Efficient electric boat

[Guest625101138]

Jeremy & Thomas
You have been very quiet here. Have you made any progress with the boats?

Jeremy - have you had the Mitrpak outboard on a boat yet?

Thomas - I am interested to know what speed you can get with what power when you get some data.

 

[MCDenny]

Efficiency experiments with a cheap trolling motor...Your critique invited.

My goal is to find cost effective modifications to an inexpensive trolling motor that maximise its performance (thrust per watt consumed) as auxiliary propulsion for a small (14' to 20') sailboat. I hope to develop a system capable of driving a 16' WL, 800# sailboat for 10 miles at 4 mph with one 60# 12v battery.

My test unit is a four year old MinnKota Endura 36, available in the US for about $110 at discount stores. It is a simple 12v model with a resistor speed control. The manufacturer claims it generates 36 pounds of thrust. It is equipped with a 9" diameter two blade weedless prop of unknown pitch.

An APC model airplane prop can be easily fitted by boring the prop hole to 0.375" and grinding a notch in the thrust face to engage the motor's shear pin. So far I have tried eight different sizes with more on the way. Best performance to date is with an 11 x 4 two blade sport prop (costing <$4.00!)

I also have a PWM speed control meant for an RC car (airplane units do not have reverse) rated for 100 amps continuous. The motor's resistor speed control is of course terribly wasteful but right now I am only concerned with motor power consumption, not battery drain.

I have rigged up a scale to measure thrust mounted to my dock so the prop operates in undisturbed water, as it would on a boat. I don't think thrust measurements made in a water filled trash barrel can be accurate with all the turbulence. I first tried actually driving a canoe and measuring speed with a GPS but the speed display jumps around too much to give readings to the 0.1 mph with much accuracy. Doing the tests on a solid dock yields much more repeatable data.

I am measuring amperage and battery voltage with a "Watts Up" meter in series with the battery cables (6ga); motor voltage is taken with a DVM connected to the terminals on the upper end of the power wires going down the tube to the motor. Thrust measurements are taken with a 0 to 300# spring scale arranged so the scale reads three times the motor thrust to improve improve resolution of the small thrust forces.

My goal is to mazimise thrust produced per watt consumed by the motor. Motor watts = circuit amps times motor voltage. The results so far:

At 200 watts to the motor the standard motor/prop produced 18.1# thrust.

Substituting the PWM controller resulted in 21#, a 16% improvement. [Can someone explain this? Is there a reason 9v steady dc is different from 12v dc pulses? Could the pulsing be affecting the voltage measurements?]

Substituting the 11x4 APC prop increased thrust to 23.6# - in total 30% more efficient than the stock motor.

Just switching props increased thrust 16% with either controller. At 350 watts I still get 16% more thrust from the 11x4 prop, 26 vs 30#.

The 11x4 drove the canoe 5.0 mph vs 4.1 for the standard prop - 22% more but as I said above the speed measurements are not as reliable as the thrust data Amp draw dropped about 10% for all props in the free running canoe compared to zero speed amp measurements on the dock.

Note the stock prop draws 35 amps, the 11x4 draws 37 so there should be no issue with overheating the motor.

Next steps are to try a few lower pitch props; to get better connectors and to wire in a bypass around the motor's speed control.

The .25" spade connectors in my test wiring plus the connectors inside the resistor controller generate a good bit of heat. At 35 amps I lose 0.6v in the motor's internal terminals, 0.4v in the Watts Up meter and terminals and 1.5v in the PWM controller and its terminals. New high amp connectors are coming with the new props.

Any thoughts or comments?

PS: I started with a 2" OD brushless outrunner (Hyperion ZS4045-14Y) that the manufacturer wired as wye for me (thanks for the wye vs delta lesson, Jeremy) to give a Kv of 117. This gets RPM on 12v in the ballpark but I gave up trying to figure out a cost effective way to waterproof it to run direct drive under water. If anyone is interested in this for your own tinkering send me a PM or email.

 

[Guest625101138]

Denny
There is little point in measuring static thrust unless you want to make a tug boat. For the best static pull go for the largest diameter prop you can find or carry and gear it down.

The only relevant test is to operate the motor at the intended speed and thrust required on the boat to achieve that speed.

Working backwards; your 60Ah 12V battery will give around 700Wh at the 10 hour rate. At 2.5 hour rate you should get 500W from it. That means 200W is available from the battery or say 150W at the motor shaft.

I have used this to look at the best prop for 1200rpm as this is the speed you should get from the motor at 12V under load. The best I can get is 12 x 4. It will give a thrust of 55N at your desired speed.

To get any significant improvement you need to gear the motor down. The attached shows at 600rpm you can get the thrust up to 65N for the same power input. The prop is 16 x 9 which should be quite easy to get. You could even go for a 16 x 10 as you might get over 4mph.

To do the gearing you could get one of the little Mitrpak boxes that Jeremy has bought. This provides a simple way to make a neat little outboard. It will handle this power and thrust.

I expect 65N would push your boat to 4mph (1.7m/s) but have not checked it.

Rick W

 

[MCDenny]

Thanks for your quick response, Rick.

My static thrust tests are to determine the best prop out of the handful of APC props I have. I am assuming the most efficient prop at 0 mph will also be the most efficient at 4 mph. 11x4 and 12x4 were pretty close, 13x4 was a little less efficient, maybe do to increasing blade area (wetted surface friction?) Also, I don't want a prop that draws more current than the standard one to insure the motor does not burn up.

I realize the acid test is to put it on an actual boat which I will do to make the final comparison to the standard unit. Some more reading of the GPS manual led me to a way to get better speed data the next time I go out in the canoe.

I want to keep this simple and inexpensive so do not want to add a gearbox. It also needs to stow in as small as possible space so I will devise a way to attach the motor (just the motor and prop) to the rudder is some sort of a easy-to-put-on, easy-to-take-off method (to be determined).

The battery will be ~60 lbs, which is around 80 ah 80% depth of discharge at 20a should take 2.4 hours. That would get me 240 watts out of the battery so I sould be able to go a bit faster or a bit longer than your calculation above.

You were kind enough to estimate the drag of my proposed Arctic Tern sail boat back on post #95 as 70N at 4.6 mph. The 56N thrust above at 150 watts. So with a little less speed and a little more power 10 miles at 4 mph looks doable.

 

[Ad Hoc]

McD
"..I am assuming the most efficient prop at 0 mph will also be the most efficient at 4 mph..."

There are only 3 things one must consider when selecting a prop size:
1) Speed of advance (ie water at the prop, taking into account hull shape etc)
2) The power at the prop
and
3) The prop shaft RPM.

Once you have "made" and tested your electrical motor and you have the figures, you can tick boxes 2 & 3. The biggest is box #1. Until you do that, no point saying what size, you're putting the cart before the horse.

Draw up your arrangement and a rough lines, if you have one. Then you can estimate/determine the wake reduction. Once you have doen that, you'll be in a better position as you can then just simply select the best prop for your application/design from a series of charts.

 

[Guest625101138]

Denny
It is the Arctic Tern right?

With the variable speed control you can limit power. The controller may even have a selectable current limit. This would protect it against over enthusiastic use.

A prop that gives the best static thrust may not give the best performance at speed. The conditions can be quite different.

Anyhow the direct drive with the 11 x 4 prop should do the job quite easily. From my previous drag estimate for the Tern the power should be under 100W at 4mph. You could expect much higher range than 10 mile off 80Ah. I think you will be surprised.

Did I point you to the tests I did with two little 12V batteries on my catamaran hull before:
http://www.boatdesign.net/forums/boat-design/electric-boat-data-20445.html
Post #4 has some actual data. I did about 80 minutes at 7kph from two little 12V 12Ah batteries. I suffered problems with the prop fouling in weeds as well. It was quite boring doing laps on the lake at 7kph to do the test. You will get close to the rating of the battery as it should last around 9 hours. In practice I think you will push it a little faster.

How long is the canoe you were doing the tests on? It could easily have higher drag than your planned sailboat.

You could probably embed a gearbox in the rudder per your original intention. They are 38mm across. Mount the motor on or in the tiller. The boxes have lip seals. I have operated them happily under water.

Rick W

 

[Guest625101138]

....
You were kind enough to estimate the drag of my proposed Arctic Tern sail boat back on post #95 as 70N at 4.6 mph. The 56N thrust above at 150 watts. So with a little less speed and a little more power 10 miles at 4 mph looks doable.

Denny
The 70N is at 2.1m/s. I have redone the prop calc using the 1200rpm. As you can see the prop power will be 218W at this rpm with 12 x 5. So about 260W at the battery - very close to 20A. You could expect at least 10nm from the 80Ah battery.

Rick W

 

[MCDenny]

Rick,

The canoe has a 14' waterline length as compared to the 16' Tern sail boat. The Tern will weigh about 800 lbs though and the canoe, as I am testing, only weighs about 300#. The formula in Gerr's book suggests the 800# sail boat will go 4.0 mph on 200w at the prop while the canoe should do 5.0 mph on the same prop shaft power.

His formulas, though, are really more like rules of thumb and may not apply at such tiny power levels. I am seeing speeds of 4.0 - 4.5 mph in the canoe at 200 watts to the motor. If the motor is 80% efficient then 250 in = 200 out and I measure speed of 4.5 to 5.0. Pretty good correlation with Gerr.

I plan to build the Arctic Tern this winter so will not have the actual boat to test until next spring.

Tomorrow is a weekday so the river should be calmer and I may be able to get some better speed data.

 

[Guest625101138]

Both of these hulls around the 2m/s mark will be just getting into substantial wave drag. I would expect the canoe to drive a little easier

I have attached the data for the Tern in a chart. It shows the battery drain you could expect with an overall efficiency around 60%. This is what I expect you will get with direct drive on the motor.

You will see it will be very sensitive around the 4 knot mark. The canoe will be lower but a similar shape and start rising a bit sooner if it is shorter on the waterline. It will not rise as rapidly though.

I expect the attached data will be very close to what you end up with. The little gearbox with 2:1 reduction would do slightly better. Maybe up to 65% overall. You will see that a small speed change will have a big bearing on range. Best will be at 4.2 knots.

Rick

 

[Guest625101138]

Denny
I would be interested in plots of speed against power for your system on the canoe if you get some good calm water data.

Most GPSs have logging functions. They can be set to give updates every second also. However in calm conditions you should be able to do some consistent runs. Do both ways at a particular speed to negate any current.

Rick

 

[portacruise]

Efficiency experiments with a cheap trolling motor...Your critique invited.

Substituting the PWM controller resulted in 21#, a 16% improvement. [Can someone explain this? Is there a reason 9v steady dc is different from 12v dc pulses? Could the pulsing be affecting the voltage measurements?]

USING AN ANALOG METER SHOULD TELL IF THE PULSING EFFECT IS GIVING BAD READINGS WITH A DIGITAL METER. THE OTHER POSSIBILITY IS THAT USING PWM GIVES HIGHER TORQUE AT ALL BUT FULL SPEED COMPARED TO A SIMPLE SERIES RESISTOR SPEED CONTROL. I THINK THIS IS BECAUSE THE CURRENT FOR TORQUE IS LIMITED BY THE RESISTOR IN SERIES EVEN AT FULL STALL FOR THE MOTOR, BUT THAT IS NOT TRUE FOR PWM. YOUR THRUST SHOULD BE ESSENTIALLY THE SAME AT HIGHEST SPEED SET BY RESISTOR OR PWM IF THIS IS TRUE.

Next steps are to try a few lower pitch props; to get better connectors and to wire in a bypass around the motor's speed control.

MODEL PROPS WILL BE MORE FRAGILE, BECAUSE OF THE GREATER LEVERAGE WITH THE LONGER EXPOSED PROP TIPS, BUT ALSO DUE TO THE WEAKNESS RESULTING FROM CENTER HOLE DRILLED OUT TO FIT YOUR TROLL MOTOR SHAFT- BEST TO CARRY PLENTY OF SPARES.

The .25" spade connectors in my test wiring plus the connectors inside the resistor controller generate a good bit of heat. At 35 amps I lose 0.6v in the motor's internal terminals, 0.4v in the Watts Up meter and terminals and 1.5v in the PWM controller and its terminals. New high amp connectors are coming with the new props.

SOMETIMES THEY WILL SKRIMP ON THE WIRING LEADING TO THE MOTOR ITSELF AS WELL- USING A LOWER GUAGE TO SAVE A FEW PENNIES. YOU SHOULD BE ABLE TO TELL BY FEELING THE WIRING GOING DOWN TO THE MOTOR FOR HEAT. SOLDERING EVERY CONNECTOR POSSIBLE IS BEST ESPECIALLY IN A SALT ATMOSPHERE AS CORROSION EVENTUALLY WORKS ITS WAY BETWEEN MECHANICAL SLIP CONNECTIONS THAT ARE DIFFICULT TO SEAL AGAINST VAPOR.

Any thoughts or comments?

I'VE OWNED SEVERAL EV'S OVER THE YEARS AND STILL HAVE A FEW AROUND EVEN NOW. MOST OF THE TIME THE LINEAR RANGE CAN VARY BY UP TO 50% DEPENDING ON CONDITIONS YOU CAN IGNORE WITH AN IC. YOU WILL LOSE MORE LINEAR RANGE THAN YOU EXPECT DUE TO LOWER TEMPERATURE AND PEUKERT, AND AGING EFFECTS ON BATTERY. ALSO, WHEN FIGHTING WIND, WAVES OR CURRENT IN AN OPEN MARINE ENVIRONMENT, LOSSES WILL BE PRONOUNCED AND GAINS EXAGGERATED. I ALWAYS LIKED THE SERIES-PARALLEL SWITCHING ARRANGEMENT TOGETHER WITH QUICK CHANGE PROPS WHICH I WROTE ABOUT PREVIOUSLY FOR THAT REASON. SAVED ME FROM BEING STRANDED ON A FEW OCCASIONS WHEN SOME SURPRISE WIND FRONTS HAD ME STALLED AT 12V, TRYING TO GET BACK RIGHT BEFORE DARK.

SORRY ABOUT THE CAPS, COULDN'T SEE MY RESPONSE VERY WELL WHEN I PREVIEWED POST WITH REGULAR TYPE.

PORTA

 

[MCDenny]

Rick,

Thanks for the speed/power prediction curve for the boat. I will be thrilled if I can get 4 kts at 150 watts out of the battery.

Unfortunately no tests today as there are gale warnings - winds predicted to be up to 50 mph. Blowing 25 right now.

I realized I can use the average speed function on the GPS trip computer screen; reset it when I am up to test speed and then note the average speed after a minute or so. That will give me a much better data point than guessing the average as the instantaneous speed number bounces around.

 

[Guest625101138]

Denny
I use cameras and the GPS for many useful measurements that I refer to later. I download all the data recorded in the GPS and can use it in various ways. My little Garmin Edge has its own software. The old Etrex was a bit harder to get data from but it was still possible. I have records for a few years now of most things I have done. They provide useful reference.

Rick W

 

[MCDenny]

Porta,

The PWM gives more thrust/watt than the resistor control when both are at full power. The resistor control just directly connects the battery to the motor at its fastest setting so how could the PWM be more electrically efficient than that? That's what bugs me.

This may have something to do with the large losses in the PWM connectors as actual max voltage at the motor is 8.5v with the PWM, 10v with the resistor control, both at max throttle. Hopefully my new connectors will fix this disparity.

With no load I get a voltage drop from battery to motor of 0.2v with resistor, 0.42v with PWM. This shows the PWM is going to 100% power with the throttle pot pegged.

The props are fragile but will be completely protected behind the sail boat's rudder. They are cheap, too, less than $4.00 each. Spares are affordable

The internal connectors MK used are the same crappy .25" spade terminals I used on the external wires. The top of the motor gets warm just from the speed control connector resistance. The actual speed control coil is down in the motor under water. Once I get the best prop chosen I will start working to reduce electrical losses.

 

[portacruise]

Porta,

The PWM gives more thrust/watt than the resistor control when both are at full power. The resistor control just directly connects the battery to the motor at its fastest setting so how could the PWM be more electrically efficient than that? That's what bugs me.




My guess is that there is additional resistance being picked up when going through the resistor speed control. This could be due to the switching contacts being oxidized or something involving the reversing switch. The old troll motor models had only as simple double pole double throw reversing switch, and the highest speed feed directly to the motor. Maybe bypassing the resistor switch with PWM eliminates those losses?






This may have something to do with the large losses in the PWM connectors as actual max voltage at the motor is 8.5v with the PWM, 10v with the resistor control, both at max throttle. Hopefully my new connectors will fix this disparity.

With no load I get a voltage drop from battery to motor of 0.2v with resistor, 0.42v with PWM. This shows the PWM is going to 100% power with the throttle pot pegged.

The props are fragile but will be completely protected behind the sail boat's rudder. They are cheap, too, less than $4.00 each. Spares are affordable




Yes on the model props, I have been using these for many years without a single failure, but at 40 watts maximum. You can get them cheap and graduate them to the optimum because of the large number of pitch and sizes available. Sometimes you find something better than what is predicted by computer through experiment. The best way to test is under fly using a knotstick http://www.knotstick.com/ if you have some clear water. These are very accurate and are even used to calibrate electronic knotmeters, and they don't jump around in their readings.






The internal connectors MK used are the same crappy .25" spade terminals I used on the external wires. The top of the motor gets warm just from the speed control connector resistance. The actual speed control coil is down in the motor under water.




Yes, the coils can be cooled in that way, but there may some slip connectors affecting things down there sometimes.






I get the best prop chosen I will start working to reduce electrical losses.

Porta