh analysis in
the engine on an aggressive
maneuver on a race track.
Photo courtesy of GM.
automot i v e co lumn
Chrysler innovation involves more
rapidly warming up transmission
fluid, said Black. Doing so stabilizes
the fluid’s temperature and confines
its properties to a narrower span.
In turn, that allows for systems to
be designed so as to take advantage
of this greater fluid uniformity over
the operating range.
This sort of thing could serve as
a model for other areas of a vehicle.
A heat exchanger, for example,
might be able to keep axle lubes or
engine oils in a narrow temperature
range. Determining if this makes
sense involves estimating the cost of
implementing such innovations and
weighing that against the potential
payback. It is much less expensive
to do this in a model initially than to
determine this first in hardware.
The computer evaluation of
these and other proposed innova-
tions is possible not only because
of improvements to computing
hardware, but also advances in soft-
ware, such as Chrysler’s Powertrain
Matching Analysis Toolset, a soft-
ware package that allows real world
data to be brought into a virtual
model. Once there, it can be part of
the calculations, enabling designs to
be effectively evaluated before any
hardware is actually built.
“Such models only go so far,
though,” Black said. In a crankshaft
in an engine block, for instance,
structures are under very dynamic
loading conditions, which causes
flexing and geometry changes. Then
there’s the fluid flowing through the
bearing system to consider.
“As it flows through into the
bearing opening and then out
through the bearing system, it’s
changing its temperature and
viscosity on the fly,” Black said.
Current modeling assumes
the bulk parameters of various
properties of the material. To more
completely model what’s going on,
though, it will be necessary to look
at characteristics on a microscopic
scale. That requires greater knowl-
edge of the properties of the various
components and materials within
the engine. In turn, that sort of
information may be something that
lubricant makers are best suited to
supply, at least for their products.
Of course, any model that is
developed has to be verified. That
is not cheap, as it means measuring
a slew of process parameters and
then comparing these to what was
predicted by the model. However,
it is necessary, as a model misfire
can wipe out any savings a virtual
engine generates.
It won’t be possible to eliminate
all friction and claim the entire
possible prize, Black acknowledged.
However, the effort to create better
models and get as close as possible
to the ideal will pay off in terms
of reducing fuel consumption
and emissions.
“We spend a lot of our gasoline
[consumption] on rubbing friction.
As we make improvements in this
area, both from an analytical and de-
sign approach, it really will have a sig-
nificant impact on the customer and
the CO2 generation,” Black said.