We did the final test on thruster version 2 today. I have posted a video that also shows the speed trace while we were playing around:
As you can see in the clip we made 6.7kph in calm conditions with a power drain of 2527W out of the batteries. 5kph was possible with a miserly 900W. The spikes in the chart are where we moved the GPS across the deck. The speed for power is roughly what was predicted but the target was to get up to at least 4kW with the single unit so we could have a good indication of what two units could do, both working at 2kW. With total of 4kW the aim was to get 5+kts (say 9 to 10kph) but we could not get the power for a raft of reasons set out below.
Note that the controller is on the top of the case now with leads exposed so we could measure the AC phase currents and voltage as well as the DC from the battery. That was to get a better understanding of what was going on and guarantee equal phase currents.
Since the previous tests the lead/acid batteries have been replaced with a 48V 60Ah lithium battery. The voltage droop under load is very small when compared to the lead/acid. The batteries are no longer a limitation. We have been on and off the mooring numerous times under electric power alone but only in relatively calm conditions. Also the scum was mostly cleaned from the hulls for the test today.
This unit uses a Mars 3001 PMSM I had laying around and a Kelly KBL48301 purchased to match the motor. The published data for the combination is shown here:
This curve shows the current going to 180A while controller is rated at 300A. Motor is only rated at 100A. At 180A it gives a speed of 2350rpm and output power of 5000W. The input power is 48X180 giving 8640W. Hence the efficiency is 58%. Note the dark blue efficiency curve that shows 68%. So first inconsistency found in these curves.
The output power of 5000W at 180A and motor speed of 2350rpm corresponds to a torque of 20Nm. It would be reasonable to assume that a 300A controller could then produce 33.8Nm. In fact it can BUT only under locked rotor condition. We could not get more than 14Nm from the motor once it had started rotating. This was confirmed by Kelly as being a preset limit. So these published curves showing that at least 20Nm is possible at 2350rpm are pure fiction. They could not be produced with the KBL48301 controller that they are supposed to show.
This was a partial answer to the puzzle but did not completely explain why the motor was bogged down so badly. It was only getting 1100rpm at 6.7kph.
The prop we are using is nylon, supplied as 540 diameter by 345 pitch. It is quite impressive - landed price $66. Has ample strength being rated for 30HP. Measurements showed the pitch to be actually 520mm. This provided the missing piece of the puzzle.
With all this new information we could get more from the motor using different gearing but so far the 3.3:1 box has worked flawlessly and we intend to change bearings to angular contact so it has more thrust capacity. At this stage this is one component meeting expectations. There are 4 identical boxes being used on the steering.
At 6.7kph the overall propulsion efficiency is only 40%. The motor is 66% efficient and prop 61%. It would improve with two units but is really unacceptable due to the motor being bogged down.
The next step will be to upgrade to motor/controller combination capable of getting the torque I expected from the PMSM motor when pushed into overload. There is a brushed motor that can do this within its rated conditions. It would also be nice to get bigger diameter props with smaller pitch that suit the existing box.
With two 680mm props (stated as having 390mm pitch) the propeller efficiency can be lifted to 75% and the brushed motors will be in their efficient range giving 89%. So overall efficiency lifts to 66% which extends the range from the batteries significantly.
The fun continues. The bigger motors and controllers are higher cost but estimate for two units still under AUD7000. That includes the lithium batteries but no charging system. Intention is to trickle charge with a couple of existing solar panels or use the existing portable gasoline generator and charger.
__._,_.___
.