Thursday, April 24, 2008
Magnetic Rotary Encoder Design
Hey RepRappers!
Inspired by Nopheads fantastic printing progress using his encoder, I decided that RepRap would greatly benefit by having an awesome, standardized rotary encoder board. The optical encoders can be finicky and hard to mount. Many RepRappers have also had great success in the past using AS50** family chips, so I decided to go that route.
I'm designing a new board around that AS5040 magnetic rotary encoder. Its nearly finished, I just need one more thing: Your input! This is one of the more advanced boards I've attempted, so if anyone has experience with the AS5040, or the Austria Microsystems magnetic rotary encoder chips in general, I'd love to hear you weight in.
Anyway, a few things to note:
* the encoder has 10 bits of resolution (1024 steps/resolution!!!)
* connector will be a 10-pin IDC header/connector to make life easy.
* the AS5040 offers many modes, and I've attempted to allow you to interface all of them. The modes are: quadrature (+index), PWM signal, analog signal, and digital shift-register.
* The pull-down resistor on CSN defaults it to quadrature mode
Thats about it. I've tested the quadrature wiring, and I'll be testing the rest this weekend. Assuming there are no bugs, I'll be laying out the board and sending it off for a prototype run next week.
Comments:
<< Home
Looks good.
A real SMT PCB would be really good - it'll make it much easier to work with. Add lots of locating marks to allow people to align the chip accurately.
Also, which side will the rotating magnet go? Above the chip, or on the other side of the PCB? If the latter, can you make the PCB single-sided, so there is a flush uncluttered face that could be bolted onto a flat plane in which the surface of the rotating magnet lies?
Either way, think carefully about mounting holes. They either need to be accurate enough to locate the PCB correctly (and probably also be a simple multiple of 10 number of mm apart), or need to be slots so you can move the thing about and then clamp it in place.
A real SMT PCB would be really good - it'll make it much easier to work with. Add lots of locating marks to allow people to align the chip accurately.
Also, which side will the rotating magnet go? Above the chip, or on the other side of the PCB? If the latter, can you make the PCB single-sided, so there is a flush uncluttered face that could be bolted onto a flat plane in which the surface of the rotating magnet lies?
Either way, think carefully about mounting holes. They either need to be accurate enough to locate the PCB correctly (and probably also be a simple multiple of 10 number of mm apart), or need to be slots so you can move the thing about and then clamp it in place.
I don't think the LED will work properly. The two outputs you have connected it to are open drain so they only pull current to ground. The LED and its series resistor should be in place of the 10K.
Post a Comment
<< Home
