Hello Guest it is October 25, 2021, 12:40:59 AM

Author Topic: Encoder Hacker Needed  (Read 6843 times)

0 Members and 1 Guest are viewing this topic.

Re: Encoder Hacker Needed
« Reply #10 on: March 05, 2008, 07:22:23 AM »
this pik-up seems to be equipep whit HALL sensors, the colored wire is positive power suply  the wire has no change resistance when aply magnetic field near is negative and the other two wires are chan A and chan B ouputs.

Josep

 
Re: Encoder Hacker Needed
« Reply #11 on: March 05, 2008, 07:33:44 AM »
Thanks Josep,
I may tinker with it later. I'll go with 5V and see what happens.
RC
Re: Encoder Hacker Needed
« Reply #12 on: March 05, 2008, 02:26:54 PM »
by all mean try it, however

if is was a simple AB encoder  why would the manufacturer not directly interface with the micro, since it has a timer/counter input
further if it is a simple AB, how does it create 5000 steps out of that toothed wheel which only has 60 or so teeth...

as said before my bet is that these are two magnetic sensors, quite likely coils and that they are exited by the custom chip
both are set at a specific distance from eachother and they can both see the same teeth of the wheel, by measuring the loading on the coils induced by the proximity of the teeth, current and timing and smart analog stuff this thing can interpolate the position.

alternatively it is a Hal effect sensor, and they measure the analog level altough that would require extensive calibration and be subject to noise (especially magnetic)

We have been playing with a magnetic sensor to measure current on busbars ( think 400A and above), great idea only the local earth magnetic fields affected the measurements...

A better explanation can be found at the newal site, the sphersyn system


 

Re: Encoder Hacker Needed
« Reply #13 on: March 05, 2008, 02:41:19 PM »
Thanks JB,
I realize now that it's not a simple encoder. This thing has a brain.
When the motor was working, it could be set to 50,000 PPR.
Thanks for the Spherosyn tip.
RC
Re: Encoder Hacker Needed
« Reply #14 on: March 05, 2008, 05:24:30 PM »
If a hall efect sensor would require extensive calibration and be subject to noise (especially magnetic) whi are used in spark ingnition in cars ??
or near coils and magnets in a brushless motor conmutation ??
in this two examples have a lot of electromagnetic disturbances and requiere very precision in trigering.
if this encoder have litle coils in the pik-up whi varies the resistance when aprox. to the magnetic disk??
i`m not enable a discuss on this theme (simply my poor opinion).

Josep
Re: Encoder Hacker Needed
« Reply #15 on: March 06, 2008, 02:54:40 AM »
first of all my last post seems harsh, it was not intended to be :-[

in most applications hall effects are used as switches, and thus either a schmitt trigger circuit is added or build-in
with the use of hall effect basically if it moves (as in cyclic) the electronics look at the flank of the signal, if it is not cyclic (like proximity) you look at a "arbitrary" level
and decide stop/move
generally in the electronics world , if you can reduce a real world analog signal into electronic timing signal it is always easier to measure/reproduce then a analog signal level

in order for the this encoder to work (and I was a factor of 10 out) it will have to interpolate to quite a small degree
so take the 50.000 CPR , that's 0.0072 degrees per click, so the positional change of any teeth is minute the change in distance change be teen the sensor and a teeth is even less, The hall effect sensor must thus be able to measure sub 0.001mm distance changes and translate in a proportional voltage level change.
this then has to be captured by the electronics, which if substantially below mV becomes hard to reproduce
this (analog 2 digital) is a large part of my dayjob and it gives me and my co-workers grey hears ..... :'(


Still these motors are cool .....intended
JB