Project Logs

Collapse

• Choosing an motor

  erik • 03/11/2024 at 13:13 • 0 comments

  [En svensk version finns i slutet av inlägget]

  To choose the right size of motor

  When it came to choosing the motor and inverter I started by selecting a maximum speed I would like to drive at continuously for 30 minutes, and a speed I would like to be able to drive indefinitely at. My initial thought was that I am willing to loose 1 Knot as maximum speed, which means a new max speed of 6 knots. And continuously I would like to not go slower than 4 knots. 

  From my calculations, that means a motor that can deliver ~10kw for 30 minutes and 2kw continuously. However, since 30 minutes is a rather long time compared with how quickly an electric engine heats up, the continuous rating will have to be close to this. 

  One more thing that I needed to know is at which RPM the engine shall be able to deliver this, and that is simply to look up the propeller rpm at the 10kW. Which, looking at the logged data, is around 850rpm. However, electric motors usually gett a high power output by spinning fast, and that is a problem since the torque goes down proportional to the rpm increase for a specific power output, since: power[w] = torque[Nm] * rotational speed[rad/s]. 

  So the motor has to be able to deliver 10 kw at 850 rpm, and the torque delivered is then around 110 Nm, which is quite a lot. To compensate for this the diameter and pitch of the propeller could be decreased, or some type of gear reduction could be installed.

  The things that I mainly considered was buying a kit or similar parts and using a toothed belt drive to increase the torque delivered by the engine. This seems like a popular DIY solution and I was about to jump on board that ship when I found a Tesla SDU (the front axel or FDU) on sale at Blocket (Swedish version of Craigslist/eBay) for a better price than the other solutions I had looked at. Before I decided to not board the usual DIY boat and sail by myself I tried to find some data on the motor and what it would mean for me. 

  Tesla Small Drive Unit

  This is what I found: 

  Tesla Small Front Drive Unit Specifications.

  Weight 90 kg (198 lbs) without driveshafts.

  Max Speed 18,000 RPM

  Transmission 9.34:1

  Voltage Range 200-420 Volts DC

  Max Current 650 Amps DC

  Max Power 220 kW (300 Hp)

  Max Torque 330 Nm (243 lb-ft)

  Output Power (12 min.)90 kW (121 Hp)

  Continuous Power 35 kW (47 Hp)

  Max Regenerative Braking 90 kW (280 Amps)

  Which clearly means that it has enough power. Apart from that I wanted to know how to control it as well as at which RPM it delivers the 35 kW.

  For control, I ended up choosing to replace the logic board with one from https://openinverter.org/shop/ It is not the simplest solution since it requires opening up the inverter, but it was the cheapest and the way that would give me the most control which is why I chose to go with it. 

  When it came to when the motor delivers 35 kW, I made a big assumption about there being no losses. 

  At around 70 kph the Tesla comes into constant power delivery, which can be seen in the picture below, so the motor should be spinning slower than then to be able to run at the optimal slip.

  
  

  Assuming standard 19-inch (0.4826m) wheels it means that they are spinning with

  rpm. This means that the motor after the gearbox is spinning at 770x9.34=7187 rpm.

  Since torque output is the main reason for heat at lower speeds and loads, therefore the max torque is more or less constant through the speed range. We can now calculate the continuous torque the motor can supply. 

  Nm out of the gearbox.

  434/9.34=46.5Nm from the motor.

  Similarly, an estimate of the torque the motor can deliver at 18000 rpm can be calculated 

  Which after the gearbox becomes: 18.6x9.34=173Nm.

  From this, it seems that the gearbox is essential to get the torque needed. This however means that I need...

  Read more »

• So it begins

  erik • 12/26/2023 at 22:27 • 0 comments

  [En svensk version finns i slutet av inlägget]

  Background 

  Two years ago I bought a boat with the hope that someday be able to electrify it. It is a displacement boat of a type called snipa or snäcka which typically has low drag at low speeds. For that reason I chose this boat type, the boat is a bit bigger than what I planned for, and I think that is the theme for most of this build. It is a Sandvik 25 and currently, it has a Volvo Penta MD3B as the engine. 

  It has a displacement of 2400kg according to the old datasheet. But it is probably more with the equipment that has been added over the years. Before starting the conversion we exchanged the rubber gasket on all windows to keep the rain on the outside of the boat since that was a bit of a problem before. Because it does feel quite important to keep the water outside of the boat before installing a high-voltage battery in it. We also repainted the bottom as it was for lakes and not for the sea.

  Initial testing

  At the end of the first summer with the boat, we went out and made a small test logging the rpm of the motor, GPS speed as well as speed through water (not used). During the test we drove back and forth with one rpm setting before increasing it to 100rpm, writing down the speed after it had stabilized after each turn. After a total of 30 turns, we were done with the test, and the average results of the data look like this: 

  My idea is to use this data to estimate the maximum power used to propel the boat at different speeds. I say maximum because it assumes that the engine can produce the power it could when it is brand new, which it most probably can not do. As well as there are other losses on the way to the propeller, which also should make the estimated power a bit higher.

  A simple generic equation for static thrust can be found in this paper, which describes the relationship between the rpm and thrust using this equation: 

  Where T_s is the static thrust produced (in newtons), K_t is the thrust coefficient, ρ is the air density (in kg/m^3), ω is the propeller speed (in rad/s), and D is the diameter of the propeller (in m). I have removed some constants and changed units since K_t will just be scaled accordingly.

  It only holds for stationary propellers, so let's hope that the propeller moves somewhat slowly enough through the water. And that there is not too much growth on it making the equation break even more. But this should only make the estimated power used to propel the boat higher, resulting in a further range after the actual conversation.

  First, we find the power produced by the engine at the specific rpm. In the datasheet for the engine, such a point can be found for full throttle.
  

  Then the rpm of the engine needs to be converted to the propeller speed, which has a reduction of 1.91 and then I also convert it to rad/s.

  Since the speed of the boat, Vmax (in m/s), and power produced by the vessel, Pmax (in kW), is known at the max rotational speed of the propeller (in rad/s). We can solve for the thrust produced by the propeller: 

  Fmax can then be plugged into our thrust equation, where it is called T_s, which we then can solve for the constant. This constant replaces all the other constants since they can be multiplied together. Resulting in this equation:

  This new constant, K, can now be used to calculate the thrust produced at a certain rpm of the vessel, and since the velocity is already known the power used can be estimated. 

  I did also do the calculations with a linear version of the formula (removing the square), to make a worse estimate at lower speeds. The electric motor will also have some inefficiency which will mean that more power will be pulled from the battery than is required to propel it forward. The results from both can be seen below:

  I have also done calculations on the range with a 10 kWh battery, which can be seen...

  Read more »

View all 2 project logs