Delta Tau GEO BRICK LV Benutzerhandbuch
Turbo PMAC User Manual
144
Motor Compensation Tables and Constants
•
$2: The 48 possible bits are from the Y-register at the specified address and the Y-register at the next
higher address. The conversion table performs no filtering.
higher address. The conversion table performs no filtering.
•
$3: The 48 possible bits are from the Y-register at the specified address and the Y-register at the next
higher address. The conversion table can perform filtering of the source data.
higher address. The conversion table can perform filtering of the source data.
•
$6: The 48 possible bits are from the Y-register at the specified address and the X-register at the
same address. The conversion table performs no filtering.
same address. The conversion table performs no filtering.
•
$7: The 48 possible bits are from the Y-register at the specified address and the X-register at the
same address. The conversion table can perform filtering of the source data.
same address. The conversion table can perform filtering of the source data.
•
$F: The data is in byte-wide pieces at the Y-register specified address and adjacent higher
address(es). In this case, the first hex digit of the entry’s second setup line must be either $2 (no
filtering) or $3 (with filtering).
filtering) or $3 (with filtering).
In all of these cases, the bit-19 mode-switch bit in the first setup line of the entry is set to 0 to place the
LSB of the source data in bit 5 of the result, where Turbo PMAC firmware treats it as a count. If the
mode-switch bit is set to 1, the source LSB is placed in bit 0 of the result, where Turbo PMAC firmware
treats it as 1/32 of a count. This alternate setting can be used for very high-resolution feedback, where
numerical velocity saturation could be a problem.
LSB of the source data in bit 5 of the result, where Turbo PMAC firmware treats it as a count. If the
mode-switch bit is set to 1, the source LSB is placed in bit 0 of the result, where Turbo PMAC firmware
treats it as 1/32 of a count. This alternate setting can be used for very high-resolution feedback, where
numerical velocity saturation could be a problem.
Bits 0 – 18 of the first setup line of the entry specify the address of the source data. In the case of
methods $2 and $3, or $F/$2 and $F/$3, some of the source data could also come from the next higher
addresses.
methods $2 and $3, or $F/$2 and $F/$3, some of the source data could also come from the next higher
addresses.
The second setup line of the entry specifies which part of the source address data is used. This line is split
into four parts: the first hex digit, the second and third hex digits, the fourth hex digit, and finally the fifth
and sixth hex digits.
Second Setup Line for Parallel Data Entries
into four parts: the first hex digit, the second and third hex digits, the fourth hex digit, and finally the fifth
and sixth hex digits.
Second Setup Line for Parallel Data Entries
Hex Digits
1
2 & 3
4
5&6
Contents
Aux Meth
Bit Width
Byte
LSB Location
In the case of methods $2, $3, $6, and $7, the first digit is always 0. In the case of method $F, the first
digit is $2 if no filtering is to be done; it is $3 if filtering is to be done.
digit is $2 if no filtering is to be done; it is $3 if filtering is to be done.
The second and third digits specify the bit width – how many bits are to be used.
In the case of methods $2, $3, $6, and $7, the fourth digit is always 0. In the case of method $F, the
fourth digit specifies which byte the data comes from: 0 for the low byte, 1 for the middle byte, and 2 for
the high byte.
fourth digit specifies which byte the data comes from: 0 for the low byte, 1 for the middle byte, and 2 for
the high byte.
The last two digits specify which bit of the source byte or word is to be used as the LSB. For byte-wide
data this is always in the range $00 to $07. For data from a 48-bit field, this value could be up to value of
48 minus the bit width.
data this is always in the range $00 to $07. For data from a 48-bit field, this value could be up to value of
48 minus the bit width.
If the source data is to be filtered, there is a third setup line in the entry, which specifies the maximum
change in the source data (in units of LSBs of the used source data) in a single servo cycle that is to be
regarded as real. If the source changes by more than this (maybe due to noise in a high bit), the result will
change by this amount only. This is an important protection against noise and other possible anomalies.
change in the source data (in units of LSBs of the used source data) in a single servo cycle that is to be
regarded as real. If the source changes by more than this (maybe due to noise in a high bit), the result will
change by this amount only. This is an important protection against noise and other possible anomalies.