diff --git a/iris/publish-model.conf b/iris/publish-model.conf
index 8201d3c..9cc177d 100644
--- a/iris/publish-model.conf
+++ b/iris/publish-model.conf
@@ -11,11 +11,12 @@ publish {
The entire event is published atomically so that all published subcomponents are synchronized.
- *Discussion: For sidereal tracking, the maximum rotation velocity of ADC prisms caused by the orientation change (= the change rate of the parallactic angle) is 0.225 deg/s, which happens when observing a target that transits 1 degree south from the zenith. The current required position accuracy is [+/- 0.28 deg](https://docushare.tmt.org/docushare/dsweb/Get/Document-57487/DN%20IRIS.IMG.ADC_REL01.pdf), so 1 Hz of update rate should be enough. If IRIS will need finer time granuality to frequently control the mechanisms, IRIS software should perform interpolation internally using the new event and the past events. If we really need to increase the update rate of this event in the future for better accuracy, there are two options: increase the update rate of this telemetry, or split "orientation" attribute into a separate event because the rotation velocity caused by the atmospheric correction is likely to be much slower than the velocity caused by the orientation change.*
+ *Discussion: This event is published at 2 Hz. If IRIS needs finer time granuality to frequently control the mechanisms in their control loops, IRIS software should perform interpolation internally using the latest event and/or the past events.*
- *Discussion: It is expected that this event is published two ticks (= 2 seconds in case the update rate of this event is 1 Hz) before the time indicated by "timestamp" attribute. This event shall be published, at least, one tick before the time indicated by "timestamp" attribute to allow IRIS to perform interpolation of "orientation". Considering the calculation time to perform interpolation and margin for communication & software delay, it should be one more tick before, hence two ticks in total.*
+ *Discussion: This event should be published > 1 second (TBC) before the time indicated by the "timestamp" attribute.*
"""
- maxRate = 1
+ maxRate = 2
+ minRate = 2
archive = true
attributes = [
{
@@ -34,9 +35,7 @@ publish {
name = orientation
description = """Orientation of the axis of dispersion due to atmospheric refraction defined in the XY plane of the FCRSIRIS-ROT.
- *Discussion: For conventional field rotation tracking observations, this value will be the same as the parallactic angle (or its negative). This is because the ADC is inside the instrument, and the angle of the instrument rotator is a combination of parallactic angle and pupil rotation. The ADC just needs to negate the parallactic angle caused by the insturment rotator. This value may include an offset if the observer has requested a particular zenith direction in the image.*
-
- *Discussion: If we need to support pupil rotation tracking mode (= fixed pupil mode, or ADI mode) in the future, this value should be the zenith direction in the image specified by the observer. This should be fixed during the observation because the instrument rotator is supposed to compensate the pupil rotation.*
+ *Discussion: For conventional field rotation tracking observations, this value is the same as the parallactic angle (or its negative) plus the position angle. However, this attribute can be a different value in other tracking modes. See "2.1.2 IRIS Celestial Tracking Modes" in TMT.SEN.ICD.13.002 for more details.*
"""
type = double
minimum = -180
@@ -364,6 +363,8 @@ publish {
The TCS publishes the timestamped current IRIS instrument rotator angle.
*Discussion: During LGS, NGS or seeing limited operations, the IRIS instrument rotator will rotate to derotate the field, which rotates as the telescope tracks a target across the sky (a combination of parallactic angle and pupil rotation). In addition, the rotator angle may include an offset if the observer has requested a particular PA.*
+
+ *Discussion: In the M1 tracking mode, this angle excludes the parallactic angle to tranck only M1 so that there is a fixed M1 pupil pattern in the IRIS focal plane. In the DM0 tracking mode, this angle only includes the position angle specified by the observer so that the rotator is static. See "2.1.2 IRIS Celestial Tracking Modes" in TMT.SEN.ICD.13.002 for more details.*
"""
minRate = 25
maxRate = 100
@@ -399,20 +400,18 @@ publish {
{
name = pupilRotation
description = """
- The TCS publishes the timestamped future pupil rotation angle.
+ The TCS publishes the timestamped future pupil rotation angle in FCRSIRIS-ROT.
- **TBD**: This event shall be issued 50 ms before the time indicated by "timestamp" attribute considering the time for the event propagation over the network and additional latency caused by IRIS software and control system.
+ *Discussion: This event should be published > 1 second (TBC) before the time indicated by the "timestamp" attribute.*
"""
maxRate = 20
archive = true
attributes = [
{
name = pupilRotation
- description = """Future IRIS pupil rotation angle in the X,Y plane of the FCRSIRIS-ROT that will be valid at the time indicated by "timestamp" attribute.
-
- *Discussion: For conventional field rotation tracking observations, this value will be the same as the parallactic angle (or its negative). This is because the cold stop is inside the instrument, and the angle of the instrument rotator is a combination of parallactic angle and pupil rotation. The cold stop just needs to negate the parallactic angle caused by the insturment rotator. This value may include an offset if the observer has requested a particular zenith direction in the image.*
+ description = """Future IRIS pupil rotation angle in the X,Y plane of the FCRSIRIS-ROT that will be valid at the time indicated bythe "timestamp" attribute.
- *Discussion: If we need to support pupil rotation tracking mode (= fixed pupil mode, or ADI mode) in the future, this value should be the zenith direction in the image specified by the observer. This should be fixed during the observation because the instrument rotator is supposed to compensate the pupil rotation.*
+ *Discussion: For conventional field rotation tracking observations, this value will be the same as the parallactic angle (or its negative) plus the position angle. However, this attribute can be a different value in other tracking modes. See "2.1.2 IRIS Celestial Tracking Modes" in TMT.SEN.ICD.13.002 for more details.*
"""
type = double
minimum = -180