Parker ACR9000, Руководство по установке

Страница: 139 / 198

Parker Hannifin
or maximum number (512) of digital I/O is reached. For example, the first
digital input on node 0 is bit 11520, and the first digital input on node 2 is bit
13568.
Each node will have an information parameter block, described later in this
text. This block will contain, among other things, the number of bytes of
digital inputs and outputs. Digital I/O are assigned in blocks of eight, so the
number of bits assigned to each node is a multiple of eight. For example,
suppose node 2 has 12 digital inputs. Node 2’s inputs would be bits 13568-
13579, even though the node status parameter indicates that it has two bytes
of inputs. The same numbering rules apply to digital outputs.
Analog Inputs and Outputs
Analog inputs and outputs are implemented by ADCs and DACs respectively,
and unlike digital I/O, the analog values represent something with units and a
range. For example, a DAC might assert –5V to 5V, or 0-20 mA, or some
range of pressure, force, or speed. The ADCs and DACs also have variable
binary resolution, e.g., 10, 12, 14 or 16 bits. All CANopen values are left
shifted to occupy the entire 16 bits as a two’s complement signed number,
even if the actual ADC or DAC is less than 16 bits. This does not increase
the analog resolution. In addition, the sign of the resulting 16-bit number is
the same as the sign of the physical quantity it represents instead of being
offset. A value of 32767 represents full scale positive for the device, and -
32768 represents full scale negative for the device.
For example a 0-10V DAC would take values of 0-32767, and a ±10V device
would take values of –32768 to 32767. However, a ±5V device would also
take values of –32768 to 32767. To translate from this raw binary number to
the range and units being controlled or measured, the ACR90x0 employs
entered offsets and gains.
An offset has the same units as the user units of the analog value, for
example volts or milliamps, and translates the center of the analog range to a
value that allows a gain to be applied. A DAC gain has the units of full-scale
binary resolution per user unit. The DAC range is 16-bit or 65536 DAC
counts, regardless of the actual DAC resolution.
For example, suppose a 12-bit DAC asserts –10V to +10V, where a value of
32768 will assert –10V and 32767 will 10V. In this case, the offset is
0V, and the gain is (65536/20 = 3276.8). If the user wants to assert 7.5V, a
value of 7.5 *3276.8 = 24576 must be written to the DAC.
The process is different for an ADC. An ADC gain has the units of full-scale
user units. For example, if the input of the analog device were a maximum of
+/- 10V, then the gain would be 10. Alternatively, if the input of the analog
device were a maximum of +/- 20ma, then the gain would be 20. Internally
the raw analog count value is normalized such that +/-1.0 represents full
scale positive and negative before the user gain is applied, and user offset
added.
The ACR90x0 automatically performs this arithmetic so that the analog
values appear to the user as user units, not raw DAC or ADC counts. The
user must know the analog range of the DAC or ADC in order to calculate the
appropriate gain for entry into the ACR90x0 parameter structure. Offset
values will usually be zero unless an actual physical offset is required. The
ACR90x0 uses default values for gains and offsets if the user does not
overwrite the defaults. All default-offset values are zero. All default ADC
gains are ten (10.0), and all default DAC gains are 3276.8.
Appendix E CANopen 139
www.comoso.com