Tags:
tag this topic
create new tag
view all tags
Verified positioners in theta reverse hardstop by hand. *Moved phi to intermediate position* Ran several images without motion in between to measure joint angle stability in MSIM. Looked repeatable. D:\PfsTests\15_09_09_14_17_55_msimCenters\Log Case Study on PID 9: <verbatim> 1639.010909, 1494.950314, 9.512194, 83.546917 1639.015469, 1494.949618, 9.507678, 83.541114 1639.026299, 1494.966767, 9.462858, 83.582771 1638.989696, 1494.927909, 9.579703, 83.498373 1638.978009, 1494.924961, 9.599850, 83.499336 1639.018310, 1494.927580, 9.544032, 83.473836 1639.007960, 1494.922499, 9.566369, 83.467384 1638.971294, 1494.946626, 9.569028, 83.568717 1638.999337, 1494.947836, 9.531346, 83.549141 1638.984517, 1494.939704, 9.564862, 83.537397 </verbatim> Commanded phi all the way out and repeated pictures. Found theta to stay at ~10deg <verbatim> 1612.001134, 1495.254202, 10.307388, 149.973990 1611.968848, 1495.261306, 10.204726, 150.227769 1612.017472, 1495.252043, 10.353620, 149.854833 1612.040316, 1495.239580, 10.452690, 149.633752 1612.016971, 1495.225077, 10.452127, 149.699773 1612.055621, 1495.222419, 10.550961, 149.435291 1612.038438, 1495.207727, 10.565453, 149.460212 1611.999091, 1495.222729, 10.419239, 149.801907 </verbatim> Commanded phi all the way in and repeated pictures. Now theta is totally different... <verbatim> 1639.735283, 1455.454917, 147.170602, 24.861867 1639.704160, 1455.516264, 146.826288, 24.822980 1639.717375, 1455.515868, 146.875580, 24.799836 1639.642199, 1455.472841, 146.766662, 25.001201 1639.711187, 1455.516682, 146.850111, 24.809704 1639.700574, 1455.509256, 146.839862, 24.840208 1639.696118, 1455.518971, 146.786981, 24.833281 1639.689207, 1455.512950, 146.784832, 24.854969 1639.710706, 1455.510546, 146.871615, 24.820013 1639.688442, 1455.509084, 146.796698, 24.862288 </verbatim> Commanded phi out 300 steps and repeated pictures. Theta back to ~10deg. Obviously error in inverse kinematics. ---++Overheating theta stages Checking general health of Theta stage on PID 9 visually Script does the following in a loop 1. Run 4000 steps fwd 1. Wait 10 seconds 1. Run 4000 steps rev 1. Wait 5 seconds In first loop iteration, Observed theta stage moving fwd correctly and back to hardstop with some overdrive. But all subsequent fwd moves fail to get out of hardstop. The motor can be heard, but no motion visible. Several full loops observed with this behavior in succession. Reran the script and the first fwd motion is successful, but the next again fails as do the following fwd motions. Reverse can't be talked to since the motor never makes it off the rev hardstop after the first loop iteration. Thinking this could be an issue with motor overheating, changed wait after reverse motions to 30sec. The fwd motions now succeed everytime. Repeated the overheat experiment with PID 31 and same results. After moving back to hardstop 4000 steps with slight overdrive into hardstop, after waiting 5 sec, the motor does not move when given a fwd command of 4000 steps. After increasing the wait between moves to 30sec, the fwd motion is successful. *Confirmed the same behavior on older positioners 15, 21 and 37 using the same script.* ---++Inverse kinematics debug continued... * Adjusted fiber arm on PID9 coupler to provide a certain positive offset when in home position. * Ran the following commands and watched by eye what was going on: <verbatim> cmd_moveMotor_Steps_all 2,2000,1 { Move to home position and out slightly on phi } cmd_moveMotor_Steps_all 1,6000,1 cmd_moveMotor_Steps_all 1,-12000,1 { Take reference image} cmd_getImageStart_NoWait cmd_getImageDone_Wait cmd_findCentroids cmd_LogCentroids cmd_setCurrentPos_all cmd_getCentroidsArmInverseKin cmd_clearOpenCV_Images </verbatim> * The first move on phi goes as planned * *The second move of 6000 fwd on theta begins, but before I notice an end to the motor noise the next command kicks off and then completed immediately and proceeds to the camera command which I hear the camera click for while the motors are still moving!! This explains why the angles we were measuring here did not correspond to the hardstop, the camera took the image mid-motion.* * confirmed now that all of the positioners are in their theta rev hardstops 1. take images, no motion 1. Move in phi -2000 steps, take image 1. Move in phi -2000 steps, take image 1. 10 iterations of: Move phi 200 steps, take image * PID 9 Results <verbatim> 1649.220964, 1473.403612, 10.006204, 24.517356 1648.831170, 1469.729178, 9.077270, 15.773651 1648.696660, 1479.761686, 10.246687, 39.614717 1644.348054, 1489.283520, 10.342301, 64.512966 1637.307440, 1495.646859, 10.254036, 87.145947 1628.207385, 1498.704456, 10.488848, 109.705394 1618.723872, 1497.899769, 11.133845, 131.328195 1612.428443, 1495.146727, 11.622437, 146.725051 1607.913950, 1491.322876, 13.454309, 157.127254 1603.966002, 1485.883795, 17.938911, 163.915913 1602.232599, 1481.051764, 26.867528, 157.646242 1602.237929, 1481.058267, 26.872470, 157.616623 </verbatim> * Positioner 21 appears to have something very wrong. Ignoring for now * *All positioners seem to go past 180deg and then their phi angles decrease again with corresponding jump in theta angle* -- Main.ChazMorantz - 09 Sep 2015
E
dit
|
A
ttach
|
Watch
|
P
rint version
|
H
istory
: r1
|
B
acklinks
|
V
iew topic
|
Ra
w
edit
|
M
ore topic actions
Topic revision: r1 - 2015-09-09
-
ChazMorantz
Home
Site map
Antarctica web
Gattini web
SPRITE web
Cahill web
Gemini web
Keck web
MOSFIRE web
NGAO web
Main web
Palomar web
GASP web
PTF web
PalmLGS web
RoboAO web
SDC web
SEDM web
SWIFT web
TMAS web
Triplespec web
WaSP web
ZTF web
SNAP web
Sandbox web
Subaru web
PFS web
TMT web
More...
PFS Web
Create New Topic
Index
Search
Changes
Notifications
RSS Feed
Statistics
Preferences
P
P
View
Raw View
Print version
Find backlinks
History
More topic actions
Edit
Raw edit
Attach file or image
Edit topic preference settings
Set new parent
More topic actions
Account
Log In
Register User
E
dit
A
ttach
Copyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki?
Send feedback