To Taper Bore or Not to Taper Bore the Reduction Nozzle - An Analytical Approach

I posted this over on PWC today in the Kawi section, but I thought I would post it here too. Just thought I would share this information. Take it for what it is worth.


The age question of whether one should taper bore or straight bore the exit of the reduction nozzle has been around for a while. In an attempt answer this question, and understand better the optimum nozzle configuration for a particular setup, I put together a simulation model. At this time the model does not take into account the added pressure on the pump intake due to the watercraft moving, however for an A to B comparison, the values should still be valid. I should be able to also look at pump cone angles too, however it was held constant for this investigation.

The simulation I put together includes an engine (with an estimated torque vs. rpm curve of an kawi 1100), pump, and reduction nozzle geometry. The model I created adjusts itself based on pump pressure (and pump drive torque) and engine torque until an equilibrium is met. So you if you incorporate a smaller nozzle, you can really see peak engine rpm decrease. Then looking at the outputs of the simulation, one could compare engine rpm, mass flow, exit velocity, and resultant thrust. I still need to validate the model with some actual test data, but it seems to make sense with a 1mm change in diameter making about a 100-200 rpm change depending on where you are at and whether you are using a straight vs. taper nozzle.

Jet pumps work by the means of change in momentum, so (mass flow)x(velocity out -velocity in) = Thrust force. The greater the peak thrust force, the faster you will ultimately be able to go since you will have more force to overcome drag. You are also able to look at how quickly the thrust force increases, which links to acceleration.

Conclusion
The conclusion I have come to, is the taper at the end of the nozzle decreases jet pump thrust. Holding all other things constant, having a straight bore for the final exit of the provides greater thrust. The simulation suggest the net thrust improvement will be in the range of 4-5% between the optimum 1.5deg taper bore and the optimum straight bore.

Some other points:
  • Kawasaki and Yamaha high pressure die cast these parts. They need to have a certain amount of draft in the casting in order to pull the part from the mold. Coincidence or not, most of the high pressure die cast aluminum parts I have worked with over the last several years have all called out a nominal 1.5 deg of draft. I have not measure it, but I have read 1.5deg is what the end diameter is at on the Kawi nozzles.
  • The optimal nozzle exit diameter will be different based on the nozzle taper angle. For example: To achieve the greatest thrust at max rpm(that the engine can spin at that pump power) - the nozzle configuration for a straight nozzle may be 82mm, however for a 1.5 degree taper nozzle, it may be 83mm. The net thrust at with the taper nozzle at 83mm will be still less than the thrust with straight at 82mm though. The simulation suggests you will always need a larger taper bore than an equivalent straight bore to reach the optimum point for each design. The nozzle size will still play a role in RPM and hookup, but the taper has shown to not be beneficial in all cases I have evaluated.
  • Like I said above the nozzle works from change in momentum. However for a higher efficiency pump system, you desire to have the greatest mass flow/velocity combo going parallel to the axis of desired motion. This is why the pump has stator veins. It is an attempt to “straighten” the water and direct it so more of it is going straight out the back. Without the stator, a great of the waters velocity will be spinning and not helping provide thrust in the forward direction. So saying this, having a taper at the end of the nozzle is directing water in a direction which is not parallel to the axis of desired motion. The outer most water in the nozzle is also crashing into the inner most water actually going more parallel.
  • As the aeration level increases in the pump, the taper nozzle performance becomes closer to the straight nozzle performance.

So hopefully some of what I said here makes sense. I could, of course, be missing some things in this simulation, but I have been working in the hydraulics field for quite some time and my expertise is in hydraulic, pneumatic, and mechanical simulation. Anyways, I thought I would share this work since it ultimately means nozzles are easier to bore, and of course you get more out of your machine.
 
I wish I could sit down with you at look at your simulations! Sounds like you put a lot of time into this and I appreciate that.
Can you play with pump vane numbers, lengths, angles, etc? And are the vanes in the reduction nozzle helpful or a hindrance?
Thanks!
 
  • Like
Reactions: DAG

DAG

Yes, my balls tickled from that landing
Location
Charlotte, NC
I wish I could sit down with you at look at your simulations! Sounds like you put a lot of time into this and I appreciate that.
Can you play with pump vane numbers, lengths, angles, etc? And are the vanes in the reduction nozzle helpful or a hindrance?
Thanks!
I would like to see this also! Is it the vanes or just the large hub of a mag that makes the difference we all can feel and enjoy. My money is on the hub diameter not the vane count
 
The simulation was put together to mainly focus on the reduction nozzle investigation. It mainly looks at the impeller, pump, and engine torque curve being constant and looking at what the expected changes happen due to the nozzle. I can use this to help understand where I should go with the nozzle. It actually was pretty quick to look together.

In order to get more detail on the pump geometry, a much more complex model needs to be completed. I have been thinking about looking at working on a much more detailed pump model, however it will probably not include the ability to include the engine response as well.


Here are some other answers from my experience and not simulation.

The vanes in the reduction nozzle will help continue to straighten the flow coming out of the pump. They will tend to increase pump pressure (compared to nozzle without vanes), however they should help direct more of the flow in the direction parallel to desired motion. You do want the vanes to match up with the vanes in the pump to help prevent any additional pressure drop. I am sure if you could get the vanes in the nozzle thin enough, like you get in a mag pump, they would be worth it on a 6 vane pump. They will probably also be better if you can have them butt right up to the pump vanes. On anything more than 6 vane they will probably only add more drag. On a 12 vane pump, you more vanes to separate and straighten flow. The vanes are blocking more of the available area, so the pump pressure is higher. This is generally why higher HP applications go with 12 vane. They can counteract the extra pressure to gain the efficiency from the 12 vanes straightening more water.

For the benefit of the magnum pump, it is should be a combination of the hub diameter to increase velocity and build pressure, more vanes to help separate and straighten the flow, and much thinner vanes to interfere less with the flow. So essentially the hub helps with keeping pressure in the pump (hookup), and the vanes (number and thickness) help increase the efficiency pump and increases thrust. You should notice this in actuality too. The pump stuffer option gets you the improvement of the magnum in terms of hookup, however the vane configuration will help with increasing thrust and peak speed.
 
  • Like
Reactions: DAG

waxhead

wannabe backflipper
Location
gold coast
Wow interesting I know a little about jet units but nothing on the scale that you have just written
In your simulation can you add a cross at the back of the reduction nossle or at least some straightening fins that go to the end. O fcourse you would have to keep them thing to reduce drag but im wondering if the straighter you get the water the more thrust you get or does it drop off at the end.
Of course there is going to be a cutoff point where the extra drag is going to over come the gains of more straigtening fins. I would also be very keen to see how moving the prop closer to the stator effects it in regards to efficency.
Great thread
 
Most likely you will have a case of diminishing returns to get you to the peak efficiency point. Then the efficiency would drop if you keep on going. You have to balance pressure (and resultant pump power) from the vanes with the added benefit you get from a greater portion of the pump water velocity going parallel to the desired axis of motion. Where that exact point is I am not sure.

The closer the impeller gets to the stator vanes should help increase efficiency by having water loose less energy by the water spinning. The longer cylinder of water you have between the rear of the impeller and front of the stator veins, you have more potential for the impeller to really get the water spinning inside the pump. The more the water is spinning, the more energy is lost while the water travels through the pump to get to the vanes. The water is not only traveling linearly, but is also traveling circumstantially. You also have the added benefit of having less volume. Pressurizing a smaller volume takes less time.

In case of the the simulation, with what I have set up right now, I am not able to evaluate flow characteristics of specific changes to the flow path. I would need to use a different modeling software to model the exact flow path in 3D. I am looking into being able to do that, I just have not yet.
 
Most likely you will have a case of diminishing returns to get you to the peak efficiency point. Then the efficiency would drop if you keep on going. You have to balance pressure (and resultant pump power) from the vanes with the added benefit you get from a greater portion of the pump water velocity going parallel to the desired axis of motion. Where that exact point is I am not sure.

The closer the impeller gets to the stator vanes should help increase efficiency by having water loose less energy by the water spinning. The longer cylinder of water you have between the rear of the impeller and front of the stator veins, you have more potential for the impeller to really get the water spinning inside the pump. The more the water is spinning, the more energy is lost while the water travels through the pump to get to the vanes. The water is not only traveling linearly, but is also traveling circumstantially. You also have the added benefit of having less volume. Pressurizing a smaller volume takes less time.

In case of the the simulation, with what I have set up right now, I am not able to evaluate flow characteristics of specific changes to the flow path. I would need to use a different modeling software to model the exact flow path in 3D. I am looking into being able to do that, I just have not yet.

Good stuff. What software are you using? I downloaded some freeware program last year, but I had alot of difficulty figuring it out.
 
theres a good reason skat and solas both use 12 or more vanes in mag pumps for higher hp.I spoke to skat about putting the prop very close to the stator,they found disadvantages in doing so for high rpm applications but no testing done for freestyle aps. In terms of straightening water,ive tried the yami 60E turn nozzle with the 4 stator veins in it. what is your opinion on that? I have not tested it much yet. There are not many turn nozzles using vanes,makes me wonder what the yami engineers where thinking when they used these?
 
Vanes in the turn/steering nozzle should improve steering depending on how much the water is spinning leaving the reduction nozzle. The vanes will help straighten the water (like the stator vanes do), but in this case it is helping to increase the velocity of the water going in line with the axis of the steering nozzle. This increase in velocity (and resultant increase in change in momentum) will increase the steering thrust.
 
Skat has a reduction nozzle with no Vanes at all .. In speaking with them they seem to over simplify the lack of need for them. Some of the skat pumps veins don't even line up correctly with veins in reduction nozzles.
 
Last edited:

Quinc

Buy a Superjet
Location
California
Skat has a reduction nozzle with no Vanessa all In speaking with them they seem to over simplify the lack of need for them. Some of the skat pumps veins don't even line up correctly with veins in reduction nozzles.

I am sure Skat is telling that is a "feature" right?
 
i cant remember what they said about why it was made without them,just that much testing was done and no noticeable diff found rather veins in nozzle or not. Wonder what type testing was done
 
Top Bottom