For those of you who have some insight, I’d like to ask if you know / guess what is the function of these hall sensors on a hoverboard motor.
It’s a Jetson hoverboard. There are three halls on the inside, which are obvious. However, I see three more pointing inwards towards the motor shaft, apparently doing nothing. Any of you have any idea what are these for?
Thank you.
Cheers,
Valentine
Is it the same on both wheels?
Yes, symmetrical. Both wheels have the same extra 3 sensors.
I’d expect there to be a magnet mounted to the axle, so the inwards facing halls could measure the orientation of the board vs. the wheels…
But maybe it is a case of a sensor board design being used on multiple models of hoverboard, with not always the same configuration in terms of which parts are used?
I didn’t find a magnet on the board.
Perhaps this is a combination of both, the board is for multiple models, and one of the models has a magnet to check on the wheel orientation?
What’s puzzling is that there are only 4 signal wires, for 6 sensors I’d expect 6 signal wires. This is very strange.
I don’t need the hall sensors, since I’m using an angle sensor but I was really wondering why they made it like that.
Thank you.
Strange indeed. It doesn’t looks like they’re positioned accurately enough to do anything useful even if there were signal lines for them and a magnet to sense.
Hello,
Maybe it simple is for an inrunner motor version, therefore an universal pcb.
The marked components are labeled ‘2n551’. Might it just be transistors used to amplify the sensor signal?
I think that’s for measuring the temperature
Hi @Valentine,
Do you know if it is feasible to place a magnetic sensor in this type of motors?
They seem to have plenty of torque at a reasonable price, but the built-in Hall sensor may not have much resolution.
Short answer yes, I’ve done it. Long answer – mechanically it’s really difficult, since the wheel is single pivot and you have no reference to put the magnet. So, what I did, was I came up with two solutions.
Placed a second pivot by literally gluing (or bolting) the pivot to the face of the hoverboard. Requires also ball-bearings, etc.
Designed from scratch (used a professional CAD program and made the parts from laser-sintered steel) and manufactured a very complex throughhole gearbox which goes on the back of the hoverboard and has inside a custom designed AS5047P PCB to hold the sensor. This requires microscopic ball-bearings and very fine motor skills and very high level of precision.
I don’t have the pictures handy to post here but you get the idea. Let me know if you want pics.
Cheers,
Valentine
Thanks for your answer 
More or less I figure out how you did it.
One thing I’ve been wanting to try is to replace the 3 digital halls with linear halls and see what sort of signals you get. They might be too saturated by proximity to the magnets, but hopefully two of the three would give good midway signals at any given time and could be transformed into an angle. I’ve got a bunch of 49E linear halls, so if anyone wants to take on the task let me know and I’ll mail you some.
Someone shared those pictures with me:
He had to file the shaft so the cables fit in the bearing:
Thanks for the info
Here is an documentation of how the typical hoverboard bldc hall sensor encoders are working that I found online:
https://www.digikey.de/de/blog/using-bldc-hall-sensors-as-position-encoders-part-1
and here is even an example code of how to interpret the sensors
https://www.digikey.de/de/blog/using-bldc-hall-sensors-as-position-encoders-part-3
