Efficient electric boat

Hi Jeremy, I am the one who posted that link to the "E-Pod" electric motor. It looks like this design takes care of the loss of power through the 90 deg gearbox but if I read your last post correctly, the actual motor is getting in the way of the proper water flow into the prop? What about a longer driveshaft on the end and a more aerodynamic shape of the cylinder?

 

Jeremy
Your comments about the attention to detail with fairing is reasonable but I do not agree with the above statement.

Have you actually calculated the thickness of the boundary layer for a small boat versus a small airplane in typical operating conditions?

Have you compared the kinematic viscosity of water and air?

Rick W

 

Jeremery

"...The odd thing with boat/prop interactions (from my perspective) is the effect that proximity to the hull has. I've only just realised that, due to the much greater viscosity of water, the boundary layer is much thicker, meaning that hull proximity actually slows the inflow velocity down..."

That is basically it in a nut shell. With a boat, the nature of the hull shape aft varies the speed of flow into the propeller. If the propeller is below the bottom of a boat the speed of water can be taken as speed of the boat...(as in Rick's case too, since the flow is undisturbed by the presence of the hulls). On the other hand, in the case of say a deep keeled yacht, the prop fitted into an aperture, the water flow will be slowed down by the friction on the hull bottom (viscosity effects). So that the propeller will be working in the wake which is in fact being dragged along with the hull.

How much the wake is affected depends upon the flow into the prop (ie hull shape aft), which in turns affects the prop efficiency.

Not familiar with the swallow boats, cant see what the underwater shape is like., but this would probably be in the region 20~30% reduction, if it is a "standard" type of hull shape aft. If the flow into the stern is improved, then this can be as low as 10%....but without seeing the exact lines aft...difficult to say for sure.

Also having the prop too close to the hull is not good too (but often cannot be avoided)...for similar reasons as an aeroplane, wont go into details. Also to aid the flow into the prop, the shaft is generally angled, which affects the performance again, less thrust. It also changes the flow into the prop and flow on the tips which alternate and the flow becomes seriously complex, not to mention the pressure fluctuations from the hull.

"...My set of compromises is pretty heavily biased towards the minimum power needed to power a small inland waters craft.."

Well, that comes down to just 2 simple variables.
1) weight
2) hull length

If you haven't already, you need to do a very serious weight estimate, and don't forget to add a margin for growth etc. After that, its plain sailing...excuse the pun

 

Jeremy,

Thanks for the explanation. I'm not clear, though, on the 36v vs 12v issue. Would running the motor with 12v input be more or less or equally efficient as 36v? 1/3voltage, 3x amperage would give 1/3 rpm, 3x torque, right?

I'm guessing 36v is better electrically but 12v operation simplifies speed reduction to the prop and the rest of the boat's electrical system. 12v with your modified towerpro 5330 would give about 1500 rpm into the 2:1 right angle box; 750 prop rpm would be a practical speed for a decently efficient prop.


Rick,

The dingy I'm envisioning is the Ness Yawl by Iain Oughtred, a light double ended lug rigged beach boat. A lines plan of a very similar boat is attached. LWL = 15.3 feet or 4.66m; wl beam is 4.33' or 1.32m. 650# all up weight with crew.

My estimate of power required is from Gerr's "Propeller Handbook"
Speed in kts=10.665 x square root(LWL in feet) / third root(weight in lbs / shaft hp).

This is an empirical formula fitted to "average" boats with "average" props. It does a very good job of predicting the speed vs power of my 25' electric launch. This is a long skinny 2000# boat with a normal 3 blade 11d x 12p bronze prop (see it at my website below).

Back to the Ness Yawl, the formula says 180w needed for 2.9 kts; 750w gives 4.7 kts. This assumes 90% electrical efficiency (too high, I know) and "average" prop eficiency. I would guess the actual output power thrusting the hull forward would be perhaps half these levels.

My idea is that a narrow 2 blade prop could be fully hidden within a fairing on the trailing edge of the rudder when sailing. Imagine the after 20% of the rudder is removable - when taken off it exposes the prop. Installing and removing the fairing would be much easier than the current state-of-the-art solution: mounting and stowing a small (ugly!) outboard motor. Lets just dismiss rowing

All,

The same book suggests a minimum tip clearance between the hull and prop of 10% of prop diameter and says 15% is even better. I designed my current electric launch with 10% clearance with a 12"d prop and had noticible vibration at high power. Going to 11"d substantially reduced vibration (and cut efficiency about 5%)

On the poor theoretical efficiency of most boat props: They are expected to have some resistance to hitting floating debris and survive an occassional brush with mud or sand bottoms. Long skinny blades cannot be as strong as short wide ones. Also, the only commercial market for low power props is the 5 to 20 hp small sailboat auxiliaries - still much more power than we are talking about here and much more blade loading. As someone above said, all designs are a compromise. When seeking ultimate performance, such as Rick's elegant HPBs, durability can be low down on the priority list and efficiency at the top.

I see the skinny 2 blade prop having acceptable durability due to low power (<750w) and protection from grounding by the rudder. Besides, the concept is not practical at all if the prop does not fit completely within the rudder when sailing. Rick is right, having the prop dragging along when sailing is out of the question.

 

AH,

Thanks for confirming what I thought. I'm still going up a steep learning curve in trying to adapt aerodynamic thinking to hydrodynamic thinking!

The hull form is rather like a flat-ish bottomed canoe, but with a stern that ends in a narrow transom that is just clear of the surface. The run aft from around a third of the way back from the bow is very flat and gentle. I believe that I can get the prop clear of the hull, although I am sure that the upper part of the prop sweep will be operating in an area of reduced inflow velocity, plus there will be some effect from the propeller shaft.

I have attached an approximate hull form, derived from a Freeship model that I derived from a rough lines drawing of the Winsome hull.

Hull weight is likely to be around 40kg, I hope, although I'm expecting the operating displacement to be of the order of 230kg to 250kg or thereabouts, with two of us on board, the propulsion system fitted and a canopy full of solar cells. LOA is about 5.3m, beam 1.1m, LWL about 5.2m.


Rick,

Using a common length of 5m (pretty typical for a small aircraft fuselage and close to the length of the hull I'm thinking of), assuming turbulent flow (probably OK at this sort of Re for a fairly smooth aircraft fuselage and fairly flat bottomed hull), assuming a cruise velocity of 80kts for the aircraft case and 2m/S for the boat (again, not unreasonable figures I believe), then I estimate the boundary layer thickness (using the simplified method, not the full Navier-Stokes method), to be about 7mm for the aircraft case and around 120mm for the boat. Using a very simple formula that appears to be used for big ship design I get a boundary layer thickness of around double this for the aft end of my hull, but this sounds to be in error, probably due to scaling problems.

I don't think there is much doubt that boat propellers are often working in a region of reduced inflow velocity, at least as far as conventional designs with the prop behind a skeg, or tucked up in the stern. Perhaps a better analysis, using a well-proven naval architecture method, might show up some flaws in my calculations. What do your calculations give, Rick, and what method did you use?

Jeremy

PS: I fully accept that my calculations might well be out, they were derived from some fairly big assumptions!

 

Denny,

I don't think there's a "one size fits all" answer to this particular issue, as my experimental system is electrically pretty sub-optimal. Generally, increasing voltage and reducing current is a pretty good thing, as the fixed resistive losses in the system (battery internal resistance, wiring resistance, switch and fuse resistance, controller FET on-resistance and motor winding resistance) are fairly constant. Motor winding resistance will be a bit higher for a higher operating voltage motor, but it's not normally that big a factor.

As the power losses are proportional to the square of the current, anything that allows the same power to be delivered with less current has to be a good thing. The problem is deciding just where the best compromise is. Controller FET on-resistance increases with higher voltage rated devices, cost rises steeply once you get over about 50V or so, safety legislation may rear it's ugly head at high voltages, and generally high voltages and water mix even less well than low voltages.

I think you're spot on that 12V simplifies things for my system as it stands. The new unit that is sitting on the desk in front of me has a 4:1 reduction between the motor and prop shaft, which should allow the motor to spin up to an efficient speed whilst turning the prop at a more sensible rpm.

I have another test session pencilled in for a couple of weeks time, and am looking forward to getting some better data. The parts I ordered yesterday (new stainless prop shaft, watertight housings and connectors etc) all arrived today, so I need to get into the workshop and start work!

Jeremy

 

Hi Jeremy,
I have access to brushless, axial flux PMSM motor prototypes a friend has been developing,
full-time, for a couple of years now. His biggest struggle seems to be the controller.

He's recommending 36 volts for my application.
I'm thinking of a total of 48 of the SLA 6-12's I mentioned previously.
So, 6 in series, then 4 of those in parallel (24) for each motor. Times two, totals 48 batteries.
The motors are unique in that they are hollow in the centre allowing a solid shaft from one to run
through a hollow shaft from the other, counter-rotating, directly to the props!
They would turn ~1000 RPM under the kind of load I'm going to have at speed, so no gearing!
I'd like to run a double (or dual, I don't know the terminology) potentiometer from a common knob to the controllers
for dual speed control without loosing the safety of double system redundancy.
The motors are a little less power than I'd hoped at 4 kw. ~92% efficient.
Efficiency may increase if he can eliminate cogging.
A "momentary on" switch would give full reverse for brakes (hopefully never needed).

Does this sound reasonable to you?

Thanks for your interest.

Tom

 

Jeremy
What values did you use for the kinematic viscosity in each case?

I have attached the comparison of the two conditions you state above.

You will see I have used constant pressure gradients. This is good for the sort of streamline flow I aim to achieve around my hulls.

Many boats have tiny props tucked behind a large bluff body and the pressure gradients are not as clean so there are significant wake factors but this will not be the circumstances in your case unless you try hard to achieve such poor arrangement.

If you read through that article on the Whio I linked to you get an idea of what can be done with good design.

Rick W

 

Denny
I will do the hull calc when I have a little more time on the weekend. The information provided is enough. I can "see" enough of the linesplan - it is not the best quality copy.

Your idea of a cowling on the rudder with the prop at maximum thickness should be very neat.

Is the shaft in your electric launch angled?

When you get down to the low velocity ratio of very efficient props then the clearance is not that critical. My props run with velocity ratio usually less than 1.03 - hardly different to free flow.

Rick W

 

Jeremey

Your hull form looks nice. I would estimate a value around 10% for wake with this hull, (perhaps as a margin, a value of 15% upper limit). Since the flow is relatively clean and undisturbed. What appendages will you have under the hull iwo the prop?

As for prop clearance, 15% is an absolute minimum, better to ahve 20% if you can.

And boundary layer. Pointless calculation. You're not going to be able to change it and its effects. The only variable you really have is the shape of the hull aft, and the prop tip clearance. (You wont be able to get the hull surface hydrodynamically 'smooth'). Other than that forget it....it is an exercise i did when a student, just to prove Prantl's theory and then to calculate the momentum thickness etc....ignore this bum steer.

Your weights.
As i suggested, you need a really good weight estimate. By that I mean listing everything down and its weight, i mean EVERYTHING. If you don't know the weight of something or can't get it from a catalogue, then estimate a "rough" idea value. Then add a margin of approx 10%. Can't do much without it, since everything comes down to what she will weigh!

 

Denny
As noted earlier your idea of the prop in the rudder is neat. The ability of the rudder to provide collision protection also has merit.

I not sure if you know that I do annual race in one of my boats in a log strewn inland river - 404km over 5 days. My prop regularly hits logs and drags over shallow areas. To get durability I have made the shaft compliant. It usually bounces out of the way. The linked photo shows the boat sitting on the ground with the shaft in a "S" curve:
http://www.boatdesign.net/forums/at...5293528-pedal-powered-boats-v11j_cf_frame.jpg

So there is more than one way to achieve durability. Once you start making things big and heavy you start down a slippery slope of inefficiency. More weight, more power, high prop loading, lower prop efficiency, more power, more weight and so on.

The two bladed prop will sometimes get caught just right and I have bent blades (I can straighten them by hand). My next stage is to make the blades folding. If I do it well they will fold out of harms way and they should automatically shed weed when the rotation stops. This is an issue for my areas of operation.

If you consider a deep keel yacht there is no reason why a huge prop like 1m diameter could not be tucked into a faired recess behind the keel similar to your rudder idea. (By the way I did a prop design for a fellow with a rudder mounted prop) A small geared electric motor inside the spinner could drive it. It would pull a barn down and efficiency approaching 90% achievable.

I like the Whio boat because this guy has taken a whole boat concept and made a nice package. Had to make his own prop. Got into all the important bits and taken nothing for granted. That is what I want to do with my coastal cruiser. I find that many compromises are made without really understanding the possible alternatives.

Rick W

 

Rick,
I've seen your video of the prop on the long shaft self centering and sort of going around the corner when you turn. I would have never believed it would work without seeing it! I did not consider the durability aspect of the flexible shaft - damn clever.

The Whio was featured in WoodenBoat magazine a few years ago - quite an impressive boat. The whole notion of slower, more fuel efficient boating is slowly gaining traction. I show my electric launch as four or five boat shows each summer and many, many people say 6-7 mph cruising would be fast enough for them. (BTW, the shaft is at about a 15 degree down angle).

I'm going to have to load java prop and get to know it. For sure common boat props are the weak link in getting the most range out of the small bit of energy batteries can provide.

 

Denny
Boating for me has always been a means of escape. When I was younger it is what we did as a family. When I got my own family my boys not have the same interest for a variety of reasons so I have found something for now that I can have fun with in short time intervals.

I am long past the twin 150HP Mercruisers slamming across the tops of waves. My personal view these days is that it is socially irresponsible to use things like this for pleasure. They are so wasteful of resources.

In retirement I will have more time and a leisurely pace of 6 to 8kts will be fine. I don't think I am alone with this view. I have seen your launch before and it is the sort of thing that encourages me. (I think I can do better but if I get to what you have done I would be happy)

With an inclined shaft you are fighting the natural self-aligning forces of a propeller. The term p-factor is used in aircraft because it causes nasty forces when climbing. It is well known in aircraft. I did not know about it until I bent a shaft for no apparent reason. I then studied the forces and also managed to eliminate an annoying source of vibration I had spent many hours of ballancing and aligning to try to resolve.

The smaller the diameter, the more blades and the lower blade aspect (meaning lower efficiency) the less you will notice the problem but it is still there.

We could back calculate from the design of your current boat what could be achieved with a better prop but a more efficient prop would need to run in line with flow. I do it with a curved shaft. Once you eliminate the bending forces due to the inclined shaft you can go down to quite small shafts to handle the torque. This reduces appendage drag and disturbed flow onto the prop so there are multiple benefits.

JavaProp is quite easy to use once you get the hang of it. It is mostly an analytical method so considerably more useful and flexible than what you have with empirical based tools. The best basic performance assessment tools I have found are JavaProp, JavaFoil and Michlet/Godzilla.

Ask Jeremy what value of density and viscosity you need to use in JavaProp when working with water medium.

Rick W

 

Re prop rampant ineffiency:
The highest selling pedal powered boat by far is the extremely inefficient paddle type you see at water resorts! Yet even the poorest designed prop types are several orders more efficient. Hard to explain that one except that paddleboats are simple and cheap?

Petrol motors came first. So props were patched together to work from this extremely high energy reservoir where one could be wasteful. Also, beyond a certain engine power the torque required to drive a large prop might be more dangerous than the serious slicing action of smaller, low efficiency, high rpm props. Maybe enough torque to flip over a sizable boat when a big prop strikes something? That's what I see FWIW, anyway....

Re weedless props:
I have a feeling that you trade off some efficiency when going to a "weedless" prop. At least that was what I gathered in conversations over a decade ago from some people at minn kota...

Re noise in electric prop drives
Planetary gear types seem to be the worst in my experience. Splined gears do help in my experience. Electronics running at higher frequency beyond what can resonate electronically or mechanically to produce something in the audible might help. My quietest brushed gear motors are essentially silent and use a special beefy nylon spur gear.

Porta

 

Rick,

It looks like you've been using one or other of the Virginia Tech Java applets to calculate boundary layer thickness. These are OK for a flat plate, but both boats and aircraft fuselage aft sections are convergent, which makes the flat plate model lose accuracy. Obviously static pressure reduces as you run aft under a boat hull with upswept lines, so this will tend to increase the thickness of the boundary layer. Also the dynamic pressure decreases too, from Bernoulli. Both effects lead to an increase in boundary thickness over that derived from a simple flat plate simulation. Anyway, as AH has said, it's a moot point, as there is a mass of practical knowledge that confirms the effect this has on conventional (i.e. close to the hull) prop installations. BTW, I used 0.000014607 m²/S as a value of v for air, 0.000001307 m²/S as the value of v for water.

I agree that the compromise on many small boats leads to props being too close to the hull or other obstructions, but sometimes these compromises are necessary for practicality. I'd love to have an efficient flexible shaft system such as yours, but have to accept that this boat will be manhandled in and out of the water in some difficult areas (river and canal banks) and will also need to tolerate grounding fairly well. Being able to tuck the prop up into the hull easily seems a worthwhile idea to expore, as does the idea of a folding prop.

I believe that it should be fairly easy to make a composite folding two blade prop. It'd be interesting to know how well such a device might shed weed, or whether it would spin out OK to go astern. I have a big (1100mm) three blade folding prop for a motor glider sat in my study. It's made of carbon fibre, with foam cores, and each blade only weighs about 250g (see picture below). Using this basic design (scaled down for use in water) I think it should be easy to make a nice, strong, prop. This configuration has the advantage that the carbon rovings can be wrapped around the bearing tube for the folding pivot, making the blade root attachment very strong.

I think we're all coming at this concept with similar interests, but differing backgrounds, which means we'll each put a different weight on certain aspects. As an aircraft designer, I tend to want to build very light, optimise propulsive efficiency as far as practical and go for as clean a hydrodynamic solution as fits with the need for stability and practicality. Had I come from a big ship design background I might well have dismissed these tiny brushless motors out of hand, simply because they look far too small to do the job.

I built and tested the 4:1 belt drive last night and it seems to work well. I'm planning on fitting the motor, reduction drive and controller into a robust diecast, sealed, alloy box. Hopefully this should keep the water out, provide a measure of added sound reduction and act as an effective heat sink for the motor. I should get to take some pictures when I get home tonight.

Jeremy