Efficient electric boat

I'm going to experiment with running it open loop, in sensorless mode. One reason I opted for using Hall sensors was so that I could use an electric bike controller, primarily as they are cheaper and have the advantages of current limiting and a simple throttle interface, with built in reversing.

I've just bought some sensorless adapter boards (about the size of a postage stamp!) that should allow me to use one of the electric bike controllers with an unmodified motor. If this works, then it should be an improvement, as it means less wires and connections down at the motor end.

I think I can get a Mitrpak OK. Their online store won't let me register, but they've been very prompt at getting back to me via email. I just need to call them later today (when the US wakes up, it's 08:00 in the morning here now!) with my credit card details.

I'm also working on a swept blade weedless prop design, that may be of interest to Mike Lampi on the pedal boat thread. I'm trying to get a way to mould it from carbon fibre composite and think I've at last found a way to make a mould that will work OK. I'll post details in a week or so, assuming I can make the method work.

Jeremy

 

Jeremy
I have doubts about a weedless prop offering you any advantage in normal conditions but it will reduce efficiency. The biggest problem I find with weed is the stuff that floats and gets fed down the shaft to the prop. It also collects on the rudder if it cannot be lifted to clean.

With your vertical leg it will just collect at the leg under the hull. It should not get tangled in the prop unless you are actually operating in shallow water.

My view is that a folding prop would be better anyhow. I am yet to make one though.

Rick

 

Now efficient electric propulsion to me is paying $US 160 for an 85 lb thrust 24v minn Kota copy. buying 4 12v 40 AH batteries from K mart on a 25% off deal and a swith to give 12 or 24v ..I buy I fit I enjoy thats efficiency

It all in the perception of efficiency .....

 

I can't argue with that philosophy regarding efficiency of purchase!

My aim is to make a casual leisure boat that looks fairly traditional and that can operate without the need for external power - I'm hoping that I can get enough solar charge capacity to run for a few hours a day. Initial cost isn't as important to me as the ability to operate free from a conventional power source, plus there's the added attraction of learning more about just what can be achieved with some careful optimisation.

A Mitrpak gearbox is on it's way, although the shipping cost was pretty high at $100! No doubt I'll get hit hard for tax and duty too............

Jeremy

 

New Motor

Jeremy,

I do hope you read the reviews on your new motor...

Perhaps at reduced loads you wont experience such trouble.

Best of luck!

Tom

P.S. I like the air conditioning mode on your flip-up drive diagram!!

 

The first thing I learned when I started playing with these cheap Chinese made motors was to pull them apart from new, replace the rubbish bearings with decent SKF or NSK ones and re-glue the magnets. It only takes half and hour or so, and turns a potentially dodgy motor into a rock solid one.

All of the Chinese made brushless motors I've seen have been a bit iffy in terms of assembly quality, but thankfully it's very easy to sort out, and the general manufacturing quality, in terms of machining etc, is usually pretty good, so they are still a good buy.

Jeremy

 

If those batteries were deep discharge with a recent date code, sounds like a good purchase. Using a switch for speed control will save on minn kotas if you buy the lower cost non PWM models. Kudos, sounds like you have found something that meets your goals, and at low cost.

Sailboats can also operate essentially without the need for external power, but I don't know how efficient they are compared to solar cell power. Neither work well on smallish, protected, low draft and shady rivers where I do 95% of my boating though. That is one thing that drives me to tinker with small prop driven e-boats. Perhaps a river boat can operate in bigger waters at least on a limited basis, but the opposite is not true, if I see things correctly.
I have a small, portable, solar flex panel that can drive my tiny boat at about 20% of full speed directly or charge batteries to extend range, but don't use it much. With just a little more money than a minn kota setup and some considerable but enjoyable tinkering time, I have something that is used for short trips twice a week on average. Just depends on what your goals are as Jeremy has indicated.

Heavy weeds are about the only constraint with prop drive I haven't made some progress on. I am reduced to using paddles in such situations.

Porta

 

I've attached a few pictures of the experimental chain drive unit for interest. I'm not at all sure that I'll go with this option in the final design, primarily as it's very fiddly to assemble/work on and probably demands too much regular maintenance. It was a very useful learning experience though, as I now have a far better understanding of the strengths and weaknesses of this approach. If I hadn't done it, then I'd have been forever wondering!

Although I have a Mitrpak gearbox on order, which I feel may be the best overall solution, I spotted some neat double universal joints and bought one to play with. It's surprisingly good at working over a very sharp angle, close to 90 degrees seems possible. Even though the torque rating has to be massively derated for high angles, it's still plenty strong enough for my application. Although this one is steel, the same company make them in stainless. I'm going to make up a test fixture and see just how smoothly this will run at a fairly sharp angle. Being a double joint it works at a constant velocity, so it's just a matter of seeing what the losses are and whether or not it's noisy. There's no reason why the output shaft angle has to be at 90 degrees to the input shaft, if I can get someething that works OK at around 60 to 70 degrees then I think that would be fine. In fact, having an inclined shaft might make it less susceptible to weed clogging.

Looks like another weekend of experiments in the workshop coming up.............

Jeremy

 

Question for Jeremy

Not had any experience of this sort of motor ..all glow and diesel in my model makeing days !!! will it run on a lead acid battery and what sort of voltage ? could you indicate the rpm they run at loaded ..I would be lookingat a simple open loop arrangement if thats possible ..???

always ask an expert ...x is the unknown and a spert is a drip under pressure .

thanks ....

 

No problem at all with running these motors on lead acid, depending on voltage. Anything over 30 volts gets expensive with RC type sensor-less controllers, but 24 volts is fine.

The bigger motors, like the ones I'm using, tend to have a Kv, that's the rpm per volt figure, of around 130 to 250. This means that they will run at an unloaded rpm of 130 to 250 rpm per volt of available supply voltage; loaded rpm will be slightly lower. For example, a 170 Kv Turnigy 63-74 motor will spin at 170 x 24 = 4080 rpm from a 24 volt supply, or 2040 rpm from a 12 volt supply. Generally, it's better to spin these motors fast on a higher supply voltage and gear them down, as the losses are a function of the square of the supply current. For a given power, half the voltage = double the current = four times the losses (which means four times the heat).

They need some sort of reduction drive to run a boat prop efficiently, but that's easy enough to do with off-the-shelf toothed belt parts. A big, slow turning prop is a fair bit more efficient at modest boat speeds than a small, fast one.

The best place to buy motors and controllers seems to be Hobby City in China. They have a big stock, the prices are OK and they have a pretty good reputation.

The motors and controllers tend to be over-rated in their specifications, certainly as far as use in anything other than a model aircraft is concerned. I tend to treat the claims of kW of power as just very short duration peak power ratings. I'm currently testing options using a motor rated at 2800 watts, but I'm running it at a maximum of around 500 watts (about (2/3rd hp) and it runs nice and cool. It might run at around 1hp or so quite happily, but I'd not want to run it for long periods at much over that. The bigger Turnigy motor I've just bought is rated at 3250 watts and looks able to run at maybe 1000 to 1500 watts continuously (it has a bigger stator and hence more ability to get rid of heat than the smaller TowerPro).

The same goes for electronic speed controllers. I tried a 40 amp controller that should have been more than big enough to drive the motor at the power level I'm running at, but it got a bit warm at just 15 amps. I switched to a cheap (around $40) 120 amp controller and that doesn't get warm at all. The modified motor I'm using is adapted to run from an ebike controller that has ratings that are conservative, it will happily deliver far more than it's rated power without getting warm.

Running these motors on standard RC electronic speed controllers is fine, but poses a couple of small problems. The first is that there's no built in reversing capability, if you want to go astern then you need to switch a couple of the motor phase wires over (easy enough with a big switch). The next problem is the throttle interface. These controllers need a pulse position modulated input to alter speed. This is easy enough to provide with a cheap servo tester unit though. Finally, RC controllers have no current limiting, so can easily be blown if the motor gets overloaded (prop jammed up with weed, for example). This can possibly be sorted by using some sort of current trip on the supply side.

Hope this helps.

Jeremy

 

Jeremy, do your motors have a PM rotor and what kind of efficiencies are you getting when running under your derated conditions? If you have any experience with unconventional configurations such as rotating field type how do they compare in efficiencies and heat disapation? Thanks for any comments you would like to make.

Porta

 

Porta,

These motors are PM ones, they use thin neodymium magnets arranged around the inside of the rotating outer drum. They are colloquially referred to as outrunners, because of this configuration. Putting the magnets outside the stator increases the torque for a given overall motor diameter, but means that the conductive cooling path from the stator has to be via one end only.

Efficiency tends to be around 90% over a fairly broad speed range. Derating a big motor tends to reduce resistive losses, which are the dominant source of loss at modest speeds. Eddy losses in the stator core tend to be fairly low for the sort of power levels and speeds I'm running at. I have a spreadsheet model for calculating motor characteristics. It uses measured values of the no-load current at minimum and maximum rpm, coupled with the motor fixed parameters (winding resistance, Kv, Kt etc) and predicts the motor theoretical efficiency. The attached plot is for the Turnigy 63-74 170 Kv motor, running at a maximum current of 20 amps from a 25.6 volt supply.

This motor is rated at 3250 watts, but I've derated it to about 500 watts in this application.

I've little experience of other types of motor, I'm afraid. All of those I've played with have been fairly standard 12 slot rotor, 7 magnet pole-pair, three phase ones, wound as distributed LRK. I've rewired a couple from delta to star (wye) to lower the Kv, but apart from that I've done little more than investigate how they work and measure their performance.

Jeremy

 

Thanks for the details, 90% sounds great for such a low rpm motor. At least my understanding with conventional motors has been that low rpm means low efficiency. How do you remove the magnets without damage and what kind of jig and glue do you use when reglueing? I've used epoxy with conventional motors myself, and it seems to hold for intermittent use but breaks up with extended heat over a period of time.

Do you think it is possible to operate the motor in a sealed can format by covering the openings when running at your present derated fashion? Or do you think the opening is critical regardless of how low you derate for continuous use? Presently, I use an exposed, unshielded, sealed can motor configuration because of simplicity, lower weight, and volume.

Thanks.

Porta

 

OOPS. See it wouldn't be advisable to run an exposed outrunner on a boat....

Do you think it is possible to operate the motor in a sealed can format by covering the openings when running at your present derated fashion? Or do you think the opening is critical regardless of how low you derate for continuous use? Presently, I use an exposed, unshielded, sealed can motor configuration because of simplicity, lower weight, and volume.

Thanks.

Porta[/QUOTE]

 

Porta,

As I understand it, low rpm does mean low efficiency if the motor is being loaded so that it is drawing a high current to keep it at a low rpm, if you get what I mean (it maybe none to clear, the way I've expressed it!).

As far as I can tell, efficiency at low commutation speeds depends primarily on the power wasted by resistive heating losses in the motor. Big, high power, motors are wound with heavy gauge wire, so the resistance is low. Under-running them tends to reduce the losses over a motor that's sized to *just* deliver enough power. The down side is that the bigger motor has greater frictional losses in the bearings, but in practise this seems to be negligible. The bigger motor also needs more current to just turn over, but the trade off still seems to work in favour of derating.

What does hit efficiency hard is trying to pull lots of torque from the motor at low speed. My first experiment only had a gear ration of 1.25:1 and the motor was drawing quite high phase currents even though the battery average current was quite modest. The result was a fairly warm controller and higher than desirable losses. The lesson learned was to gear the motor down, as spinning it at the highest practical speed seems to give the best efficiency. Theoretically, the motor I have should give peak efficiency at around 75% to 80% of maximum rpm, but in practice the efficiency curve is so flat in this region that it's not that important.

I don't bother to remove the magnets to re-glue them, I just drip thin cyanoacrylate glue into the gaps between them and it runs into any voids. To stop it dripping out, I stick the rotor in the lathe and gently spin it, whilst giving it a quick squirt of cyanoacrylate activator. I did one motor with epoxy in a similar way, running it in between the magnets and then spinning it up in the lathe to make sure it all flowed out evenly. As you rightly say, epoxy is not too good a choice, as it does tend to soften when it gets hot.

I don't think there is any need to seal these cans up, as there are no exposed electrical connections inside. I haven't done it yet, but I intend to coat the windings and the magnets with conformal coating, to help keep corrosion at bay.

I think that airflow through the motor does help cooling a bit, but in my experience it's the motor base plate that tends to get slightly warm, so I think that conduction is probably the main means of getting rid of stator heat. In my application the motor waste heat is probably going to be less than 10 watts on average, so it should be easy enough to get rid of.

Sealing one of these motors would mean fitting an extra can over the whole unit. In my case I've opted to mount the motor and reduction drive inside a sealed diecast alloy box for the latest experimental unit, which achieves a similar effect, but allows some internal air circulation to allow (fairly inefficient) heat exchange between the air flowing though the motor and the thick alloy case.

Hope this helps.

Jeremy