Rotary encoders

[Michael GilliganParticipant @michaelgilligan61133]

(3)

Nice thought … but not quite that simple.

The track on a CD is a single spiral; and the "pits" have constant linear spacing.

MichaelG.

[Andy from WorkshopshedParticipant @andyfromworkshopshed]

MikeW, I know of Renishaw from a long time back when I was an apprentice. Their kit is top quality. I just had a look on their site and the magnetic modules and chips have a "AM" prefix on them so it's quite likely they are sourced from Austria too. Obviously the advantage of having an encapsulated device is all the alignment and positioning of the magnet issues have already been resolved.

One thing to note is the use of a non magnetic holder for the magnet so I'll likely use plastic as I've got a rod of that.

There are some nice sections in their data sheets about linearity which I'll re-read as it's likely to apply to my chip too. Which chip in particular were you referring to?

http://www.renishaw.com/en/magnetic-rotary-encoders–9801

There's also a point about the limit of resolution being determined by the electrical noise from the Hall sensors so that's likely to be the case for all sensors using this technique, 13bits being the limit.

Following on from the CD idea and discussions on mice, perhaps an optical mouse could be used to "read" an LP without any physical contact to the disk.

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Andy – look in pm's

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

**LINK**

+ multiple further links and data / explanatory download pdf's .

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

CD encoder disc – this was the general idea :

(1) Save data onto disc in patterns of 0 1 0 1 etc etc to give the required number of counts .

(2) Data is stored in sectors – each successive sector 0 or 1.

(3) All read head has to do is detect the transitions from 0 to 1 to 0 etc as each sector passes .

(4) Individual read / write cells on the disc are minute compared with the size of the sectors needed even at quite high resolution .

(5) Every sector contains a very large number of cells each with same logic state .

(6) A relatively coarse resolution read device could detect average over a number of cells with no loss of accuracy and be very insensitive to random bad cells

Or a standard resolution read head could read multiple cells and do error checking .

(6) Any normal tracking system on the disc could be used but with complete command control available and suitable choice of disc technology a task specific tracking system could be devised .

(7) There is no real need to use modern high density discs – older technologies might work quite adequately .

(8) Auxilliary encoding could make availabe useful data eg sectors could be numbered or one identified as the home position .

(9) Sector numbering would make possible absolute encoding rather than incremental .

(10) Gray coding would also be possible .

MikeW

Edited By MICHAEL WILLIAMS on 16/05/2013 13:18:31

[Anonymous]

As it says on the Renishaw website, their ICs are made by RLS, in Slovenia. RLS are an associate company of Renishaw, which I assume means that Renishaw have invested, in order to ensure continuity of supply chain.

From the datasheets on the Renishaw website their magnetic rotary encoders are claiming accuracies on the order of 0.1-0.2°, which is about what one would expect, looking at the datasheets for the RLS chips. Not surprisingly that's consistent with what you might expect from the AMS ICs. Of course the Renishaw optical encoders are orders of magnitude more accurate.

It would be rather difficult to write a distinct data pattern to a CDROM that actually coincided directly with the pattern of pits as CDROMs use a non-return-to-zero inverted encoding scheme, along with cross interleaved Reed-Solomn coding to minimise the effects of random and burst errors during the read process. In simple parlance the data to be written is encoded across a significant part of the CDROM, so that despite damage to a localised area the data can still be reconstructed.

Regards,

Andrew

[Cornish JackParticipant @cornishjack]

Ummmm??? (the youngsters won't understand this) "What did Horace say, Winnie?

Rgds

Bill

[Anonymous]

Correct, I had to google it. Way before my time.

Andrew

[Author]

Posts