Delta Tau GEO BRICK LV User Manual
Turbo PMAC User Manual
188
Setting Up the Servo Loop
Motor 1 is the first source motor (the row motor), using its desired position; each row represents a row
span of counts of positions of Motor 1, and each column represents a constant raw position of Motor 1.
span of counts of positions of Motor 1, and each column represents a constant raw position of Motor 1.
Motor 2 is the second source motor (the column motor), using its desired position; each column represents a
column span of counts of Motor 2, and each row represents a constant raw position of Motor 2.
column span of counts of Motor 2, and each row represents a constant raw position of Motor 2.
Motor 3 is the target motor; the corrections are applied to Motor 3.
The span of each row is 20,000 counts, so there is a spacing of 20,000/20 = 1000 counts (of Motor 1)
between entries along the row dimension. The span of each column is 15,000 counts, so there is a spacing
of 15,000/15 = 1000 counts (of Motor 2) between entries along the column dimension.
between entries along the row dimension. The span of each column is 15,000 counts, so there is a spacing
of 15,000/15 = 1000 counts (of Motor 2) between entries along the column dimension.
The next (15+1)*(20+1)-1=335 numerical constants sent to Turbo PMAC are entered into this table.
Why this number of entries? Because there are entries in each row and column at both the zero position
and the maximum position (hence the n+1 terms), but there is no explicit entry at the origin of both source
motors (hence the final -1).
Why this number of entries? Because there are entries in each row and column at both the zero position
and the maximum position (hence the n+1 terms), but there is no explicit entry at the origin of both source
motors (hence the final -1).
In this example, the first entry would be the correction at Motor 1 (raw) position 1000, Motor 2 (raw)
position 0 – i.e. at (1000, 0). The second entry would be at (2000, 0). The 20
position 0 – i.e. at (1000, 0). The second entry would be at (2000, 0). The 20
th
entry would be at (20,000,
0). The 21
st
entry would be at (0, 1000), the 22
nd
at (1000, 1000), and so on. The 335
th
and last entry of
the table would be the correction at (20,000, 15,000). If I30 were set to 1, this value would also be the
correction at (0, 0). Typically the correction at the origin is made zero by definition, to serve as the
reference point for the other corrections.
correction at (0, 0). Typically the correction at the origin is made zero by definition, to serve as the
reference point for the other corrections.
If there is any possibility of motion going past the declared span of the table, whether for purposeful
rollover or not, the entries at both ends of each row should be the same; likewise for each column.
Otherwise, there will be a discontinuity in the correction at the edge of the table.
rollover or not, the entries at both ends of each row should be the same; likewise for each column.
Otherwise, there will be a discontinuity in the correction at the edge of the table.
Note three things to be careful about in the entry of a 2D table. First, the number of rows and number of
columns is separated by a period, not a comma. Second, the number of rows (15 in the above example) is
entered first, before the number of columns, but the spacing (in counts) between rows is determined by
the span of a column (15,000 in the above example), which is entered after the span of a row. Finally, to
permit efficient computation in Turbo PMAC, both Row and Column 0 as well as Row and Column n
must be entered.
2D Table Example
The following example shows the entry of a simple 2D table, shown in a form that makes for easy reading
by a user. (Turbo PMAC does not require that the table be entered with each row on a separate line, but
this is recommended for readability.) Note that with an implied correction value of 0 for the zeroth entry,
Rows 0 and 4 are identical, as are Columns 0 and 5.
columns is separated by a period, not a comma. Second, the number of rows (15 in the above example) is
entered first, before the number of columns, but the spacing (in counts) between rows is determined by
the span of a column (15,000 in the above example), which is entered after the span of a row. Finally, to
permit efficient computation in Turbo PMAC, both Row and Column 0 as well as Row and Column n
must be entered.
2D Table Example
The following example shows the entry of a simple 2D table, shown in a form that makes for easy reading
by a user. (Turbo PMAC does not require that the table be entered with each row on a separate line, but
this is recommended for readability.) Note that with an implied correction value of 0 for the zeroth entry,
Rows 0 and 4 are identical, as are Columns 0 and 5.
#3 DEFINE COMP 4.5, #1D, #2D, #3, 50000, 40000
38 45 –22 –35 0
; Row 0, Columns 1-5
24 56 13 –34 –8 24
; Row 1, Columns 0-5
18 43 –9 –65 32 18
; Row 2, Columns 0-5
-6 28 22 –38 12 –6
; Row 3, Columns 0-5
0 38 45 –22 –35 0
; Row 4, Columns 0-5
In this example each row covers the span of Motor 1 (0 – 50,000 counts) at a constant position of Motor
2; each column covers the span of Motor 2 (0 – 40,000 counts) at a constant position Motor 1.
2; each column covers the span of Motor 2 (0 – 40,000 counts) at a constant position Motor 1.
Enabling and Disabling Tables
All position compensation tables (as well as backlash and torque compensation tables) are enabled when
I51 is set to 1. All of these tables are disabled when I51 is set to 0.
I51 is set to 1. All of these tables are disabled when I51 is set to 0.