Encoders w/128 pulses per turn?

File: rotaryenc_4r8u904r8.jpg (129KB, 1196x750px) Image search: [Google]

129KB, 1196x750px

I have never seen one that is thru-hole/board mount (like OP pic) that had more than 24 pulses. They wear out relatively quickly too, and don't move smooth because they have detents in them.

There are rotary encoders pretty cheap on the China sites. Looking fast I didn't find any with 128 steps per turn, most are 100, 200, 400 or 600 steps.
If you got a 400-step one then you could just use 3 pulses of the encoder as one 'actual' pulse, that would take 384 pulses on the encoder--which is pretty close to one actual turn.

Do note: the types of encoders pictured don't have detents, so they turn totally smooth.
And they don't have end stops either--they spin round and round.
They output 2-phase gray code and it is easiest (IMO) to just put a arduino (any 16mhz) on there to monitor the two encoder lines and put out the digital value on other pins. This is cheap and you can DIY it easily.

If you want an absolute optical encoder--one that *never* loses tracking and puts out a digital signal already--they cost a LOT of money. Figure $400-$500 at least, for just a 128-pulse unit. Absolute optical encoders are industrial-level gizmos with industrial-level prices; there are no cheap ones anywhere.

>>968941
>Figure $400-$500 at least, for just a 128-pulse unit.
Okay,,, I was a bit high in that estimate.
Here is an Omron-brand 256-pulse absolute encoder at Mouser for "only" $301
http://www.mouser.com/ProductDetail/Omron-Automation-and-Safety/E6CP-AG3C-256-2M/?qs=TwPrcXezm71FKQKJ%252bh0w3g%3D%3D

Omron makes a lot of different rotary encoders, but (looking on their website) the 256-pulse one is the lowest-pulse-count absolute one they appear to make.

A $12 China encoder and a $2 arduino pro is still way cheaper than $300, and can do -almost- the same thing.... The main advantage that an absolute encoder has is that it doesn't lose its position when powered down.

>>968898
You can have 128 positions or any other arbitrary number using a 24 step per rotation rotary encoder. A rotary encoder turns continuously and you keep track of position in code. You can count as high as you like before rolling over. You'll have to do multiple full rotations to go from 0-127, but that's fine if you show the number on something like an LCD display instead of printed on the panel behind the knob.

Honest question, why can't a potentiometer work here? Couldn't you read the analog value and send a midi signal accordingly? That would be a lot cheaper I suppose, but I don't know what I'm talking about.

>>969017
>Honest question, why can't a potentiometer work here? Couldn't you read the analog value and send a midi signal accordingly?
this doesn't work well when you want lots of tiny divisions, as the accuracy that is detectable is limited to the range that the pickup device's own voltage regulator works over.

You can use a stepper motor as a control knob too, if you want the cogging feel it has. A normal stepper motor has 50 "detents" and 200 pulses per turn.

>>968898
Use a planetary ratio converter to couple the control knob to the 20 step encoder. I've stripped a few of these out of old equipment. They can be used in either direction: increase or decrease knob to encoder ratio. You will want to set one up to turn the encoder 10 turns for each 1 knob turn.

The converters I have are made with ball bearings (as opposed to toothed gears). Because of this and the turns conversion, your device will not have absolute position information (the bearings do slip a bit). But you just track that in software.