Are there any off the shelf chips like the mct8316z, that offer inputs for four hall sensors as opposed to 3? Or anyway to hack one of them?

 

Why do you need 4 sensors?

 

For increased resolution for increased starting/direction change smoothness m, robustness, and snappiness.

 

How does that work???

 

If you have 3 hall sensors they are 120 degree out of phase. If you have four you have 90 degree. You get 33% more resolution.

 

Is there any off the shelf motors with that option??

 

I haven’t seen one before but we are designing/building our own.

 

I would be interested to see the switching sequence, I would have thought if this was an asset, there would be at least one manuf that used it?
Does it apply to motors in the 3ph mode as well as BLDC?

 

If you have 3 hall sensors they are 120 degree out of phase. If you have four you have 90 degree. You get 33% more resolution.

https://www.mathworks.com/help/sps/ref/pmsmfourphase.html

What is the resolution of a three phase motor?

 

For increased resolution for increased starting/direction change smoothness m, robustness, and snappiness.

I am used to using AC & BLDC versions in CNC servo applications where I can achieve fractions of a thou resolution using PID loop. The DIY route is achieved with Galil multi axis/robotic products.

 

I am used to using AC & BLDC versions in CNC servo applications where I can achieve fractions of a thou resolution using PID loop. The DIY route is achieved with Galil multi axis/robotic products.

The OP seems to think there is some modern advantage to poly-phase (>3) motors. There once was before modern electronics. OEMs been using accelerometer feedback with linear synchronous motors to control wafer processing linearity and stepping for 40+ years using trapezoidal signal drivers.

A 30+ yo machine linear beam scanner (to generate a X scan correction factor) using 3-phase AC servo motor with encoder positioning.

With modern FoC drivers, with the ability to easily generate up to 32-bit resolution 3-phase sinusoidal drives signals from digital synthesis, there is no practical advantage for 4 phase drive systems.

 

Use a resolver for motor position and speed feedback, as has been done for years.

 

Use a resolver for motor position and speed feedback, as has been done for years.

I tended toward quadrature encoders.

 

If you really want to do this your best bet is an FPGA.

 

If a resolver doesn't float your boat, a Hyperface encoder should achieve all your requirements ?

 

For years the quadrature encoder has been used in most comercial and DIY CNC/ motion control etc.

 

The problem with a quadrature encoder in this application is, it doesn't provide rotor position.
Which the OP requires.

A simple resolver resolves that issue .

A Hyperface encoder provides a digital level of position information. (It actually uses a resolver in part).

 

I didnt read that in his requirement?
Quote: " For increased resolution for increased starting/direction change smoothness m, robustness, and snappiness. "

Once The quadrature version is referenced, it can provide a position.?

 

Use a resolver for motor position and speed feedback, as has been done for years.

We use a resolver and encoder for high precision direct drive motor systems. A high resolution encoder (absolute or relative) is better for positioning, and today, the quadrature resolver signal is converter to digital.
https://forum.allaboutcircuits.com/...-motors-and-their-accuracy.100445/post-754228


https://www.celeramotion.com/optical-sensors/


360,000 counts per shaft revolution ( .001° or 3.6 arc seconds ) accuracy.



26-bit encoder ring absolute encoder for a telescope mount.

https://www.renishaw.com/en/resolute-encoder-series--37823

 

I didnt read that in his requirement?
Quote: " For increased resolution for increased starting/direction change smoothness m, robustness, and snappiness. "

Once The quadrature version is referenced, it can provide a position.?

No.

A incremental quadrature encoder cannot provide position data.
The maximum it can provide is, how much a rotor has moved and in which direction.

What is meant by quadrature version referenced?
If the motor has to move to achieve this, the 'starting/direction change smoothness' requirement' cannot be met as the rotor will already have moved?