Need help with drive shaft design

My boat project is a 11 ft L x 48"W planing hull micro-skiff, it is a Carbon fiber/Basalt fiber skinned foam blank. I have decided to build an electric surface drive I will be using a Maytech, MTI 120116 brush-less motor that puts out 18.8 KW max or about 25hp. I am building a motor and shaft unit that
pivots up and down to change the angle and depth of the prop and use a Levi drive type shroud/rudder for steering. The drive shaft will be 48" long or possibly a little longer if needed.

I am now at the point where I could use some advice on shaft design. The motor comes with a 15 mm shaft that I would have to couple to. I am considering a stainless tube for the drive shaft to run through and either bushings or bearings to reduce down to the drive shaft size.
What type of bushings or bearing would be best ?
Should I have one bearing at the top and one at the bottom end of the shaft or have a bearing in the middle?
Should I be using a cutlass bearing at the prop end of the shaft?
Should I use a thrust bearing at the motor end to keep from putting pressure on the internal motor bearings?
I have seen electric drive set ups for different applications where they used a curved jaw coupling to alleviate vibration and alignment issues, would that be advisable? Or should I consider 2 of these couplings in the drive design?
I also need to decide the prop end design, splined or threaded for prop attachment?

I'm pretty good on boat design and construction but I am pretty clueless about this shaft design problem and I am doing a lot of experimental stuff in my overall design and want to keep error in my drive trial and error to a minimum.

 

Not that relevant to your question, but what is supplying this 25hp electric motor ? An 11 x 4 boat won't carry much battery storage.

 

Welcome to the Forum JB.
This sounds like an interesting project - however you never get a free lunch with electric power - as Mr E says, you need to get the oomph for the electric motor from somewhere.
I am also intrigued by the 15 mm (diameter?) shaft on the electric motor - is this a typo error?
I do not see how you can transmit 18 kw of power through such a narrow shaft?

 

Surface piercing propellers are very sensitive to thrust loading and speed. Your approach is backwards engineering. It does not work in this case; you must have more facts on hull weight, resistance and attainable speed before you can specify anything about the drive. Thrust at hump and max speed determines the prop dia. Power and prop dia determines rpm. Power and rpm determines torque. Torque and material determines shaft diameter. Shaft diameter, torque and shaft length determines bearing distances. And, yes, unless the motor bearings are dimensioned for axial load you must have a thrust bearing. There is a homework to be done; with that finished, we may come up with realistic advice.

 

Thanks for the responses let me address the questions posed.
Mr Efficiency - The recommended power is 10s to 16s Lipo, I do not plan on running this at nearly 18+ KW or 25hp I think I can run over 20 kts. at about 8hp. I elected to switch from a 6hp outboard to electric and about 16 kts max would be expected from the outboard. This boat is designed to be used as a fishing platform and to be car topped or trailered and launched as you would a canoe or kayak and to fish within 5 miles of the launch point. I will be slow cruising and site casting shallow flats and marshes at very low speeds so the battery capacity will be designed to meet these parameters. I realize to run at full power for lengths of time I would need huge battery capacity and the heat loads would be another big problem. At full speed with the right battery set up I could be hitting 40 kts but it is a tiny boat and running at those speeds is not what the boat is configured for.
bajansailor - That is not a typo the motor is delivered with a 15 mm shaft. It is a tiny motor for is claimed performance, it will fit in a 5 inch cube, one of the primary uses of these is are electric surfboards and E-foil boards I think most of these applications use the submersible little brothers to this one which run 7 to 10 KW of power.


baeckmo- The boat hull will weigh less than 100lbs but fully loaded , person, gear batteries etc. should be just under 450lbs, it will draft about 3.5 inches at rest. The shaft will need to be 48" in length to meet my goal of 5 deg. with prop shaft at the surface it could be longer but shorter would put the angle higher. The motor and boat as designed would have potential speed of 40 kts + but for my purposes a cruising speed of 20 kts would be good. The motor puts out 20 N*M of torque at 60% capacity so around 11 hp. and I believe that 5700 rpm will be attainable when loaded from looking at their literature. The maximum spin speed for the motor is 13k RPM with maximum voltage and no load.
The hull is shaped and ready for glass so those parameters are set, the motor housing unit is built and is ready for glass, and I have chosen the motor so I am well off the drawing board. I know prop sizing prop shape shaft angle etc. will be very much trial and error but I have to start somewhere and the mechanics of the drive shaft are next on my list. I could follow the long-tail mud motors and the surface drive mud motors in this area, they have a lots of hours of use and are well tested.

 

So, let me get this clarified - you are hoping to propel a 450 lb boat at 20 knots with an engine driving a propeller shaft that is 48" long, and 15 mm (a bit less than 3/4") in diameter?
Have you thought about the torque on this shaft?

 

Draft 3.5" ? So a small flat bottom punt ? I think some pictures of this who arrangement are needed, when you say the shaft is 4 feet, one assumes it does not project much behind the boat, or the steering set-up will get complicated.

 

The motor puts out about 18 kw and 20N.m of torque max and the motor shaft is 15mm the drive shaft can be larger but the motor shaft is by the manufacturer. These motors are made for marine use, one of the smaller motors in this line is 9.5kw with a 12mm shaft and is shipped with a prop up to 7.7" x 7 which is commonly used for E-Foil boards and electric surf boards and is meant to have the prop mounted directly to the motor shaft. I don't think I will be swinging a prop much larger than that.

 

Ok, there are many unknowns in this project as it has been presented here. Perhaps that is why there are no answers to the question? Anyway, since it might be of interest to someone else, I’ll give it a try. In a European perspective this is an engineering, or construction problem, not a “design” dito, and most engineers have their own preferences for the process of solving; this is my personal approach for the rough, very first input to the construction spiral (suitable for a back-of-a-barnote analysis).

First, what power is needed to propel the actual mass at the desired speed?

If you look at a resistance diagram, you find a Drag versus Speed relation for various Slenderness ratios. If you non-dimensionalize drag and speed to Drag/Weight and volume Froude number, you can get D/W as a function of Fnv. After some arithmetics you arrive at the following expression:

V=(P/(constant*displacement^(5/6)))^0.5;

here V is speed in m/s, P is power in HP and displacement is m3. With a propulsion efficiency of 50 %, the constant is roughly 0.45 to 0.55 for “normal planing hulls” in the Froude range ~3 to ~6.

So, for your 250 kg and 10 m/s we arrive at ~16 Hp. This means a thrust of (16*736*0.5/10)=589 N.

Second, what propeller size is needed to produce this thrust at the required speed?

You were aiming at a “longtail” arrangement with a surface-piercing prop, for which there are no “Twitter-quick-fixes” available. But there are some different formats in use for preliminary estimates for both supercavitating and transcavitating screws. I prefer the format (kt/J^2), where kt is the normal thrust coefficient and J is the advance ratio. After some manipulation it takes the form:

kt/J^2= T/(density*D^2*V^2);

where T is thrust, D is prop disc diameter and the density is for the water, of course. The good thing here is that the kt/J^2 – value is rather constant for rotors operating with part submergence, ventilation or cavitation. For fully submerged supercavitating propellers with special blade profiles the value is about 0.145. For your case with something like a Newton-Rader propeller operating surface-piercing, a reasonable start value would be 0.08 to 0.085. That will result in a propeller diameter of 0.263 m.

Third, what is the required shaft speed for this propeller to produce the required thrust?

A very rough rule of thumb shows that the torque coefficient (kq) for props operating with a gas cavity over its suction side must not be over ~0.023.

kq=M/(density*n^2*D^5); and M=P/(n*2*pi);

where M is shaft torque, P is power (in W) and n is shaft rps.

with the values at hand, your shaft speed is now 40.2 rps, ie 2409 rpm.

Fourth, what is a suitable pitch here?
A starting value is P/D~1.3*J, and J=V/(D*n).


Fifth, what shaft diameter is needed for this propeller?
We use DNV, saying that the diameter in mm is a constant*(P/(1000n))^(1/3);

For SS AISI 316, the constant is 23, giving a minimum shaft diameter of 15.3 mm. Now a propeller in surface piercing mode produces a bending moment in the shaft, requiring some beefing-up. A 20 mm shaft has about 2.2 times the bending strength of the 15.3 mm shaft; should be ok. If in doubt, check rotating bending moment.


Finally, what is the maximum bearing distance and prop overhang?

There are three criteria; the resonance speed from the overhung propeller mass, the resonance speed of the shaft flexing between bearings, and the compression bending from the asymmetrical propeller thrust.

In this case I consider the first criterion to dominate. This resonance speed should be >2*shaft speed in order to avoid damaging vibrations.

w^2=3*I*E/(m*(l+c)*c^2);

where w is angular velocity in rad/sec, I is moment of inertia (=pi*D^4/64), E is modulus of elasticity (steel ~2e11 Pa), m is overhung weight, L is bearing distance and c is overhang distance.

With m = 2 kg, diameter 20 mm (I=1.57e-8), L =0.7 m and c = 0.1 m, the resonance speed is 550 rad/s, or 5250 rpm, which is ~2 times the shaft speed, but exactly equal to the blade frequency of a two-bladed screw. As seen in the formula, the overhang comes in squared, so reduce overhang to a minimum.

And then what?

Two problem areas remain; there is no flywheel effect in a small elecromotor. The propeller torque in this case is pulsating with blade frequency, and without a flywheel, the motor will respond by consuming its current with a corresponding pulsation, which translates to a torque pulsation that goes into the structure. How do you attack that problem?

And then the hump thrust problem. You have a boat with about 0.25 m3 on a waterline length of, say 3 m, giving a slenderness ratio (Lwl/displ^0.33) of ~4.8, resulting in a marked hump drag. This will demand some compromise in propeller selection with ventilated props; best remedy is a longer wl, but you may get over it if you can increase submergence without stalling your motor (torque will be high).

So, there you are, shooting fish..... or?

 

What he (Baeckmo) said ....

 

Thank you Baeckmo this gives me a great deal of information and other parameters to consider. I truly appreciate you comprehensive input, I am getting close to glassing the hull and finishing some components and while I am in that process I will be figuring out my drive components, shaft, bearings,propeller, lifting and steering mechanisms, what you have so generously provided will help a great deal.
.

 

A switch in units? hp m/s?

 

Hello Eagle-eye Barry ; I confess that I am not altogether SI units here. Since all engines still are presented with power labels in Hp, and m/s is very close to knots/2, I chose to be pragmatic instead of dogmatic. Together they result in the constant 0,5, which is about as many decimals as my brain RAM will store nowadays. Used as a first estimate, it gives a fair idea of the hydraulic quality of a boat "you just saw".

 

Thanks for the formulas. Another addition to my powering formulas.

I am trying to write this in Excel. What do I use for resistance diagram? I can't find anything referring to slenderness ratio. I have Dave Gerr's displacement, semi, and the Crouch graph.

 

There have been some graphs presented in this forum before (Ad Hoc et al); I'll come back with a few examples later.