Nedd cheap(er) panel mount optical encoders

Hi folks,

I'm having to repair a device that has six 1000ppr panel-mount optical quadrature encoders. The encoders are failing more and more because the device is a box which gets dropped on concrete floor a bit too often.

I'm looking for replacement encoders but at that resolution (500ppr would do, too) things get really industrial-grade and ball-beared with associated cost ($100+ apiece). The encoders are hand-operated with knobs so precision is not an issue.

Suggestions, anyone?

bob

If 'precision is not an issue', what are you doing using a device that can potentially resolve 4000 points round a circle?. One reason the devices have ball races, is because any shift in the shaft, will be seen as an encoder movement, and with a typical encoder of this type, a sideways movement of just 0.01mm, will be seen as a step. Consider if you could provide the input knobs and shafts yourself?. If so, you could use much cheaper 'bearingless' encoders, like the push-on ones from US Digital. Some of these are under $20 each. US Digital do some types that sound significantly cheaper than your current units, with the 'H1' (ball bearing unit up to 1250ppr), coming in at $57. If you could come down even lower on the resolution (unless the knob is very large, do you really think people can generate/resolve movements finer than perhaps 1 degree?), some of the digital potentiometer units (usually 90ppr, to give 1 degree resolvable), are also significantly cheaper.

Best Wishes

When it is reasonably large (say 2 inches diameter) it is amazing how well a round knob can work for fine control, if there is some feedback. An experienced radio operator can tune an SSB receiver within about

1/1000 of a turn (around 10 Hz resolution at 10 kHz/turn, for example). I suspect a motion control system might do even better, where the eyes are used for feedback rather than the ears.

Steve

Try US Digital

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Ah, this is interesting: The computer to which the current encoders are connected registers 1000 counts per turn, which is why I specified 1000 ppr. But I guess 250 is enough for me, then.

"Precision is not an issue" is, of course, an unprecise statement. What I mean is that I can live with possible loss of counts or similar errors. To my surprise I have found that 1000 steps per revolution is not completely overblown for a hand-operated device with visual feedback, but half the resolution is enough.

I guess that's the reason why it is hard to find the combination "many steps with plastic mechanics".

Absolutely. You'll be surprised when you try it (I was).

robert

Thanks!

Exactly, and the application in this case _is_ a motion control system.

robert