Encoder for Micro Metal Gearmotor with extended shaft


I am looking for an encoder to be used with the 100:1 Micro Metal Gearmotor HP with Extended Motor Shaft.
I don’t want to use the pololu wheel encoder instead I wan’t to have an encoder on the extended shaft. Do anyone know of an good encoder for this?

Are you still interested in something like this? What kind of resolution are you looking for?


I am looking for an encoder for shaft of “100:1 micro metal gearmotor with extended motor shaft” item #2204.
the encoder should be small and fit the extended motor shaft NOT the wheel.


I’m also interested in those encoders. Micro-metal gear motors extended shaft is very thin, so it is dificult to find something that fits. Does Pololu or anybody know of something that can fit? I’m looking for an 128 pulses per revolution at least, to do a god speed control of the motor. But if somebody can advise me of an less ppr model, it will do the job at the moment.

Austria Microsystems AS5045 and AS5048 are nice sensor/processor chips for position encoding. They are just the chips, though – you have to mount a magnet on the axle, and mount the chip next to the magnet, and supply it with voltage and control signals, to get a full solution.

Digi-Key has eval boards: digikey.com/product-detail/e … ND/3694172
But they’re not in stock.
The chip itself is $13 in quantity 1: digikey.com/product-search/e … ?k=AS5048B

If you know how to design circuit boards in something like Eagle, you could perhaps send off a board to a place like oshpark.com/ to make one that’s small enough? Seems like a fair bit of work, though.

Hi jwatte, I have some AS5040, but they are difficult to work with. First you need special magnets, they need to be diagonally magnetized. Attachment to shaft and calibration it is not trivial. You also have to very acurately place the magnet at the exact point over the chip. And so on…

Too many dificulties. I was looking for an easier way to do the job.

Interesting! However, they do claim that AS5048 is easier to insert, doesn’t need hardware calibration, and is more error tolerant, so perhaps they learned :slight_smile:
I agree: if someone else already did all the hard design-build-test cycle, that saves a lot of time and is much more convenient!