Nissan Leaf motor at 60V, 1/6 the power?

[yabert]

Still playing with different parameters to fine tune VESC.
One of my best result : 9 kW at 690 rpm.
It's really nice to see that phase amps are not to high at around 230A.

Edit: turn PA to 350A just for fun: 9.6 kW at 725 rpm

Edit 2: Turn ME1507 PA to 1200A: 10 kW at 680 rpm

I think it's time for me to find a 16S (60V) battery.

 

[mxlemming]

Still playing with different parameters to fine tune VESC.
One of my best result : 9 kW at 690 rpm.
It's really nice to see that phase amps are not to high at around 230A.

Edit: turn PA to 350A just for fun: 9.6 kW at 725 rpm

Edit 2: Turn ME1507 PA to 1200A: 10 kW at 680 rpm

I think it's time for me to find a 16S (60V) battery.

View attachment 378303

Your duty cycle seems unhappy... It should be smooth at 95+is all working well at max power.

Check you're on the latest 606 (yes there's been another 606 release...) there was a major change to the duty cycle controller very recently that got ported back to the release...

That's actually more power than i thought you'd get into that motor at such low voltage.

What parameters for the FW are you running?

 

[yabert]

What parameters for the FW are you running?

Ummm! I'm not sure of what I'm doing, so here a picture

 

[mxlemming]

I see you set FW to 300A and 10%duty start! This is a very unusual setup...

Field weakening is something you add once mtpa and normal operation is stable. They are quite different in behavior despite both being negative d axis current. Field weakening is there to deal with not enough duty available (duty is the proportion of max voltage used) so optimally it would only be applied at 95%... But for stability we choose to ramp it up sooner.

MTPA is an algorithm that creates the most efficient operation point... Assuming the observer is working optimally.

Definitely still worth getting the encoder working even though you might be feeling like you're winning.

 

[mxlemming]

Ummm! I'm not sure of what I'm doing, so here a picture

View attachment 378308

Ok this appears to be no field weakening atall...

 

[yabert]

No FW at all since I tried MTPA.

Definitely still worth getting the encoder working even though you might be feeling like you're winning.

Absolutly, I trust you and I will take the effort to integrate an Encoder to the Leaf motor if that can help.
At the moment, my goal is simply to know if I can output some power from the leaf at low voltage... and for the moment, the answer seem: yes.

 

[yabert]

That it!!!!! 60V tests done and I like the result a lot.

ME1507 pA 600A, Leaf output 9.7 kW at 1100 rpm.
ME1507 pA 800A, Leaf output 10,8 kW at 1015 rpm.
ME1507 pA 1000A, Leaf output 12 kW at 900 rpm.
ME1507 pA 1200A, Leaf output 12,6 kW at 810 rpm.

To add, voltage sag, on the poor battery I have, was significant despite regen, so with a 100 kWh battery in my boat, result should be even better

I'm very happy by this result as rpm and power match very well my goal for direct drive propeller at the Leaf output shaft.
No belt, no pulley, no gearbox, simply max efficiency put into the water at ''low'' rpm.

 

[yabert]

My test bench in detail.

 

[scianiac]

Ok this appears to be no field weakening atall...

I have a suspicion he has MTPA set to target, which seems to apply all of the phase amps set into D axis if given the chance. I'm sure I'm oversimplifying or misunderstanding that but I know you are correct and target will spin a motor very very fast like you applying crazy amounts of field weakening. Maybe you know why this is the case, I'm very curious to understand why this happens.

As it pertains to this application I think the risk here is how much voltage will be fed back into the controller if there is a fault while running at an RPM that seems to be well above free spin RPM, I guess just the VESC here already has some headroom but something to think about.

 

[yabert]

Talking about field weakening, I continue to play with settings (don't really know what I'm doing) and at a moment, rpm reach over 3000 without load
That should be quite a lot of FW as the ''base'' rpm is probably around 800-900.

So, I disable another time FW and continue to play with settings

 

[mxlemming]

I have a suspicion he has MTPA set to target, which seems to apply all of the phase amps set into D axis if given the chance. I'm sure I'm oversimplifying or misunderstanding that but I know you are correct and target will spin a motor very very fast like you applying crazy amounts of field weakening. Maybe you know why this is the case, I'm very curious to understand why this happens.

The target vs measured... Basically MTPA applies a simple 2 line equation to the requested current, with a deterministic output.

The input to the equation is selectable, either mapped to the throttle or to what's actually happening in the motor.

If you select the throttle, aka target, then it applies the d axis current regardless of whether q current is achieved. This isn't really MTPA...

With target, it sees how much q current is actually going on (which is reduced by power settings, battery current settings and hitting max duty) and bases the d current on that.

The one that causes infinite speed is when you hit the duty limit and even though there's no q current it still apples a lot of d current which is... Field weakening...

So the best solution is to use MTPA measured and field weakening, and set an rpm limit.

 

[yabert]

Thanks a lot for all that mxlemming.

Last test tonight: 14 kW at over 1100 rpm
Battery was at 61V before sag.

 

[yabert]

Quick question for motor expert here.
My ME1507 is rated to 600 pA max 60 seconds.
Is there chance I've damage something with my tests of 1200 pA for 5 seconds?

 

[dominik h]

Thermaly [ I²xT ] it should be ok.
I do not know if magnets or windings suffer from to much current.

 

[mxlemming]

Quick question for motor expert here.
My ME1507 is rated to 600 pA max 60 seconds.
Is there chance I've damage something with my tests of 1200 pA for 5 seconds?

Probably ok as long as you let it cool well.

Neodymium is very hard to demagnetize except with heat so i wouldn't worry about that. I've pulsed motors with 20x their rated current with no lasting effects.

The really big question for me it's why it needs so much current... Is the leaf motor just so much higher torque?

 

[yabert]

The really big question for me it's why it needs so much current... Is the leaf motor just so much higher torque?

Exactly!
ME1507 is give at 0.22Nm/A and based on my tests, the Leaf is around 0.6Nm/A

I mean: 14 kW at 1100 rpm imply 122Nm (less efficiencty losses) with only around 300pA

 

[yabert]

Is someone have advices to how find information do do a display for the VESC?

Now than I'm convinced the Leaf motor will output up to 15 kW and 10 kW all day long with water cooling, I would like to setup a display for the VESC controller in order to have power, voltage, SOC, rpm, etc.

Where to find information to do this?

 

[A-DamW]

Is someone have advices to how find information do do a display for the VESC?

Now than I'm convinced the Leaf motor will output up to 15 kW and 10 kW all day long with water cooling, I would like to setup a display for the VESC controller in order to have power, voltage, SOC, rpm, etc.

Where to find information to do this?

Here are some I have found over the years, in no particular order;

Gallery

WheelDash - The ultimate Amazfit and Garmin app for EUC riders.

blkfribourg.github.io

GitHub - charclo/roxie-firmware: Firmware for Roxie, a device for electric vehicles that displays data read from VESC based speed controllers.

Firmware for Roxie, a device for electric vehicles that displays data read from VESC based speed controllers. - charclo/roxie-firmware

github.com

DAVEGA X – esk8 telemetry

GitHub - SimonRafferty/VESC_ESP32_Display: A simple display for a VESC based on a TTGO-Display board from LilyGo

A simple display for a VESC based on a TTGO-Display board from LilyGo - SimonRafferty/VESC_ESP32_Display

github.com

GitHub - aka13-404/VSETT-LISP: Repo with lisp scripts for vesc for Kaboo/VSETT/Iohawk scooters

Repo with lisp scripts for vesc for Kaboo/VSETT/Iohawk scooters - aka13-404/VSETT-LISP

github.com

GitHub - r3n33/vesc_display

Contribute to r3n33/vesc_display development by creating an account on GitHub.

github.com

RTV Display For Vesc (software 2026) IT ON SALE WONT LAST

This Display only work for vesc controller Functions The display has 3 gears in which you can adjust the maximum speed, percentage of maximum current, and power You can select the default gear in the settings, which will be selected when you turn on the VESC you can turn on auto parking...

evpro.world

GitHub - azhinu/VESC-dash: Dashboard for VESC based ESC built on TTGO T-Display

Dashboard for VESC based ESC built on TTGO T-Display - azhinu/VESC-dash

github.com

reddit VESC display

GitHub - DDev247/vesc-m365: VESC Xiaomi m365 Pro dashboard + BMS interface

VESC Xiaomi m365 Pro dashboard + BMS interface. Contribute to DDev247/vesc-m365 development by creating an account on GitHub.

github.com

GitHub - NetworkDir/SimpleVescDisplay: 🚀 Simple VESC Display for ESP32-2432S028R

🚀 Simple VESC Display for ESP32-2432S028R. Contribute to NetworkDir/SimpleVescDisplay development by creating an account on GitHub.

github.com

casainho's opensourceebike project

VESC DISPLAY

Trampa Boards : Custom mountainboards, kiteboards and parts.

trampaboards.com

 

[yabert]

Here are some

Nice! Huge thank you for that!

 

[yabert]

Your duty cycle seems unhappy... It should be smooth at 95+is all working well at max power.

6.06 update is done and my duty cycle is still unhappy.
Is there some parameter to tweak?
Where I should start?

 

[yabert]

New high power reach: 16 kW at 1080 rpm
58V battery. Regen motor set at 1000 pA

I'm not sure what is overmodulation factor and why it could be bad to set it over 1.15, but there is huge difference in power and RPM in my set up.
I tried setting at 1.0, 1.15 and 1.30. New power record was at 1.3.

I clearly don't know why pA are so close to battery Amps when set at 1.3, but it seem nice.


1.0

1.15

1.3

 

[yabert]

I'm not sure what is overmodulation factor and why it could be bad to set it over 1.15, but there is huge difference in power and RPM in my set up.

I'm still fully interrested to learn more about VESC overmodulation if someone know more about this.

I've received the encoder MT6816 and manage to integrate it to Leaf motor.
I've rework the shaft to remove threads for the resolver.
Plan to use a 10x3 magnet on the shaft. Open to advice about how to fix it. Epoxy?

So, I have to build the encoder support and I will be able to test it.

 

[eee291]

I would test the encoder on the controller first. On the Hyundai hsg I replaced the M6 screw with a aluminum one. The magnet was glued with loctite. According to some other encoder datasheet, it is recommended that you use some non ferrous material to hold the magnet as iron/steel will change the shape of the magnetic field.

 

[yabert]

it is recommended that you use some non ferrous material to hold the magnet as iron/steel will change the shape of the magnetic field.

What? Do you mean the magnet cannot be fixe on the Leaf rotor steel axle?

 

[eee291]

Read the datasheet of the encoder I recommended a some posts back. I don't know how if it will still work with a metal backing. Maybe it will still work if the board is perfectly centered and close to the magnet.