Browse Prior Art Database

Counting for Incremental Encoders

IP.com Disclosure Number: IPCOM000079639D
Original Publication Date: 1973-Aug-01
Included in the Prior Art Database: 2005-Feb-26
Document File: 4 page(s) / 58K

Publishing Venue

IBM

Related People

Portig, H: AUTHOR

Abstract

This arrangement protects against error insensitivity due to jitter back and forth between two adjacent counts of an incremental encoder, and against miscounts due to electrical noise because of feedback.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 41% of the total text.

Page 1 of 4

Counting for Incremental Encoders

This arrangement protects against error insensitivity due to jitter back and forth between two adjacent counts of an incremental encoder, and against miscounts due to electrical noise because of feedback.

An "incremental" encoder, Fig. 1a, provides discrimination of a high number of different positions, while using only two-bit positions and an external counter. By following the changes of the two bits, it is possible to tell whether the encoder moves in an "increasing-count" or a "decreasing-count" direction, and by how many counts it has moved.

Table A shows the pattern of bits for increasing counts. Table B shows the pattern of bits for decreasing counts. If the count were increasing and decreasing alternately, table C results:

(Image Omitted)

Note in table C that the transition from "01" to "10" implies direction of increasing count, since a decrease in count by one would never result in a change from 01 to 10.

In a typical "incremental-encoder and counter" system, a decision is made to count up or down every time that a bit is changed, and correspondingly increment or decrement the counter. If the encoder oscillates rapidly from, say, 01 to 10 and back, the transitions may occur so quickly that the counter loses track. If this occurs, the counter may read incorrectly because it is out of step with the encoder.

However, since the incremental encoder has four distinctly readable positions, it is possible to detect whether the counter has "lost track" of the encoder. For the examples shown above, if the counter value is written in base- two (e.g. binarily), then the last two bits of the counter value would agree with the two bits of the encoder.

(Image Omitted)

If the counter lost track by one (1) or two (2) or three (3) positions, either in the positive or the negative direction, the last two bits of the counter would no longer agree with the encoder, and it is known that an error existed, and the error may then be corrected.

When the last two bits of the encoder are compared with the encoder bits, there is a possibility of ambiguities. A -1 error is indistinguishable from a +3 error, a -2 error from a +2 error, or a -3 error from a +1 error. A +4 or a -4 error are unnoticed. Similar problems occur with errors greater than 4.

Since the main problem causing errors is the inability of the counter to follow jitter between two adjacent positions, the most common error will be by +1 or -1. As an example, if the counter says 6(10) = 110(2) and the encoder says 11, then the counter must be reading low, and it is corrected to 7(10) = 111(2).

1

Page 2 of 4

An incremental encoder is used here in unison with a counter tv obtain velocity and position formation. However, instead of using a "simple binary code", a "cyclic code" is used as in Fig. 1b. In this particular case of two-bit encoder the cyclic code is also a Gray code. This code has the characteristic that:
1) The two-bit bands are identical, but s...