Rotary encoders

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

The subject of making or buying rotary encoders for DIY projects comes up often .

In reality they are quite easy to make in a variety of ways but reading the ' Aldi calipers ' thread suggested another possibility .

The working principle of all these types of calipers is simple and could be adapted to rotary position encoding . All that would be needed is two round printed circuit boards – one with a circular form of the scale itself and one with two ( or better four ) reader segments arranged also in a circular format ..

Would need good mechanical design to ensure one PCB ran true relative to the other but that's easy enough .. Overall size could be as required but there is the possibility of making very small ones .

Michael Williams .

Edited By MICHAEL WILLIAMS on 28/04/2013 12:48:49

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

[Ian PParticipant @ianp]

Michael

Whilst it would be possible to make your own it would be no easy task. The internals of a digital caliper look deceptively simple but in reality have quite complex electronic circuitry and PCBs with very precise track layouts.

Rotary encoders using capacitive techniques already exist, see

**LINK**

Ian P

[John Stevenson 1Participant @johnstevenson1]

Another way is what is used for threading on a CNC lathe.

This has 90 slots and 3 hall effect switches the large middle one detects the long slot as an index pulse. The two outer ones detect leading edge and trailing edge of a slot, doesn't need to be the same slot even.

The electronics then read this as a quadrature signal so those 90 pulses are now 360 per rev.

Any number can be chosen provided it fits on the disc.

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

I'm familiar with existing designs of rotary encoder .

[Ian PParticipant @ianp]

Posted by MICHAEL WILLIAMS on 28/04/2013 13:16:09:

I'm familiar with existing designs of rotary encoder .

Michael

I wasn't doubting your knowledge but if one was considering making a DIY encoder, I think one working on capacitance might be the one of last resort.

Ian P

[Sub MandrelParticipant @submandrel]

The calipers use an analogue quadrature signal that allows resolution far higher (say 100-200 times) than the size of the capacitive elements.

A standard diogital quadrature decoder only allows a resolution of four times the 'tooth spacing'.

In principle I imagine you could peel the sensor ofl a large digital caliper/DRO and fit it around the rim of a circular wheel, but there might be issuse at the join and the reading head might not adapt to any significant curve.

Neil

[Les Jones 1Participant @lesjones1]

Hi Neil,
I think these angle displays use tha same princple as digital calipers.

Les.

[Sub MandrelParticipant @submandrel]

Thanks les

You may well be right, they have a resolution of 3600 steps which would seem the same order as caliper.

"Technical information coming soon"

Neil

[jason udallParticipant @jasonudall57142]

One thought.. digital calipers will have a ( fairly low) read rate..which in rotary mode might be readily exceeded….

part of why rotary encs. are more power hungry and expensive.. but for low speeds and experimental purposes why not ( the digi mikes I have scraped disassembled have a capacitive rotary encoder and count /resolve rotations so a capacitive method is certainly valid at low rpm.)

[jason udallParticipant @jasonudall57142]

thought 2

how about the digical reading a cam?

[Ian PParticipant @ianp]

Posted by jason udall on 29/04/2013 12:52:02:

thought 2

how about the digical reading a cam?

Jason

Please help my brain work this out!

If digical is not a misspelling of digital, what is reading what? Is cam something to do with a camera looking at a vernier scale?

Ian P

[jason udallParticipant @jasonudall57142]

Sorry Ian..just a bit too terse thattime eh?

Digital caliper "measuring" some sortof cam( varing radius thing) thus translating varing angle to displacment..indicated by caliper…….

just an idea …probably not worth the amplifiction..

regards

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

I find that a carefully measured dose of Felyn Foel Double Dragon helps enormously when trying to understand knotty problems :

1 pint – everything becomes clear and you get that calm and slightly smug feeling of having solved a problem .

2 pints – everything starts to become very mixed up again and truth and reality become both uncertain and irrelevant . Random motions of person and room tend to occur .

3 pints – you won't care any more .

 

Edited By MICHAEL WILLIAMS on 30/04/2013 07:58:38

[Ian S CParticipant @iansc]

Don't know if it's relivent to the subject, but one bit of research I did for one thread on here a year or two ago was that an optical mouse actually uses a vidio camera as its sensing unit. Ian S C

Edited By Ian S C on 01/05/2013 12:00:29

[Sub MandrelParticipant @submandrel]

Hi Ian,

I would dearly love to find an app note for how to hijack those cameras.

Neil

[jason udallParticipant @jasonudall57142]

Posted by Stub Mandrel on 01/05/2013 20:15:16:

Hi Ian,

I would dearly love to find an app note for how to hijack those cameras.

Neil

saw (youtube) one using an audrino..seems to be 1 bit 6 x6…..but within that does work….

[Ian S CParticipant @iansc]

Put into google Hacking an optical mouse, and you will find what there is to know, Like Micro positioning using a cheap mouse. 'Fraid the technoligy is a wee bit above me for now, if theres anthing there worthwhile someone will find it. Ian S C

[Andy from WorkshopshedParticipant @andyfromworkshopshed]

I've been doing some experiments with optical mice, they seem quite fussy with regards aligning with a suitable wheel. In theory they can do 1000 DPI which should be plenty if it can be made to work.

I've also been looking at other options such as a multi-turn potentiometer and a rather clever magnetic chip.

That Arduino video Jason mentioned might have been this one that I made last year, if not I'd be interested to know

[Andy from WorkshopshedParticipant @andyfromworkshopshed]

I've been doing some experiments with optical mice, they seem quite fussy with regards aligning with a suitable wheel. In theory they can do 1000 DPI which should be plenty if it can be made to work.

I've also been looking at other options such as a multi-turn potentiometer and a rather clever magnetic chip.

That Arduino video Jason mentioned might have been this one that I made last year, if not I'd be interested to know

[Andy from WorkshopshedParticipant @andyfromworkshopshed]

I've been doing some experiments with optical mice, they seem quite fussy with regards aligning with a suitable wheel. In theory they can do 1000 DPI which should be plenty if it can be made to work.

I've also been looking at other options such as a multi-turn potentiometer and a rather clever magnetic chip.

That Arduino video Jason mentioned might have been this one that I made last year, if not I'd be interested to know

[jason udallParticipant @jasonudall57142]

Andy.. no not that one…will try any track down the one I found…it revolved around using mouse sensor as looow res. camera

[jason udallParticipant @jasonudall57142]

Andy.. no not that one…will try any track down the one I found…it revolved around using mouse sensor as looow res. camera

[jason udallParticipant @jasonudall57142]

Andy.. no not that one…will try any track down the one I found…it revolved around using mouse sensor as looow res. camera

[Andy from WorkshopshedParticipant @andyfromworkshopshed]

Ah yes, I've heard that can be done, understand now. Here's the link http://www.youtube.com/watch?v=bci7Gi05BNc

Edited By Andy from Workshopshed on 09/05/2013 17:17:45

[Author]

Posts