this graph shows two systems which are the same except one has a motor with double the kv and that system is showing to have almost double the losses at this speed and output. the losses accounted for with the simulation are esc and motor. what percent of the losses do you think are in the esc and what are in the motor?
both motors have the same copper losses if the km (mass constant), torque and rpm are the same, so in the higher kv system, 100% of the “extra/additional losses” would be in the ESC… in other words the circuitry in the esc of the doubled kv motor is producing 4 times as many watts of heating as the other esc, since it’s using twice as much motor current to produce the same torque with the doubled kv motor.
also @Hummie it isn’t clear in your simulation whether the doubled kv motor has 1/4th the electrical resistance of the other motor as would be the case if the km of both motors is the same.
So it seems and the" motor amps" are about double w the x2 kv. but "total wattage " shows higher. Don’t remember seeing that term on there defined. That’s a lot of heating the esc no?
If you’re married to the specific impeller, which implies a specific range of rpms, and you dont have the option of gearing, then you will want to pick a motor such that Kv * battery_voltage = max rpm. Furthermore, you practically always want to run as high a battery_voltage as possible (given your ESC, space, and weight constraints) as this will mean lower current draw, and less ohmic heating.
If you aren’t married to the chosen impeller, I’d say you should really talk to the quad-copter / airplane / boat guys, as they will know a lot more about picking a motor+impeller combo. You say you dont have gearing, but I get the feeling that impeller diameter is very similar to gearing: higher diameter = more water pushed per rpm = high torque, low speed motor.
Edit: on second thought I think you’re going to need to do the math. This problem is enough outside our typical domain, that I dont trust my intuition. The big difference is that in our application, the amount of torque it actually takes to sustain top speed on flats is pretty low. We’re not moving fast enough for aerodynamic drag to overpower our motors. But in your application, you have to overcome hydrodynamic drag of the board AND the torque it takes to just spin an impeller in water, in addition to aerodynamics. Frankly, 14500 rpm sounds like a LOT in water, I’m guessing its going to take a LOT of torque to hold speed, and I’m not sure our motors will be able to sustain that power output without going above the typical 10-12s voltages. Out of curiosity, do you know what the torque required to spin that impeller is, and how it varies as a function of rpm? On the electromechanical side, motor torque = Kt * Current, and Kt = 1/Kv. So higher Kv motors also produce less torque per Amp. If you pick a motor with a relatively high (for eskate anyway) ~325 Kv (which should hit 14500 rpm @ 12s under no-load conditions), the Kt might be so low that to get enough torque to actually hit that rpm figure in water, you’ll draw so many amps that you’ll definitely fry the ESC if not the motor. You want to pick a high Kv motor to spin the impeller at efficient speeds, but if you go too high you wont get enough torque to actually spin the impeller at the desired speed. Its a maximization problem that doesnt exist in our skateboard applications, so I doubt any of us have done the homework yet.
Also in answering your original question about Kv and max amps, yes, you are right that low Kv motors have thinner, longer wire and more internal resistance. Roughly speaking, motor wire resistance should scale with Kv^2. Since max current is mostly a function wire heating, which is a function of internal resistance, max current should scale with 1/Kv^2. Most of our motor manufacturers dont reflect this in their technical specs, but I think those specs are probably pretty made up anyway. A higher Kv motor will be able to handle more current.
@Hummie do you have a good way to estimate the losses at the ESC? For comparison, a 10s board with dual 6354 motors, geared to 50 km/h max speed, climbing a 10% hill (steady speed, not accelerating ) will draw ballpark 40A total (20A per motor), and have motor internal resistance 0.039 ohm. That comes out to 20A^2 * 0.039 ohm = 15.6 Watts of ohmic heating. My guess is that ESC inefficiencies will be negligible in comparison, but frankly I have no idea.
i assume the vesc wattage measurement is just what is coming from the battery before its losses. maybe not. but if so then maybe possible to add a wattmeter on each phase? i dont know how people isolate esc and motor losses and thats a guess.
i’m curious how your motor will show on the grin simulator! you can adjust the wheel diameter, and go into the motor selection and at the bottom there is “custom” and plug in resistance and a bunch of iron loss stuff and other things you can get off the vesc tests or running in real time, even a gearing option. wish i had a motor running to see. also can even adjust frontal area to incorporate aerodynamics somewhere on there. this tool is pretty awesome for boards too it seems. super excited to see if i can beat its predictions at 30mph with my best aero tuck and see how we stack up verse bikes at speed in a tuck. going for lowest wattage used at 30mph on a flat. and then you have to lay down and go for the record.
That’s a really cool simulator, but I haven’t educated myself on iron losses to configure custom motors correctly yet. Next in line I guess
resistance (i think phase to phase, if not then double it) is on the vesc foc test, as well as the inductance, and the others you can get in the real time screen going full speed doing the no load speed, and get the current you see then, and some other slower speed see what current is happening then.