Surface drive with large diameter propeller

Discussion in 'Boat Design' started by Barry, Feb 29, 2024.

  1. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Hi Barry, you opened quite a wide “window” regarding design SOR, so the answer can not cover details. If you have a specific task in mind, I’d be happy to assist. In the late 1970:ies I dug quite deep into the pro’s and con’s of various propulsors for a project around workboat design, including local commuters.

    Long story short; I collected as much open info, reasonably validated, that could be found, including some very good Russian data on some of their hydrofoil propulsions. The best comparison format for preliminary design evaluation is the kt/J^2 form, where kt is the usual thrust coefficient and J is the advance coefficient. After some arithmetics, it boils down to: T/(rho*D^2*Va^2); where T is thrust in N, rho is fluid density in kg/m^3, D is propeller diameter in m and Va is the advance speed in m/s.

    As you can see, the shaft rotational speed is eliminated and the prop dia remains as the main design driving factor. If you look closer at the formula, you will see that although some constants are omitted, it actually describes a limiting pressure (thrust/disc area) in relation to (rho*Va^2), which refers to dynamic pressure due to the forward speed. This means that it is based on true first principles of physics and can be used for further analysis.

    So, I made a rude classification of prop types as Submerged, Submerged supercavitating, Surface piercing fixed and Surface piercing articulated. The submerged includes both straight shaft and (most) I/O. The two versions of SPP’s should rather be called “short” and “long”, respectively, since the prop position in relation to the transom is important for acceleration and operation in low-to-medium speed modes (more on that later on).

    Enclosed, you find a graph showing kt/J^2 over advance speed in knots. It includes data from some 150-ish different vessels and tests and in spite of the “wide sweep” in the collection, there are three quite distinct trends to be seen. For a preliminary design estimate, this format seems to give reasonable results, provided you avoid values in the upper quartile for each prop type. Note that even a slight ventilation calls for a blade section profile with pressure surface camber in some form; standard profiles simply don’t work!

    To make the comparison universally non-dimensional, I have also plotted kt/J^2 over propeller Froude number [Fnp=Va/sqr(g*D)] in the second diagram. Here, the spreading between the two ventilated types is smaller, indicating a common behaviour that is not obvious in the first graph. In both images, I have hidden the Submerged supercavitating results up my sleeve, since we are now doing some development along those lines.

    Now, hitting the ballpark for a design point is the easy part; making it work over the full speed envelope with realistic hull drag is the really tricky business, and I’ll have to come back to that later.

    upload_2024-3-5_20-14-13.png

    upload_2024-3-5_20-15-12.png

    Edit: BTW is there anyone out there who has access to old EMB-reports? For my personal curiosity I tried, but failed, to find the two reports on "Hickman type propeller" that were tested there:

    "Tests of Partially Submerged Propellers of the Sea-Sled Type", EMB report R-132, April 1919.

    And:
    "Tests of Partially Submerged Propellers of the Sea-Sled Type" , EMB report R-136, April 1932.
     
  2. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    @baeckmo posts and it's like Christmas!
     
  3. jehardiman
    Joined: Aug 2004
    Posts: 3,899
    Likes: 1,265, Points: 113, Legacy Rep: 2040
    Location: Port Orchard, Washington, USA

    jehardiman Senior Member

    Hummm... since you reduced it to a function of D:Va then for a given geosim we should be able to expand the 2D graph line into a 3D volume with a D axis......giving a "cliff edge" where cavitation or bending moment takes over for a given D:Va (a concept very similar to a propeller calculator program I wrote once)..... I see a paper waiting to write itself....or bringing back the old Prohaska Log propeller charts.
     
  4. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    I must confess I'm having a hard time interpreting the graphs. "Potens" are?
     
  5. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Sorry about that; it refers to the trendlines for the three prop types. A simple curve fit of the form Y=const*X^(exponent). Could probably have found better fit to the data clouds, but they serve their purpose at this stage.

    So, if you check the first Hickman double screw installation, let's assume that he lost 1 hp in the cogs, that leaves 8hp (ie 5900 W) to each prop. Speed achived ~8 m/s (~16 knots), propeller dia 0,56 m, propeller efficiency guesstimate 55 %. Thrust is then 5900*0.55/8; ie 406 N/propeller. Then kt/J^2 = 406/(1000*0.56^2*8^2); result ~0.02. This is a much lower propeller load than we use today, but it certainly showed the way forward. There has been mention about Hickman's use of cambered blades, but I think that it might not have been overly refined at that first experiment.

    Plot the resulting value (0.02) along the 16 knot speed line, and you have the comparison between the "original" and later developments. My intent here is to respond to Barry's post, and show the variation of diameter as a function of thrust and speed. As you can see, ventilated prop diameters, particularly the "short" installations, have to be considerably larger than submerged props when design speed goes down, even with modern blade section profiles.

    Edit; One reason to use this format is that in many cases when you find info on a prop installation in an existing vessel, there is no info on shaft speed. But Hickman provided a number on that; ~850 rpm (~14 rps). That gives an advance coefficient of 1, which is a very reasonable value, even with today's standards.
     
    Last edited: Mar 6, 2024
    BlueBell and DogCavalry like this.
  6. philSweet
    Joined: May 2008
    Posts: 2,790
    Likes: 548, Points: 113, Legacy Rep: 1082
    Location: Beaufort, SC and H'ville, NC

    philSweet Senior Member

    Shouldn't there be some area correction for the surface piercing props? I think I used an arbitrary 40% for fixed surface props intended to run with the shaft dry. I remember going though the entire exercise of rederiving the standard prop curve analysis in light of variable submergence, but it was ages ago.
     
  7. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

  8. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    Once upon a time I was obsessed with helicopter design. I worked the numbers compulsively, but you can't get past the fact that thrust is momentum transfer, which is to say proportional to mv, but to change v of a given m requires power proportional to mv². A merciless relationship but definitely explains Hickman's 1911 performance. Move more mass, with less acceleration if you want efficiency. The ducted fan guys had a mantra: more mass flow.

    But of course the increase in kq points in the opposite direction.
     
    Last edited: Mar 7, 2024
  9. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    Actually that would be a worthwhile comparison: kt vs kq. I asked @gonzo a few years back where increasing prop diameter stopped being worthwhile.
     
  10. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Oh yes, that is the next step, figuring out the "effective working area", which is not as simple as it looks. But again when looking for a trend in the raw data, what you have is speed and dia. It has been shown that as long as you can define the effective area, the "data cloud" gets denser. Now, this is the main mechanism explaining the seemingly higher propeller loadings with the "long" ("articulated" in the diagram); they often operate slightly deeper, ie with a larger proportion of the disc area submerged.

    Getting a grip on that difference was one of the reasons to do the research back then.
     
    DogCavalry likes this.
  11. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Naah, you got to be careful there: Move more mass with less velocity increase, if you want efficiency!
     
    Barry, BlueBell, jehardiman and 3 others like this.
  12. portacruise
    Joined: Jun 2009
    Posts: 1,498
    Likes: 187, Points: 63, Legacy Rep: 218
    Location: USA

    portacruise Senior Member

    Curious if having more mass (and thus momentum) is helpful with efficiency with everything else (except maybe hull shape) being the same, when well below hull speed?
     
  13. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Yah, I like that logic, I think a 3D graph would be more instructive than Prohaska, since you can have, say D, V, and cavitation sigma (or shaft submergence et c.) on the graph sides and then blade bending stress as colour bands on the resulting 3D surface.

    Edit; Donald Blount and David Fox wrote some papers on formats for propeller data back in the -70-ies that have inspired, and also the original paper by Newton and Rader on "their" high speed prop still holds value, and may give hints on how to present the additional "dimension" of submergence.

    Blount et Fox; "Design Considerations for Propellers in a Cavitating Environment"; Marine Technology, Vol 15, April 1978.

    R.N. Newton, H.P. Rader: "Performance Data of Propellers for High-Speed Craft";
    RINA, Quarterly Transactions, April 1961, Vol 103.
     
    Last edited: Mar 7, 2024
  14. baeckmo
    Joined: Jun 2009
    Posts: 1,726
    Likes: 738, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    What, exactly is "hull speed", and why would it change the impulse mechanics?
     
    Barry, jehardiman and portacruise like this.

  15. DogCavalry
    Joined: Sep 2019
    Posts: 3,309
    Likes: 1,668, Points: 113
    Location: Vancouver bc

    DogCavalry Senior Member

    Corrected
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.