Why A Jet Drive?

There is boundary layer information provided in JavaFoil. Again it has to be taken as trends rather than absolute because it is only 2D and the pressure gradients are higher than in the 3D case.

The attached compares the boundary layer for different Re#. You can see there is flow separation at the tail occurring as the Re# reduces. I expect that separation would be similar for 3D but at slightly higher Re# than the 2D case.

At the design speed for a 500kg vessel of 22% thickness the boundary layer is approximately 1 mm thick where intake ports would be located. It becomes thicker when separation occurs at lower Re#.

Maybe Tom can ask Martin Hepperle to produce JavaBoR so he can optimise his shape. The underlying maths is long lost for me.

The last image is an approximation of the lowest drag form for Re# around 10E7 and JavaProp correlates well with this. Again the result relies on stable flow conditions not ducking and diving. You will see the Cda is quite a lot lower than the previous shapes. This has a maximum thickness of 30%.

The Cdv for this shape is about 0.008. For 500kg at 8.2m/s this give drag of 170N. Using the 2D Cda from JavaFoil the drag is 300N. The 170N is lower than you see in the real world but then there are no hulls or marine shapes that approximate this. The only laminar flow hulls that have been built that I know of are in the lower Re# range and these are finer with thickness around 20% or a bit lower.

The laminar flow hull shape for Re# above 10E8 is not dissimilar to the shape of large whales so provides some credence to the low drag of these shapes. All the laminar flow hulls aim to provide a pressure gradient that shifts the transition as far aft as possible. At low Re# the maximum thickness is behind centre but moves forward at higher Re#.

Rick W

 

I'm afraid that a boundary layer thickness of 1 mm on a vessel moving with a Re of 1.6 to 1.8 E7, and a length of 2-2.5 m is not near a realistic value. We should see something like 28 mm of turbulent thickness in the rear part of the body, and the shape of the boundary streamline is not congruent with what is observed on similar bodyshapes in windtunnels. See f.e. Hoerner or Schlichting. The streamlines appear all too much alike the theoretical non-viscous source-sink flow.

So unfortunately there is either a serious bug in your software, or a serious handling error. In either case the output is of no use for prediction purposes, it may in fact lead to dangerous mistakes! Before someone is getting in serious trouble by using ANY output from this source, the error must be localized!

 

Hmm

one could say that was the more conveniant sort of: fuc.k this idiots playstation "advice".

 

Rick W good morning
Details of submarine propulsion systems are not easily obtainable ,for obvious reasons ,but if you Google "Rolls Royce submarine water jets" you will see that the latest British submarine ,one of three to be built ,is powered by a new version their water jet .although it's main claim to fame will probably be the end of "hot bunking". I accept of course that you are quite right about the superior efficiency of the propeller in this application.

 

Tom, one more word of warning when it comes to comparing the drag of different shapes. When the functional criterion is a specified VOLUME, the performance must be compared in terms of "drag/volume", not drag per transverse section area or total surface area, as done above!!!

In the examples, with a diameter/length ratio of 0.22, and your diameter target of 0.45 m, the enclosed volume of a body of revolution with such a shape would be less than your specified 0.3 m3, causing this sub never to surface again, once launched. To reach your volume target, the overall dimensions of the craft have to be increased, simultaneously increasing surface area and thus final drag.

This is not to say that the short low-drag shapes are out of question per se, but the drag with your volume capacity will be greater than indicated just by the calculated drag figures shown.

A more pragmatic comment: those shapes with a continuous curvature midships are extremely sensitive to longitudinal cog for trim in surface mode. This is a good reason for having a cylindrical midships section, connecting a spherical or ellipsoid nose with a conical rear.

--. -- -.--

 

Confused...

Baeckmo,

I never specified a volume of 0.3 cubic meters. Where did you get that number from?

I stipulated I am not interested in surface operation so I'm not sure why you're addressing it.

If you find Cliff's R300 to be of less than optimal shape, what shape would consider better?

Can you see any advantages in using a jet drive for my sub?

What about using a larger diameter jet?

-Tom

 

Hi Tom, been gone for a week....now back to harass ya! In your note #15 you specify a weight of 300 kg. In order to float those kg's your hull has to displace at least 300 l of water. That is 0.3 m3. Unless you plan to leave the sub parked onto the sea bottom and never reach the surface again, the total volume of your vessel must be greater! Otherwise you have to winch her home!!!!

See why I adressed the subject ......?

There is no way a waterjet of this power can get you even close to the efficiency (=operational range) of a decent propeller.

 

Welcome back baeckmo,

Longitudinal cogging is not a concern as there will be minimal surface operation, no need to address it.

Objects that are neutrally buoyant don't sink to the bottom nor do I intend to park on the bottom.

In sea water, a 300kg vessel with a volume of 0.3 m3 will have positive buoyancy, but don't concern yourself with this.

No dialogue on your opinion that the R300 could have a more optimal shape? Like what?

You have no comment on ANY other advantages a jet might have in my application over a prop?

That was my original query.

All the best,

Tom

 

OK, guess you have control over "floatability". I mentioned it more to show that there is no general "best shape", but that optimisation will follow different paths, depending on the object or function that is to be contained. That said, if the enclosed volume is the deciding parameter, a l/d ratio of ~5.5 to 6 is near optimum. Optimum shapes were studied when airships were developed around 1910........ See fex a classic: G. Fuhrmann, diss Göttingen 1910, (=Jahrbuch d Motorluftschiff-Studienges. 1911/12 pp 63).

The usual reasons to use a confined propulsor (waterjet) in the low speed range, where its propulsion efficiency is lower than a corresponding open propeller are:

1/ No rotating parts outside the pump housing; less risk for damage.
2/ Excellent maneouvrability with nozzle moving in two dimensions and split duct reverse bucket. However, the r-bucket is difficult to "hide" from flow in fwd position in a submerged vessel, and may cause very high drag.
3/ Wj rotor may work with higher rpms than its free prop cousin, simplifying transmission if main engine or motor is high speed. Note that wj's often fail due to cavitation caused by too high rpms, though!
4/ Pressure wave (=sound) radiation can be significantly lower than with a free prop

 

baeckmo, thank you. I have an expanding number of specific questions

before I will be content to leave this research. I sense you are not

interested in specifics but knowing where I'm going with this, is there a

particular book (or books) you would recommend that may help me answer

my ponderings? Diameter, RPM, inlet designs, torque, matching drag/

velocity/thrust curves optimally, etc, etc.

-Tom

 

baeckmo,

Okay, so you don't want to share that information with me, that's okay.

What do you think of a nose inlet with internal channelling to a larger

impeller and a matching larger nozzle?

Positive feed pressure could be achieved, increased efficiency from a

larger, longer impeller. Motor cooling would be made available.

What is considered "high" RPM which could induce cavitation?

How large an impeller would be needed to hit my 500 Newton bogey with

8kw? What kind of prespinning benefits might be gained with directional

vanes in the gullet?

-Tom

 

Sorry Tom, don't get pissed; its not that I don't want to continue this communication!!! Its just that right now I have a tight schedule with deliveries for those who pay my invoices, so please have some patience!

There is no "Popular waterjet design manual" as far as I know. It is a question of blending turbomachine fluid mechanics with ship hydrodynamics in a double optimizing process, so you need an understanding in both disciplines. I'll prepare some additional info on inlets and efficiency limits later, but probably no time to do so before X-mas.

 

No problem baeckmo, I'm not pissed.

It's just that I've noticed you're not really into dialogue on this thread

but chatty on other threads. Proirities I guess.

Have a good Christmas.

-Tom

 

baeckmo,

Are you really coming back to my questions?

If not, please let me know, don't just leave me hanging.

If you don't have anything to add, or don't know the answers, let me know.

How was your Christmas?

-Tom

 

Yep, Tom I am preparing some sketches for you plus some answers; I'll be back in due time with those. Don't feel "betrayed" because I dealt shortly with the pwc jet issue; there is in fact a common wisdom regarding inlet area sizing.

I am also involved with a research project on the cold adaptation of dogs, and we have now an intensive data collection period, that keeps me away from much of the waterjet fiddling right now.

And thanks, still topped up with good stuff from the X-mas menue now that we prepare for the new year. So:

Happy New Year to you!!