changed/fixed behaviour of stop commands in PPMS
- make sure new target created by the stop command is between old and new target - fixed bad behaviour of level reading Change-Id: I484c0902c694c0edb81e2d9238985c05f92e04f4 Reviewed-on: https://forge.frm2.tum.de/review/c/sine2020/secop/playground/+/22100 Tested-by: JenkinsCodeReview <bjoern_pedersen@frm2.tum.de> Reviewed-by: Markus Zolliker <markus.zolliker@psi.ch>
This commit is contained in:
parent
d442c5b6bf
commit
f0a3306f9c
@ -81,7 +81,7 @@ def clamp(_min, value, _max):
|
||||
"""return the median of 3 values,
|
||||
|
||||
i.e. value if min <= value <= max, else min or max depending on which side
|
||||
value lies outside the [min..max] interval
|
||||
value lies outside the [min..max] interval. This works even when min > max!
|
||||
"""
|
||||
# return median, i.e. clamp the the value between min and max
|
||||
return sorted([_min, value, _max])[1]
|
||||
|
||||
@ -39,6 +39,7 @@ from secop.modules import Module, Readable, Drivable, Parameter, Override,\
|
||||
from secop.datatypes import EnumType, FloatRange, IntRange, StringType,\
|
||||
BoolType, StatusType
|
||||
from secop.lib.enum import Enum
|
||||
from secop.lib import clamp
|
||||
from secop.errors import HardwareError
|
||||
from secop.poller import Poller
|
||||
import secop.iohandler
|
||||
@ -54,7 +55,6 @@ except ImportError:
|
||||
class IOHandler(secop.iohandler.IOHandler):
|
||||
CMDARGS = ['no']
|
||||
CMDSEPARATOR = None # no command chaining
|
||||
READ_BEFORE_WRITE = False
|
||||
|
||||
def __init__(self, name, querycmd, replyfmt):
|
||||
changecmd = querycmd.split('?')[0] + ' '
|
||||
@ -127,11 +127,12 @@ class Main(Communicator):
|
||||
class PpmsMixin(HasIodev, Module):
|
||||
properties = {
|
||||
'iodev': Attached(),
|
||||
'general_stop': Property('respect general stop', datatype=BoolType(),
|
||||
export=True, default=True)
|
||||
}
|
||||
|
||||
pollerClass = Poller
|
||||
enabled = True # default, if no parameter enable is defined
|
||||
_last_target_change = 0 # used by several modules
|
||||
_last_settings = None # used by several modules
|
||||
slow_pollfactor = 1
|
||||
|
||||
@ -303,11 +304,9 @@ class Level(PpmsMixin, Readable):
|
||||
"""
|
||||
|
||||
def analyze_level(self, level, status):
|
||||
if status:
|
||||
self.status = [self.Status.IDLE, '']
|
||||
else:
|
||||
self.status = [self.Status.ERROR, 'old reading']
|
||||
return dict(value=level)
|
||||
# ignore 'old reading' state of the flag, as this happens only for a short time
|
||||
# during measuring
|
||||
return dict(value=level, status=[self.Status.IDLE, ''])
|
||||
|
||||
|
||||
class Chamber(PpmsMixin, Drivable):
|
||||
@ -416,10 +415,6 @@ class Temp(PpmsMixin, Drivable):
|
||||
Parameter('drive timeout, in addition to ramp time', readonly=False,
|
||||
datatype=FloatRange(0, unit='sec'), default=3600),
|
||||
}
|
||||
properties = {
|
||||
'general_stop': Property('respect general stop', datatype=BoolType(),
|
||||
export=True, default=True)
|
||||
}
|
||||
# pylint: disable=invalid-name
|
||||
TempStatus = Enum(
|
||||
'TempStatus',
|
||||
@ -447,12 +442,10 @@ class Temp(PpmsMixin, Drivable):
|
||||
|
||||
channel = 'temp'
|
||||
_stopped = False
|
||||
_expected_target = 0
|
||||
_expected_target_time = 0
|
||||
_last_change = 0 # 0 means no target change is pending
|
||||
|
||||
def earlyInit(self):
|
||||
self.setProperty('general_stop', False)
|
||||
super().earlyInit()
|
||||
_last_target = None # last reached target
|
||||
general_stop = False
|
||||
|
||||
def update_value_status(self, value, packed_status):
|
||||
"""update value and status"""
|
||||
@ -462,13 +455,6 @@ class Temp(PpmsMixin, Drivable):
|
||||
self.value = value
|
||||
status = self.STATUS_MAP[packed_status & 0xf]
|
||||
now = time.time()
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
self._stopped = False
|
||||
else:
|
||||
self.status = [self.Status.IDLE, 'stopped(%s)' % status[1]]
|
||||
return
|
||||
if self._last_change: # there was a change, which is not yet confirmed by hw
|
||||
if now > self._last_change + 5:
|
||||
self._last_change = 0 # give up waiting for busy
|
||||
@ -477,13 +463,23 @@ class Temp(PpmsMixin, Drivable):
|
||||
self._last_change = 0
|
||||
else:
|
||||
status = [self.Status.BUSY, 'changed target']
|
||||
if self._expected_target:
|
||||
if abs(self.value - self.target) < self.target * 0.01:
|
||||
self._last_target = self.target
|
||||
elif self._last_target is None:
|
||||
self._last_target = self.value
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
status = [status[0], 'stopped']
|
||||
else:
|
||||
status = [status[0], 'stopping (%s)' % status[1]]
|
||||
if self._expected_target_time:
|
||||
# handle timeout
|
||||
if self.isDriving(status):
|
||||
if now > self._expected_target + self.timeout:
|
||||
if now > self._expected_target_time + self.timeout:
|
||||
status = [self.Status.WARN, 'timeout while %s' % status[1]]
|
||||
else:
|
||||
self._expected_target = 0
|
||||
self._expected_target_time = 0
|
||||
self.status = status
|
||||
|
||||
def analyze_temp(self, target, ramp, approachmode):
|
||||
@ -500,7 +496,7 @@ class Temp(PpmsMixin, Drivable):
|
||||
|
||||
def write_target(self, target):
|
||||
self._stopped = False
|
||||
if abs(self.target - self.value) < 2e-5 and target == self.target:
|
||||
if abs(self.target - self.value) <= 2e-5 * target and target == self.target:
|
||||
return None
|
||||
self._status_before_change = self.status
|
||||
self.status = [self.Status.BUSY, 'changed target']
|
||||
@ -518,19 +514,18 @@ class Temp(PpmsMixin, Drivable):
|
||||
return None # change_temp will not be called, as this would trigger an unnecessary T change
|
||||
|
||||
def calc_expected(self, target, ramp):
|
||||
self._expected_target = time.time() + abs(target - self.value) * 60.0 / max(0.1, ramp)
|
||||
self._expected_target_time = time.time() + abs(target - self.value) * 60.0 / max(0.1, ramp)
|
||||
|
||||
def do_stop(self):
|
||||
if self.isDriving():
|
||||
if not self.isDriving():
|
||||
return
|
||||
if self.status[0] == self.Status.STABILIZING:
|
||||
# we are already near target
|
||||
newtarget = self.target
|
||||
else:
|
||||
newtarget = self.value
|
||||
self.log.info('stop at %s K', newtarget)
|
||||
self.write_target(newtarget)
|
||||
self.status = [self.Status.IDLE, 'stopped']
|
||||
if self.status[0] != self.Status.STABILIZING:
|
||||
# we are not near target
|
||||
newtarget = clamp(self._last_target, self.value, self.target)
|
||||
if newtarget != self.target:
|
||||
self.log.debug('stop at %s K', newtarget)
|
||||
self.write_target(newtarget)
|
||||
self.status = [self.status[0], 'stopping (%s)' % self.status[1]]
|
||||
self._stopped = True
|
||||
|
||||
|
||||
@ -583,7 +578,7 @@ class Field(PpmsMixin, Drivable):
|
||||
|
||||
channel = 'field'
|
||||
_stopped = False
|
||||
_last_target = 0
|
||||
_last_target = None # last reached target
|
||||
_last_change = 0 # means no target change is pending
|
||||
|
||||
def update_value_status(self, value, packed_status):
|
||||
@ -595,14 +590,7 @@ class Field(PpmsMixin, Drivable):
|
||||
status_code = (packed_status >> 4) & 0xf
|
||||
status = self.STATUS_MAP[status_code]
|
||||
now = time.time()
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
self._stopped = False
|
||||
else:
|
||||
self.status = [status[0], 'stopped (%s)' % status[1]]
|
||||
return
|
||||
elif self._last_change: # there was a change, which is not yet confirmed by hw
|
||||
if self._last_change: # there was a change, which is not yet confirmed by hw
|
||||
if status_code == 1: # persistent mode
|
||||
# leads are ramping (ppms has no extra status code for this!)
|
||||
if now < self._last_change + 30:
|
||||
@ -616,6 +604,16 @@ class Field(PpmsMixin, Drivable):
|
||||
self.log.debug('time needed to change to busy: %.3g', now - self._last_change)
|
||||
else:
|
||||
status = [self.Status.BUSY, 'changed target']
|
||||
if abs(self.target - self.value) <= 1e-4:
|
||||
self._last_target = self.target
|
||||
elif self._last_target is None:
|
||||
self._last_target = self.value
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
status = [status[0], 'stopped']
|
||||
else:
|
||||
status = [status[0], 'stopping (%s)' % status[1]]
|
||||
self.status = status
|
||||
|
||||
def analyze_field(self, target, ramp, approachmode, persistentmode):
|
||||
@ -624,34 +622,49 @@ class Field(PpmsMixin, Drivable):
|
||||
# not always sent to the hardware
|
||||
return {}
|
||||
self._last_settings = target, ramp, approachmode, persistentmode
|
||||
return dict(target=target * 1e-4, ramp=ramp * 6e-3, approachmode=approachmode,
|
||||
return dict(target=round(target * 1e-4, 7), ramp=ramp * 6e-3, approachmode=approachmode,
|
||||
persistentmode=persistentmode)
|
||||
|
||||
def change_field(self, change):
|
||||
if change.doesInclude('target', 'persistentmode'):
|
||||
if change.doesInclude('target'):
|
||||
self._last_target = self.target # save for stop command
|
||||
self._stopped = False
|
||||
self._last_change = time.time()
|
||||
self._status_before_change = list(self.status)
|
||||
else:
|
||||
# changed ramp or approachmode
|
||||
if not self.isDriving():
|
||||
return None # nothing to be written, as this would trigger a ramp up of leads current
|
||||
return change.target * 1e+4, change.ramp / 6e-3, change.approachmode, change.persistentmode
|
||||
|
||||
def write_target(self, target):
|
||||
if abs(self.target - self.value) <= 2e-5 and target == self.target:
|
||||
return None # avoid ramping leads
|
||||
self._status_before_change = list(self.status)
|
||||
self._stopped = False
|
||||
self._last_change = time.time()
|
||||
self.status = [self.Status.BUSY, 'changed target']
|
||||
return target
|
||||
|
||||
def write_persistentmode(self, mode):
|
||||
if abs(self.target - self.value) <= 2e-5 and mode == self.persistentmode:
|
||||
return None # avoid ramping leads
|
||||
self._last_change = time.time()
|
||||
self._status_before_change = list(self.status)
|
||||
self._stopped = False
|
||||
self.status = [self.Status.BUSY, 'changed persistent mode']
|
||||
return mode
|
||||
|
||||
def write_ramp(self, value):
|
||||
if self.isDriving():
|
||||
return value
|
||||
return None # change_field will not be called, as this would trigger a ramp up of leads current
|
||||
|
||||
def write_approachmode(self, value):
|
||||
if self.isDriving():
|
||||
return value
|
||||
return None # change_temp will not be called, as this would trigger a ramp up of leads current
|
||||
|
||||
def do_stop(self):
|
||||
if not self.isDriving():
|
||||
return
|
||||
self.status = [self.Status.IDLE, '_stopped']
|
||||
newtarget = clamp(self._last_target, self.value, self.target)
|
||||
if newtarget != self.target:
|
||||
self.log.debug('stop at %s T', newtarget)
|
||||
self.write_target(newtarget)
|
||||
self.status = [self.status[0], 'stopping (%s)' % self.status[1]]
|
||||
self._stopped = True
|
||||
if abs(self.value - self.target) > 1e-4:
|
||||
# ramping is not yet at end
|
||||
if abs(self.value - self._last_target) < 1e-4:
|
||||
# ramping has not started yet, use more precise last target instead of current value
|
||||
self.target = self.put_settings(self._last_target, 'target')
|
||||
else:
|
||||
self.target = self.put_settings(self.value, 'target')
|
||||
|
||||
|
||||
class Position(PpmsMixin, Drivable):
|
||||
@ -684,7 +697,7 @@ class Position(PpmsMixin, Drivable):
|
||||
|
||||
channel = 'position'
|
||||
_stopped = False
|
||||
_last_target = 0
|
||||
_last_target = None # last reached target
|
||||
_last_change = 0 # means no target change is pending
|
||||
|
||||
def update_value_status(self, value, packed_status):
|
||||
@ -697,12 +710,6 @@ class Position(PpmsMixin, Drivable):
|
||||
return
|
||||
self.value = value
|
||||
status = self.STATUS_MAP[(packed_status >> 12) & 0xf]
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
self._stopped = False
|
||||
else:
|
||||
status = [self.Status.IDLE, 'stopped(%s)' % status[1]]
|
||||
if self._last_change: # there was a change, which is not yet confirmed by hw
|
||||
now = time.time()
|
||||
if now > self._last_change + 5:
|
||||
@ -712,6 +719,16 @@ class Position(PpmsMixin, Drivable):
|
||||
self.log.debug('time needed to change to busy: %.3g', now - self._last_change)
|
||||
else:
|
||||
status = [self.Status.BUSY, 'changed target']
|
||||
if abs(self.value - self.target) < 0.1:
|
||||
self._last_target = self.target
|
||||
elif self._last_target is None:
|
||||
self._last_target = self.value
|
||||
if self._stopped:
|
||||
# combine 'stopped' with current status text
|
||||
if status[0] == self.Status.IDLE:
|
||||
status = [status[0], 'stopped']
|
||||
else:
|
||||
status = [status[0], 'stopping (%s)' % status[1]]
|
||||
self.status = status
|
||||
|
||||
def analyze_move(self, target, mode, speed):
|
||||
@ -727,7 +744,6 @@ class Position(PpmsMixin, Drivable):
|
||||
return change.target, 0, speed
|
||||
|
||||
def write_target(self, target):
|
||||
self._last_target = self.target # save for stop command
|
||||
self._stopped = False
|
||||
self._last_change = 0
|
||||
self._status_before_change = self.status
|
||||
@ -736,17 +752,14 @@ class Position(PpmsMixin, Drivable):
|
||||
def write_speed(self, value):
|
||||
if self.isDriving():
|
||||
return value
|
||||
return None # change_move not called: as this would trigger an unnecessary move
|
||||
return None # change_move not called as this would trigger an unnecessary move
|
||||
|
||||
def do_stop(self):
|
||||
if not self.isDriving():
|
||||
return
|
||||
self.status = [self.Status.BUSY, '_stopped']
|
||||
newtarget = clamp(self._last_target, self.value, self.target)
|
||||
if newtarget != self.target:
|
||||
self.log.debug('stop at %s T', newtarget)
|
||||
self.write_target(newtarget)
|
||||
self.status = [self.status[0], 'stopping (%s)' % self.status[1]]
|
||||
self._stopped = True
|
||||
if abs(self.value - self.target) > 1e-2:
|
||||
# moving is not yet at end
|
||||
if abs(self.value - self._last_target) < 1e-2:
|
||||
# moving has not started yet, use more precise last target instead of current value
|
||||
self.target = self.write_target(self._last_target)
|
||||
else:
|
||||
self.target = self.write_target(self.value)
|
||||
|
||||
@ -54,7 +54,7 @@ class PpmsSim:
|
||||
self.status = NamedList('t mf ch pos', 1, 1, 1, 1)
|
||||
self.st = 0x1111
|
||||
self.t = 200
|
||||
self.temp = NamedList('target ramp amode', 295., 1, 0, fast=self.t, delay=10)
|
||||
self.temp = NamedList('target ramp amode', 200., 1, 0, fast=self.t, delay=10)
|
||||
self.mf = 100
|
||||
self.field = NamedList('target ramp amode pmode', 0, 50, 0, 0)
|
||||
self.pos = 0
|
||||
@ -75,6 +75,7 @@ class PpmsSim:
|
||||
self.start = self.time
|
||||
self.mf_start = 0
|
||||
self.ch_start = 0
|
||||
self.t_start = 0
|
||||
self.changed = set()
|
||||
|
||||
def progress(self):
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user