Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

What's a mixed-flow jet pump ?

340 views
Skip to first unread message

TEAMULTRAC

unread,
Sep 2, 1997, 3:00:00 AM9/2/97
to

>>carl, care to toss another bone?

DESIGN & DEVELOPMENT: JET-PUMPS
by Carl Camper @ ULTRAC Performance Systems
http://www.interlog.com/~jlogan/ultrac
1500 N.W. 62nd Street, #510
Ft. Lauderdale, FL 33309
954-351-1943 / 1944 fax

This article will examine variations in pumps and their applications. The
following list details a few of the subjects that will be discussed.

1) Axial Flow vs. Mixed Flow vs. Centrifugal Flow Pumps
2) Single Stage vs. Dual Stage
3) Pump Sizes and "X" Dimension
4) Variations
5) Materials
6) Future Development

Note: Many sections of this article have been deleted for proprietary
purposes.

PUMPS:

The jet pump is a water accelerator. It functions as a medium to convert
and/or transmit horsepower into thrust... it is a drive system. In the case
of Personal WaterCraft Vehicles, it not only creates thrust, but enables
handling. Handling is a by-product of vacuum, which is created at the
intake source.
The jet pump is a derivative of two separate technologies that when
combined, produce favorable results.... the propeller and a ducted
environment coupled with Bernoulli's venturi principles. There are several
variations of jet pumps that have evolved over the years. Most drive
systems were created to meet the demands of a given application,
requirement, or powerplant.

The jet pump is really several components combined into one. The Inlet or
Intake Gullet/Housing, the Ramp or Intake Grate, the Impeller and Shroud,
the Stator Section, the Venturi and Convergence Cone and last... the
Steering Nozzle or Thrust Deflector. A Gimbal Ring may also be found in
cases where a trim system is utilized. Some designs incorporate segments
together, i.e., the stator section and venturi as one integral piece. Each
one of these components plays an important role in the efficiency of the
jet-pump. Changing one of the components will induce change into other
parts of the system. But the pump by itself is not the sole ingredient.
There are many other factors that effect its performance including the
pumps vertical placement within the hull ("X" Dimension), the design and
displacement of the hull and how conducive it is to feeding the pump within
its given operational parameters and the powerplant that will drive the system.

Many factors within the pump will effect performance. Impeller RPM and
pitch, stator trajectory and length, venturi reduction and bowl area,
volumetric efficiency and laminar flow of the intake housing (including
laminar flow within the entire unit), "X" Dimension, length of travel and
compression (acceleration) ratios. Current jet pump design is no accident.
It is the culmination of many years of research, not unlike the modern
reciprocating engine or jet-turbine. What makes jet-pumps unique as a drive
system is their efficiency, reliability, safety and low drag factor. What
makes jet-pumps interesting to me is the plethora of designs that can
accomplish the same task. Again, much like the combustion engine. Consider
the numerous and diverse number of combustion engines and you have a good
idea of the potential of the jet-pump. (or what may have already been
developed!) ;-)

The jet pump has evolved as a drive system that is probably more valuable
today than its original designers anticipated. It is environmentally
friendly and litigation resistant. It is also produces much greater
efficiency within its given operational parameters in comparison to its
exposed propeller counterpart. For these reasons and more, the jet-pump is
considered by marine experts to be the drive system of the future, hence
its proliferation over the past several years into watercraft, mini-jet
boats, mega-yachts and commercial vessels. It has been the focus of my
tenure with contractors to the Defense Department for the last 15 years.

The origins of the jet pump can be traced back to Italy, where Riva
Calzoni produced the first model in 1932. By the late 1950's, functional
waterjets were in production by the late Sir William Hamilton and Dr.
Franco Castoldi. There has been much research put forward by well known
jet-pump manufacturers such as Berkley, Jacuzzi, American Turbine,
Hamilton-Stoddard, AmJet (American Hydro-Jet), and KaMeWa. In recent years,
several foreign companies have penetrated the commercial and military
market including Marine Jet Power and Kvaerner Eureka from Sweden, Ulstein
From Norway, And FF-Jet from Finland. Most of these companies have
concentrated on the refinement of conventional pump technologies. But there
have been quantum leaps made over the past decade by military contractors
including Pratt & Whitney, Boeing, Lockheed and Ameriquest Technologies.

If you assumed jet-pump applications were only suitable for small light
vessels such as personal watercraft and California style big-block jet
boats.... you're in for a rude awakening. Jet Pumps have become the drive
system of choice for the some of the largest vessels in production. In
almost every application, with the exception of very high speed surfacing
hulls, the jet pump is considerably more efficient and more powerful than
conventional drive systems. On this note, I've heard engine builders and
"gurus" within the watercraft market make statements about how inefficient
the jet pump is. To this statement I must ask.... compared to what??

MYTHS AND TRAITS:

There have been many misconceptions regarding pump efficiency, the least
of which has been pump blueprinting. Theoretically, a smooth finish within
the pump walls and stator vanes would dictate a less disruptive flow. In
reality, the water exiting the impeller blade has been disrupted and
aerated so much... that the surface within the pump does not adversely
effect thrust density or velocity to the degree that one might expect. A
typical die-cast finish, void of casting flash, is sufficient. In the
1980's, my group pioneered a concept of porous walls within a pump housing
to extract air (reduce aeration), thus increasing thrust density. This is
"blueprinting" in my book! Not the balogny that engine builders pass off to
naive consumers in the watercraft market, thus lining their wallets!

Some pump designs will attempt to "straighten" or "true" water trajectory
more so than need be. While it's the function of the stator section to
accomplish this, it need not be perfectly true. There is a point of
diminishing returns due to drag against accelerated water. A slight
spiralling of expelled water is acceptable. Further attempts to "true" its
trajectory accomplish nothing, but slow expulsion velocity. However, there
is value in altering the pitch (and size) of the stators at different
mixture density levels for increased efficiency. Because the mixture of air
to water increases with speed (due to ventilation), the stator section can
provide a way to increase the volumetric efficiency of the pump when less
water is present thus increasing or maintaining pressure as needed. But
again, in performing this task, other components must vary in conjunction.
More on this later...

The efficiency of a jet pump (at various speeds) is highly dictated by
intake and expulsion dimension, hence the well known variable... increasing
venturi orifice size results in more acceleration and decreasing the same
results in more top speed. But there is yet another variable....increasing
or decreasing the volumetric area of the intake gullet to maximize vacuum
at varying speeds. A good example of this would be the comparatively small
intake diameter of the Yamaha WaveRaider, which is conducive to high top
speeds because it creates higher vacuum at speed, which draws in more water
with less ventilation. On the flip side, this same intake gullet is not
conducive to the best acceleration because it lacks the ability to take in
sufficient water mass for the best acceleration. For a better understanding
of the principle, look at the performance characteristics of the specially
built Raider Race Hulls, which used a larger intake gullet. This gave them
competitive acceleration, but resulted in a much slower Raider.

AXIAL FLOW PUMPS:

The Axial Flow Pump is basically a direct drive system. It accelerates
flow on the horizontal plane of the impeller. It is merely an impeller
within a shroud that encapsulates water spiraling outward, and forces it to
go backward, through a set of straightening vanes within the pump known as
"stators". Stators "true" the trajectory of the spiraling water and create
a catapult effect, further increasing the velocity in which water exits the
impeller blades. Velocity is then further increased by the venturi prior to
expulsion (on all pumps). Axial Flow pumps work against the laws of
hydrodynamics to some extent. They control water more than working with
water. They do not really take advantage of centrifugal force like their
mixed flow counterparts. The axial flow pump works much like a conventional
marine propeller, while the mixed-flow is more like a turbine, it builds
pressure. Axial flow pumps lend themselves to watercraft installations
because of their overall smaller diameter and because they work well in
high RPM/low torque environments.

The Axial Flow pump is far less costly to manufacture and enjoys a variety
of aftermarket impellers to choose from at this time, therefore enabling
more rapid research & development of increasing engine displacements. As
engine size stabilizes among the major O.E.M.'s, more effort will be geared
towards the drive system.

CENTRIFUGAL FLOW PUMPS:

In contrast, the Centrifugal Flow pump capitalizes on the natural
direction water is traveling *after* it leaves the spiraling blades of an
impeller (outward). Likewise, the Mixed Flow principle is a hybrid of the
two logic's. Centrifugal (and Mixed-Flow) pumps take advantage of impeller
rotation, and thus the centrifugal force created by this rotation. They
offer a path of least resistance for water to travel when exiting the
impeller blades and rely more heavily on internal pump pressure created by
the impeller to create thrust. This type of system is torque driven, as
opposed to axial flows, which are more RPM driven. They rely on engine
torque to maintain adequate pressure within the pump and venturi. While
Centrifugal or Mixed Flow offer a path of least resistance for water
exiting the impeller blades, it also creates an indirect path to the
venturi that requires water to make an abrupt change of course to converge
within the venturi. It is by no means a direct system. But it has certain
advantages. Centrifugal Flow pumps build what is referred to as a "head" of
pressure, to the volume maintained within the pump and venturi, that can
give a split second more thrust when a loss of vacuum has taken place. In
other words, when your running across choppy water and encountering
cavitation due to air entering the intake cavity, the Centrifugal Flow unit
is holding a "reservoir" of water within the pump that it is still trying
to process. The downside to this is the increased time necessary to fill
the pump once the hull re-enters the water.

The Centrifugal Flow pump design incorporates more area within the pump
for acceleration, as opposed to an Axial Flow pump which is more direct
(remember, distance between two points is shortest in a straight line).
Because this type of pump makes water travel outwards before entering the
convergence cone (or venturi) a greater distance is realized and therefore
more area for acceleration is provided. It also incorporates a specific
design characteristic that has great value, and is only now being realized
by OEM and aftermarket manufacturers.... a gradual reduction in area
through the stators and convergence cone. The drawback to this style of
pump is the increased overall diameter of the pump, which is inevitable, to
allow water to "spread outward" for acceleration before expulsion.

MIXED FLOW PUMPS:

As previously indicated, the Mixed Flow Pump is a hybrid of the two
extreme design ideas found in Axial Flow and Centrifugal Flow Pumps. Mixed
flow is more direct than Centrifugal, but not as direct as Axial. It does
maintain some of the desirable characteristics of the Centrifugal pump,
such as increased area and working with water's natural flow and building
pressure, while still succeeding in delivering a relatively compact overall
size.

For reference purposes, Sea-Doos, Polaris's, Yamaha's and most other PWC's
on the market use Single Stage Axial Flow Pumps. Reportedly, Kawasaki has
adapted the Mixed-Flow on the new ZX series. This is not new for Kawasaki,
it utilized this configuration in early model stand-up Jet-Ski's.

STAGES:

There are pro's and con's to each design. (like everything in life) The
Single Stage (one impeller) pump is most common due to it's low
manufacturing costs and ultimate reliability. The Dual Stage (twin
impeller) is a more efficient unit, but is more complicated, costs more to
manufacture and is inherently heavier due to its increased parts count and
size.

A dual stage pump works somewhat like a transmission, but without gears.
Water is brought into and through the first impeller (stage) and
accelerated to a velocity conducive to the pitch of the second impeller
(stage). This substantially increases velocity over a single stage pump,
but there are inherent problems. When water exits the first stage, its flow
has been violently disrupted, therefore, the second stage is receiving an
aerated mixture of water, thus reducing the efficiency of the second stage.
This is compensated for by the increased pressure that the dual stage pump
creates as a complete unit. This can also be compensated for with flow
feeding by-passes, to help saturate the last stage, but there are other
complications in doing this, such as varying density levels at different
power levels. There are many ways to refine these types of drive systems,
but but it is not the subject of this article and these drive systems may
be antiquated based on evolving new technologies. (no comment) Also, the
rotation (direction) and the speed (rpm) and the stagger (size, pitch,
placement) of the impellers within a dual stage pump is critical and
variations will effect other areas of the system. There is another stage in
pumps that is not in-line, like a dual stage system. I hope to share this
with you in the future. (again, no further comment)

Pumps have plus's and minus's in comparison to "exposed" propulsion
systems. They are inherently inefficient at higher speeds when placed in a
recessed setting within the hull due to ventilation or air induction. There
are ways to make pumps more efficient, but this would require an all new
hull design to take advantage of surface piercing technology, the placement
of a specifically designed pump beneath the planing surface of the hull.
Pumps are *BY FAR* more efficient at producing thrust, as long as the
ventilation factor is removed from the equation, such as low to medium
speed operation. The differences between pumps are the applications for
their use and the power plants that will drive them. There are many
variations available in pump designs that will contribute to their
effectiveness in a given application.

CURRENT VARIATIONS:

The most significant variation in the Axial Flow pump for personal
watercraft vehicles is the incorporation of reduced area through the stator
section by increasing hub diameter. This essentially increases the velocity
though the stator section and reduces the amount of water needed to keep
the venturi primed. Whilw this will give a definate increase in vessel
speed, it may lack adequate mass for a level of acceleration we've become
accustomed to.

Another trend is larger diameter pumps. You can expect them to proliferate
as the increased weight of today's current PWC offerings continues to
escalate. Reason? If you increase hull weight (mass) you need more water
mass to offset the loss in acceleration due to the increased hull weight.
Bombardier recognized this fact and has addressed it with the new GSX
Limited, which features a 155mm inside pump diameter, as opposed to the
original 139mm pumps on all Sea Doo models to this date. A 16mm increase in
size may seem rather minute, but consider that a 1mm increase or decrease
in a venturi orifice makes a significant effect. Now you can appreciate how
sensitive pumps are to change. Same with props. One or two degrees equates
to fractions of an inch, but it can make or break performance.

Another significant design adaptation is Bombardier's half-hearted attempt
at surface piercing pump placement, as witnessed by the new GSX, GTX and
97XP models. In effect, if they had kept the original flat pad Sea Doo
keel, such as the runabout line, the placement of the pump in the new
models reflects nearly a two inch drop below original Sea Doo pump
locations. Because of the new deep "V" hull design, it appears the pump is
still in the same place relevant to the keel. In reality, it is much deeper
than previous models.

MATERIALS:

The most common materials used to produce pumps are aluminum, stainless
steel, bronze, and composites (such as graphite re-inforced nylons). The
difference in these material are wear factors, resistence to the elements,
weight reduction, construction wall thickness and cost to produce. Of these
materials, composite is clearly outstanding. It is uneffected by the
elements, it is light weight, it can be constructed with wall thickness
equivelent to aluminum (due to its geometric configuration) and per piece
cost is significantly lower (in quantities that parallel the PWC market).
Certain composites offer yet another commodity that their metal based
counterparts cannot, the ability to flex. This alone makes them the
material of choice for the future of jet-pumps.

THE FUTURE:

The future of the jet-pump has never been brighter, due to its wide
acceptance by manufacturer's that have the financial means for further
development and the incentive to do so. Ironically, much of the research
that is now being conducted by certain OEM's is merely a duplication of
previous work, but they need to find out for themselves. Actually, some of
the research that I've witnessed is quite archaic, in such as they are
using tools and measurements not unlike racers, the aftermarket and engine
builders have developed for drawing their own conclusions. There is an
enormous amount of data accumulated by purpose built marine/naval oriented
research facilities that should ensure information relevant to the design
and advancement of jet-pumps in the near future. Within the not so distant
future, expect an entirely new concept in pumps to be adapted into your
favorite personal watercraft. That's all for now.

FORMULAS:

For those interested, the calculations (or math) is basic, but is somewhat
complicated and diluted by the additional factors of drag, displacement,
weight, etc.

The following legend is for reference purposes....

T = Jet Thrust
Vb = Boat Velocity
m = Mass Flow
Vj = Jet Stream Velocity
HPi = Input Horsepower
EHP = Effective Horsepower
n = Number of Jets
EHPj = Jet Effective Horsepower
nj = Jet Efficiency
np = Propulsive Efficiency

The jet propulsion principle is the acceleration of mass flow through a
nozzle. It is represented by the equation:

T = m x Vj (Thrust (T) equals mass flow (m) x jet exit velocity (Vj).)

Water jet propulsion works on the same principle except that it does not
carry its water mass. Instead, it is drawn through an intake on the hull's
bottom. The penalty for this is that water has to be accelerated to the
vessels speed in order to be expelled at a greater speed. Thus we begin
one of many more new equations:

T = m x (Vj - Vb) (Thrust (T) equals mass (m) minus boat velocity (Vb).)

The EFFECTIVE Horsepower (EHP) of a given vessel at a given speed is
calculated as Hull Drag X Speed. The efficiency of the propulsion system
(NP) of the vessel is measured by dividing the EHP by the actual power
delivered by the engine (Hpi) multiplied by the number of engines and jets (n):

EHP
NP = --------
n X Hpi

The same formula can be applied to measure the efficiency of a given jet drive:

EHPj = T x Vb

and the efficiency equals:

T x Vb
Nj = ------------
Hpi

The formula for balancing the units of measurement is:

T(lbs) x Vb (KTS)
Nj = -------------------------
Hpi x 323

I hate math.

Carl Camper, President
ULTRAC Performance Systems
http://www.interlog.com/~jlogan/ultrac
1500 N.W. 62nd Street, #510
Ft. Lauderdale, FL 33309
954-351-1943 / 1944 fax
1-888-ULTRAC-1


0 new messages