
Interpret the Attribute Tables
Chapter 3
Rockwell Automation Publication MOTION-RM003I-EN-P - February 2018
91
Attribute Unit
Applicable Units
Semantics of Values
Filter Frequency Units
Hz
Hertz
Counts
Fundamental control unit for distance.
For example, feedback counts or planner counts.
See also
This table provides descriptions of the CIP Data Types related to the CIP Motion
Control Axis.
Data Type
Name
Data Type Code
(hex)
Description
Range
BOOL*
C1
Boolean
0 = FALSE; 1 = TRUE
SINT
C2
Short Integer
-128 SINT 127
INT
C3
Integer
-32768 INT 32767
DINT
C4
Double Integer
-2
31
DINT (2
31
– 1)
LINT
C5
Long Integer
-2
63
LINT (2
63
– 1)
USINT
C6
Unsigned Short Integer
0 USINT 255
UINT
C7
Unsigned Integer
0 UINT 65536
UDINT
C8
Unsigned Double Integer
0 UDINT (2
32
– 1)
ULINT
C9
Unsigned Long Integer
0 ULINT (2
64
– 1)
REAL
CA
Single Precision Float
See IEEE 754
LREAL
CB
Double Precision Float
See IEEE 754
BYTE
D1
bit string – 8-bits
N/A
WORD
D2
bit string – 16-bits
N/A
DWORD
D3
bit string – 32-bits
N/A
LWORD
D4
bit string – 64-bits
N/A
SHORT STRING
DA
{length, 1-byte characters[n]}
N/A
*
BOOL data type is defined by CIP standard to be an 8-bit unsigned integer with enumeration of 0 for False and 1 for
True.
See also
The variations in Control Mode and Control Method result in a set of basic
Device Function Codes that help organize the many attributes of the Motion
Control Axis. Device Function Codes are designated by using a letter identifier or
a combination that you can use to determine what attributes are required for
implementation of a given CIP Motion device. The list of Device Function Codes
is as follows:
CIP Data Types
Device Function Codes