|
Fuel Payload Slosh
Noise
On 10 March 2000, we made a
presentation to the "Automotive Fuel
Systems Beyond 2000" Conference at
Dearborn near Detroit, Michigan. The
title of the presentation was "Fuel
Payload Slosh Noise".
The presentation focused on the
control of Fuel Payload Slosh Noise.
The other issues dealt with
were the:
-
interaction of
experimentation and CFD
-
design of Fuel Filling
Systems
-
role of forecourt filling
nozzles in the design
process. |
Click to download AVI (1.0 MB)
|
Slosh Noise is a new Noise Vibration
Harshness (NVH) issue. A Tier 1
supplier had a problem with a new Blow
Moulded Plastic Tank they had
developed. The noise of the fuel
Sloshing was judged by the OEM to be a
fail on their NVH Quality Level. The
OEM demanded improvement from the
supplier or the contract would be
lost. We were asked to investigate the
problem by the supplier.
We developed a tilt rig capable of
simulating the forces applied to the
tank due to the braking of a vehicle.
A data acquisition unit was used to
measure the pressure in selected
positions in the tank and assessed the
Slosh Pressure over a five-second
period after the tilting of the tank.
Theoretical Analysis suggested the
application of the Impulse-Momentum
Principle:
S F.dt = D
(m.v)
Impulse = Change in
Momentum
As a result of the investigations we
were able to make recommendations for
a modified tank which was adjudged a
NVH Quality Level pass. The modified
tank had lower Slosh Pressures and no
"aftershock" which was a
characteristic of the original
tank.
Future projects need to embrace both
CFD and empirical investigation. We
are developing a comprehensive CFD
Slosh tool. The aim is to have a full,
3D, CFD tool to predict Slosh. This
will ensure our clients do not require
the production of multiple prototypes
of their products.
Fuel Systems Beyond
2000
Request a copy of the
PowerPoint presentation on CD or via
email contact
us.
-
There is a requirement to develop
the understanding of the
Fundamental Fluid - Mechanics of
Slosh
-
What is the payback for 100%
understanding?
-
Lower Production Costs
-
Simpler designs with less Slosh
control inserts
-
Shorter Development Times
-
and OEM confidence in Tier 1
Client
-
Good Experimentation, allied to
good CFD, will deliver the
ever-higher quality products
demanded by the OEMs.
| Fuel tank and filling pipe |
We undertook a series of experiments
using a 2D tank. The purpose of the
experiments was to establish Design
Rules for features on blow moulded
fuel tanks. The effect of the
following features on fluid flow was
examined:
-
Roof to Wall Radii
-
Sloped Roof
-
Roof Baffle
-
Floor Baffle
-
Beach Area
We compared data gathered for each
condition with that of an empty tank
and could use the results to predict
the behaviour of the fluid within the
tank as a result. Using the results we
obtained we were able to reduce our
customers development time and
prototype tooling costs.
What is the Payback from Tank
Manufacturers having Slosh-Avoidance
Expertise?
-
Less prototype
tooling
-
Avoiding $100k to $200k per
tool
-
Faster Development Time
-
Lower Prices to the client
-
OEM Confidence in Tier 1 Client to
deliver product
Fuel Filling System
Design
CFD & Appropriate
Technology
|
CFD requires good input data. The GIGO
Principle applies i.e. If you put
garbage in you will get garbage
out! If the Physic is wrong:
then the answer will be wrong
.
To design a fuel filler system there
is a need to understand Forecourt
Nozzle characteristics. Particular
attention must be paid to the effect
of Nozzle Aeration. A common
misconception is that all forecourt
nozzles are the same. We have undertaken
research in the past into nozzle
design in general and the causes and
effect of nozzle aeration in
particular. |
 Velocity profile through two types of nozzle |
|
To enable CFD to successfully model a
fuel filling system the correct fluid
characteristics must be known. Our
research identified that the
Volumetric Flow Rate of different
nozzle types varies from 103% to 180%
of the Fill Rate. We also found the
Nozzle Aeration Rates Vary from 3% to
80% depending on the type of nozzle.
This information is essential when
using CFD to predict system
performance.
For more information on our range of
services
contact
us. |
|