What’s the point of higher voltage? (Split from “Sherman-S 3600wh: 100V, 20", suspension, 97lb”)

[sbb]

7 minutes ago, The Brahan Seer said:

in a complete electrical circuit.

Point of order: this part is not necessary.

[The Brahan Seer]

This is a great slow motion of a race start, it looks like everyone is accelerating pretty much the same for a second..

 

[Timwheel]

On 3/23/2023 at 4:13 PM, soulson said:

If you have a 2kw (power) motor that is rated at 84v (voltage) and another 2kw motor rated at 134v, with "all other things being equal", the 84v motor will have more torque than the 134v motor.

This counterintuitive result is because the torque output of a PMSM is proportional to the quadrature axis current of the motor's stator windings. If you hold the power constant at 2kw, and power = voltage * current, then when you increase the voltage, the current must decrease.

However, power = torque * angular velocity too. Since the power stayed the same and the torque decreased, the angular velocity must increase. So the 134v 2kw motor will be faster, with "all other things being equal." (Faster in the sense of top speed; it will take longer to reach the top speed on the 134v motor without loss of balance because there is less torque available, so you won't be able to lean as hard.)

You are right, but based on my faulty explanation. You are absolutely right based on my poor answer, and I'm glad you caught that.

I was careless in my answer, as I said that "everything else being equal" but as you pointed out it isn't the case. For a given motor, you can't apply higher voltage without also applying higher current, as once the system is at crusing speed the motor can be described by a simple resistance since inductance is no longer part of the equation, with the well known Joule's law. Applying double the voltage to a given motor will also lead to double current and therefore quadruple power. I assume however that current is limited by the controller (I think I've seen some overcurrent protection), and I doubt any of us regularly hit those limitations (the master is rated at 240amp max, but I assume it is phase voltage).

Therefore, since we can't access all the torque of the wheel on a regular basis, I'd say that the torque is pretty much the same as it is rider limited, but it should be possible to push higher current with higher voltage. I think that may explain why most people feel that the higher voltage wheels feel a bit zippier (even though I'm not sure I agree on that feeling).

We should be careful for those torque discussions, as a BDLC at very low speed behaves close to short circuit, meaning you can push insane amounts of current from 0 speed. And most discussions are around that "starting" torque. However, it is indisputable that the only way to have torque at higher speed is through higher voltage : you can full tilt on a master at 70 km/h, but do that on the same motor with an RS HT and you'll faceplant as the sag will get the voltage low enough that the real top speed will be much less than the advertised 79 km/h free spin.

[mrelwood]

3 hours ago, Timwheel said:

For a given motor, you can't apply higher voltage without also applying higher current, as once the system is at crusing speed the motor can be described by a simple resistance since inductance is no longer part of the equation, with the well known Joule's law. Applying double the voltage to a given motor will also lead to double current and therefore quadruple power. I assume however that current is limited by the controller

I’m not sure if I understood you correctly or not. If not, all of the following should be discarded and apologies requested. But:

You are describing a straightforward DC current with a simple resistive load. That’s not how BLDC motors are operated though.

 First, each coil group gets powered on only for ~6 degrees of rotation. So each coil group gets powered on and off for about 500 times per second when riding at 50km/h.

Whether the wheel accelerates or not is dependent on the voltage that the motor is driven with. A higher voltage than it’s back-EMF voltage, and it will try to accelerate. The length of each coil group’s power on phase is what determines how much total current is being used and effort being made. It’s not voltage dependent though, you can use a lot of power going at a steady speed as well. Accelerating from zero still uses a low voltage, but the PWM pulse width is large, so the coils are powered on for longer.

 Yes, the controllers have a current limiter. 

 

Edited March 27, 2023 by mrelwood

[misterdalto]

10 hours ago, Timwheel said:

You are right, but based on my faulty explanation. You are absolutely right based on my poor answer, and I'm glad you caught that.

I was careless in my answer, as I said that "everything else being equal" but as you pointed out it isn't the case. For a given motor, you can't apply higher voltage without also applying higher current, as once the system is at crusing speed the motor can be described by a simple resistance since inductance is no longer part of the equation, with the well known Joule's law. Applying double the voltage to a given motor will also lead to double current and therefore quadruple power. I assume however that current is limited by the controller (I think I've seen some overcurrent protection), and I doubt any of us regularly hit those limitations (the master is rated at 240amp max, but I assume it is phase voltage).

Therefore, since we can't access all the torque of the wheel on a regular basis, I'd say that the torque is pretty much the same as it is rider limited, but it should be possible to push higher current with higher voltage. I think that may explain why most people feel that the higher voltage wheels feel a bit zippier (even though I'm not sure I agree on that feeling).

We should be careful for those torque discussions, as a BDLC at very low speed behaves close to short circuit, meaning you can push insane amounts of current from 0 speed. And most discussions are around that "starting" torque. However, it is indisputable that the only way to have torque at higher speed is through higher voltage : you can full tilt on a master at 70 km/h, but do that on the same motor with an RS HT and you'll faceplant as the sag will get the voltage low enough that the real top speed will be much less than the advertised 79 km/h free spin.

this is interesting. 

i actually rebuilt my RS HT this week and dropped in a 3500W C38 motor from a master, while my freespin speeds are marginally the same, 79-82km/h; but riding speeds slightly more. 

using the same RS controller with 100v battery architecture
my average top speeds:
2600W C38 ~56km/h

3500W C38 ~62km/h

my firmware allows voltage drain to 3.0v/cell, and my speeds are ~45km/h @ 15% battery (75.1v under load) with ~20% safety margin using the higher nominal powered motor v the stock. 

Last night i hit a peak voltage of 11.5KW and was just like, lololol

the thing is a fully custom build at this point 😆
RS PRO 100V 3500W C38

attached some screenshots in a google photos album https://photos.app.goo.gl/aXUrvM8eBNpjRzYV8

definitely been fun to play around and tinker with. feedback from the my fellow racers here in the PNW are very impressed with the result. its nice to actually get to test some of these thought experiments, and, am delighted to be pleasantly surprised xD 

[0000]

2 hours ago, misterdalto said:

using the same RS controller with 100v battery architecture
my average top speeds:
2600W C38 ~56km/h

3500W C38 ~62km/h

Very interesting, assuming all other dimensions are the same it must be a slightly higher KV motor (different copper winding).

2 hours ago, misterdalto said:

attached some screenshots in a google photos album https://photos.app.goo.gl/aXUrvM8eBNpjRzYV8

Digging the painted shell and color scheme there. Street tire really fills it out nicely too.

2 hours ago, misterdalto said:

feedback from the my fellow racers here in the PNW are very impressed with the resul

I kinda want to see you take it even further if you haven't already with high performance and low resistance cells such as the Molicel P42As. If you upgraded the packs and even the wiring from the packs to the motor controller, even the phase wires themselves, you'd gain a little in terms of less voltage drop and heat loss making for slightly more efficient and faster laps. It'd be interesting to see where those kinds of mods would stack up next to your datalogging results now if you haven't already upgraded your packs.