Autosteer - Brushless Motor

These look interesting for very simple, safe steer motor usage? Obviously the correct wattage would need to be selected. This seems hugely powerful!!

1 Like

I really think it works, maybe a little expensive :rofl:. I was thinking of trying a hooverboard wheel, you can find it used and cheap, and building it a steering column adapter and a low steering wheel maybe it could work :thinking:

What do you think about this servo motor:

@Alan.Webb @giotoma
Yes looks interesting as a direct drive but quite expensive. I initially thought about the washing machine direct drive motors, but they would probably need rewinding for lower voltage. In the end I looked at 6354 120kv and 6374 brushless motors with sensor, using a 4:1 gear ratio with the main gear bolted to a universal steering wheel adapter:
71 tooth herringbone gears 101pcd
//open scad code (add it to the end of gears.scad):
herringbone_gear (modul=2, tooth_number=71, width=20, bore=58, pressure_angle=20, helix_angle=20, optimized=false);
//weight/material saving hole pattern
for (r=[0:60:360]){
//bolt hole pattern
for (r=[0:60:300]){

The bolt hole patterns should fit standard 70mm, 74mm and 101 mm steering wheel bolt patterns.
I did some tests on starting torque and these compact motors will lift a 2kg load wrapped round the 63mm diameter body at just 1amp at 12 volts using the flipsky vesc. So with a 4:1 reduction the torque available is massive considering that my steering wheel can be turned with just 1kg force at the rim. It would probably need the torque or current limiting in the vesc. The gear ratio will help reduce any cogging effects and match the speed requirements better, and bringing the motor well clear of the adapter hub.
Adapter spacer hubs are readily available on ebay etc in 2, 3,and 4 inch. You might need to cannibalise your steering wheel for the center hub fixing if you can’t get or make one as they come in a variety of non standard flavours , splined, tapered, keywayed etc.
PS only in design phase at present, I might change the helix angle to 0 to give a straight spur gear, which would simplify assembly and alignment problems.

My steer motor has current limiting applied, in the voltage step up unit. It makes steer less aggressive on drive components and protects the IBT2. I run a 12v Gimson motor at 22 volts. It is VERY powerful at 22v and plenty quick enough!

I like the direct drive as I figure it is probably the safest steer method apart from the friction steering wheel rim motors.

The multi motor designs made by @GregFaRmTeK look really neat too but there is still some steer jamming possibility if one roller was to break up for instance.

It’s all a personal preference thing really because you have to be responsible for your system should there be an accident and the possibility of it being implicated.

Selling semi amateur steer motor systems is a HUGE NO NO as the legal implications are wide ranging.

1 Like

Agree you need to current limit , I was just surprised how little mean current was needed, I found the skateboard outrunners 63mm diameter normally designed for 24-48 volts work more than adequately at just 12volts without stepup. The power needed to propel a skateboard being vastly more than that required to turn a steering wheel.

My voltage hike was purely for extra speed. The Gimson motor was a bit slow on headland turns. After reading an article on motor voltages I realised with our very low duty cycle it would be perfectly safe at 24+ volts.

So I guess the Gimson is a DC motor with gearhead, hence the need to speed it up. I am looking at a single brushless motor, no gearhead. A single 120kv brushless motor at 12 volts gives 12x120=1440 rpm hence the need for final gear reduction of 4:1 giving potentially 6 revs/sec at the steering wheel possibly too high, but this can always be limited in the VESC. Torque is more than adequate. I like the friction drive of Profile - GregFaRmTeK - AgOpenGPS it should reduce gear noise and tolerate for slight mounting eccentricity.

Looks cool. Nice rs485 interface. Would need running at 24 volts to get 4revs/sec. Torque at 5N-M is only just OK, but not much safety margin, and might depend on just how good the power steering is. I measured about 2 N-M to turn a steering wheel on one machine with power steering on. I would feel happier with 10N-M for the motor.

I made some test using a shorter motor, the 6354 190kv (63 mm diameter, 54 mm length) with a compact VESC Flipsky Mini FSESC4.20 50A.
They are all available with EU shipping on banggood.

20210103_140006 low
With very little tweaking using FOC and the hall rotation sensor, the duty cycle seems very usable.
The minimum stable speed is around 30 tpm without load (2-3% duty cycle, below torque not stable)
In this mode , the motor behavior is very similar to a brushed motor with variable input voltage, with smooth torque even when locking the motor to zero speed.

motor.txt (7,9 Ko)
Config moteur en txt au lieu de xml car pas autorisé par le forum.

The current from the battery and to the motor can be easily limited in the software (to avoid blowing fuse or wires). I’ve limited my maximum motor/battery current to 25A for most of my test.
This limit also the maximum available torque to avoid mechanical damage and limit injuries risk when testing (be careful, this is already power full).
The motor is capable of 65 A continues and the ESC 50A continuous, so i’ve a lot of margin. The esc has an integrated temperature safety measurement. (The one on the motor does not work).

The maximum speed can be adjusted with limiting the duty cycle below the standard 95% and the maximum no load rpm.
With my discharge battery at 9.8V, i can easily achive 1500 tpm with no load, so more than enough with the same gears as the phidgets without needing 24V.

The motor seems quite silent but I should use a better support.

The motors turns easily when not required to run. I can launch the motor by hand for a few turns.

This need to be tested on a real autosteer system but this seems a very good compact alternative to the phidget.

Few remarks (from this motor) :
Sensored motor is mandatory for me
FOC control mode with hall sensor give the smoothest torque output
BLDC with hall sensor (sensored or hybrid) is more noisy and torque is not stable when stalled.
HFI is smoothed if motor is not stalled but very noisy at low speed.
FOC without sensor does not provide a stable torque when stalled.

Low Kv
Low kv motor is important to have the most torque at the lowest current.
More than 25 A is not so easy to get reliably from a tractor.

Smaller motors
Smaller motor, with kv equal or lower should be usable.

Speed regulation
I did not test speed regulation extensively as the duty cycle was behaving the same as a dc motor and seams to be sufficient.
It will need manual tweaking with load applied to behave correctly.

Dust protection
This motor doesn’t have ventilation holes and is designed for skateboard use so it should not be sensible to dust in our tractor cab.

Thank you for your feedback on the topic, we are getting improvements :slight_smile:


Heii everyone,

I like the Idea to use this Motor. Does anyone have testet it yet?
Regards, Peter

Look at this cerea guy:


Ecco qui un altro video che mi ha mandato Luca, per farmi vedere di quanto è soddisfatto del kit - YouTube

May be this brushless motor works also for AOG

My system works with AGO.

If you like have a look at Using AgOpenGPS with a brushless motor - YouTube


This SteerWheel system you have bought in Aliexpress?

Yes, thats right.

maybe this is the missing link:
The HT 1225 seems even stronger


Can you Post the Link of the Motor?
Regards, Peter

Yes I think it can work, this week I will also test with Aog, but it is not a standard motor. With cerea work perfectly look this

Same motor Motore brushless 6374 1,8Kw utilizzato per Cerea autosteer - YouTube

there is this one too but the price :hot_face: