Came in and switched the LED control back to Andor camera.
Hrand updated MSIM
Processed new centers and link lengths from last night with PID7 working and new ontimes for PID7 and PID5 phi fwd.
D:\PfsTests\07_15_14_17_27_35_centers\Images\fullmtrlgy_071514_1.xml
NOTE: Had to eliminate some pairs of center images and their reference images because the reference images didnt contain any good centroids. Moved these to the bad subdirectory. Theory is that the camera was not allowed enough time to start exposure before the LED was strobed. So changed the the wait after start exposure command from 2 seconds to 3 seconds in 2014_07_15_Centers01
Loaded the new XML into MSIM and saved it:
cmd_load_Xml_file_to_Database fullmtrlgy_071514_1
cmd_save_Database
Created a new motormapping MSIM script with updated ontimes: 2014_07_16_msimMotMap
After running the script, noticed that PID5 is not getting around on Phi past the known sticky spot which the streaking ontime calibrations were able to overcome. This reaffirms that small moves successively cannot sometimes push past sticky points which the streaks could at the same ontime.
Here's forward phi getting stuck at around 68deg
SNIPPET FROM: D:\PfsTests\07_16_14_09_14_52_msimMaps\Log\PhiFwMap_mId_1_pId_5.txt
X-Coord Y-Coord Theta Phi
1022.344620, 928.863039, 1.066594, 49.908787
1019.806583, 929.137693, 1.554324, 54.781756
1017.245208, 929.145309, 2.086638, 59.613513
1014.640122, 928.938527, 2.530994, 64.564504
1013.318199, 928.816508, 2.574786, 67.183715
1012.914089, 928.772120, 2.576107, 67.993175
1012.886978, 928.766897, 2.580762, 68.045056
1012.851173, 928.755463, 2.596969, 68.108146
1012.848043, 928.762723, 2.579885, 68.123665
1012.848839, 928.761405, 2.583048, 68.120358
1012.843171, 928.775024, 2.551059, 68.149039
1012.849163, 928.757954, 2.590859, 68.115483
Here's reverse phi getting stuck at around 71deg
SNIPPET FROM: D:\PfsTests\07_16_14_09_14_52_msimMaps\Log\PhiRvMap_mId_1_pId_5.txt
X-Coord Y-Coord Theta Phi
994.295845, 916.376000, 6.010464, 110.272672
995.977243, 918.653630, 5.575428, 105.203033
997.848648, 920.737854, 5.154254, 100.140876
999.917699, 922.570214, 4.854682, 94.974550
1002.167968, 924.258719, 4.461959, 89.785720
1004.315528, 925.569431, 4.127701, 85.099620
1006.665386, 926.719586, 3.795188, 80.184663
1009.028498, 927.608765, 3.497133, 75.396232
1010.750009, 928.172132, 3.149613, 72.054069
1011.160449, 928.270224, 3.106738, 71.243865
1011.191703, 928.296372, 3.062277, 71.204062
1011.196193, 928.286530, 3.085453, 71.182702
1011.211267, 928.286047, 3.092447, 71.148468
1011.203714, 928.285037, 3.091668, 71.164180
1011.213820, 928.290894, 3.082909, 71.148343
Cranking up the ontime for PID5 phi fwd and reverse to 0.08 from 0.07 in 2014_07_16_msimMotMap_pid5retest
Updated the ontimes for PID 5 phi stage in the XML and saved it again as:
D:\PfsTests\07_16_14_11_52_47_msimMaps\Log\fullmtrlgy_071514_2_5phiOT.xml
Results of PID 5 restest on phi are in \07_16_14_11_52_47_msimMaps and the theta logs were moved to this folder as well from 07_16_14_09_14_52_msimMaps so that a full set of maps can be made from the matlab script which expects them all in the same directory.
While running processMsimMaps, PID9 phi forward had issue with the first line in the log being greater than the next and script couldn't handle it.
PhiFwMap_mId_1_pId_9.txt
295.792488, 195.304886, 187.768081, 2.424764
295.843566, 193.636088, 10.851856, 1.171929
295.951235, 191.660117, 3.199196, 6.183946
Renamed the log to BAD_PhiFwMap_mId_1_pId_9.txt and made a copy with original name from which the first line was removed.
Ran process maps successfully and saved resulting XML as 071614_setC_allmaps.xml in D:\PfsTests\07_16_14_11_52_47_msimMaps\Log
The new motor maps on PID5 look better and cover the expected range of motion on phi versus before where it was stalling.
PID7 INVESTIGATION
PID7 seems to now be stalling at ~152deg on theta in reverse. This is seen on the motor map plot and also confirmed in the log where an excess of 50 entries are stalled til the end at that value.
SNIPPET OF thetaRvMap_mId_1_pId_7.txt in D:\PfsTests\07_16_14_11_52_47_msimMaps\Log
X-Coord Y-Coord Theta Phi
663.051912, 542.148634, 177.732329, 30.162379
662.604466, 542.525138, 175.297696, 30.148315
662.297766, 542.806109, 173.567377, 30.132969
661.905363, 543.108958, 171.468976, 30.245900
661.477423, 543.575053, 168.851295, 30.204839
660.956158, 544.106784, 165.723539, 30.295500
660.464118, 544.720824, 162.449350, 30.312446
659.996008, 545.290064, 159.355810, 30.440995
659.675817, 545.921876, 156.486920, 30.297543
659.401817, 546.352925, 154.347650, 30.356768
659.235809, 546.659857, 152.897755, 30.366253
659.137323, 546.789450, 152.209617, 30.433682
659.144624, 546.806661, 152.171393, 30.401675
659.144894, 546.800172, 152.193701, 30.407620
659.150648, 546.839679, 152.076955, 30.356526
659.155297, 546.813431, 152.176657, 30.373297
659.163779, 546.823974, 152.163592, 30.345587
659.151275, 546.826299, 152.123260, 30.368592
659.155861, 546.823625, 152.144125, 30.361972
Looked back in motor map script and confirmed that ontime was correctly set at 0.2ms for PID7 theta rev
cmd_moveMotor_DurationInterval 1,7,1,-60,0.2,2.5
Took the ontime streak testing script for theta rev and cut out the lower ontimes so that it goes from 0.16 to 0.3. Saved as 20140716_pid7thetaTesting
Replaced lens filters for streaks and ran script
Started a new centers script for today based on yesterdays. 2014_07_16_pid7thetaTesting2
Changed center iterations to 1 just to be able to run this and see if the small step ranges of motion are not making it as far as the streaks.
Result looks like it still makes it around all the way. Note that each move was 400 steps. The motor map script does incriments of 60steps for each move and that takes too long to capture in one image so we'll have to just try again now and look at the log.
Creating a new motor map script just to look at 7: 2014_07_16_pid7thetaTesting3
Results are best summarized by the image below.
- On the upper left is the streak generated by cmd_moveMotor_DurationInterval 1,7,1,-7000,0.2,2.5
- On the lower left is a streak with periodic pauses caused by running 20 iterations of cmd_moveMotor_DurationInterval 1,7,1,-400,0.2,2.5 with 1.7sec between each move
- On the right is the tail of a log for motor mapping on PID7 theta reverse using command cmd_moveMotor_DurationInterval 1,7,1,-60,0.2,2.5 with about 4sec between each move depending on camera response time
For the streak of 7000 steps and moves of 400 steps, the theta goes all the way around. But when doing small steps of 60steps, it stalls clearly at ~150deg.
To try and see if the streak could be stalled by increasing offtime, the script 20140716_pid7thetaTesting4 was made to just do a reverse streak at .2ms ontime and vary the offtime to see when it might stall out at 150deg and prove that steps could be discrete and independent moves from one another just like the 60step moves are when we do the motor maps.
The following offtimes were tried with the .2ms ontime but still the theta stage made it all the way around:
ontime |
offtime |
steps |
result |
directory |
0.2 |
2.3 |
-7000 |
full ROM |
07_16_14_14_13_45_onTimeTuning |
0.2 |
3.8 |
-7000 |
full ROM |
07_16_14_15_22_44_onTimeTuning |
0.2 |
7.8 |
-7000 |
full ROM |
07_16_14_15_30_24_onTimeTuning |
0.2 |
14.8 |
-7000 |
full ROM |
07_16_14_15_35_40_onTimeTuning |
Ran the exact same motormap script as before to get stuck again, 2014_07_16_pid7thetaTesting3. Succeeded in getting stuck at 151deg just like before
Now trying to manually get past by successively doubling the number of steps we are commanding.
rev steps commanded |
theta |
-60 |
151.683927 |
-120 |
151.717830 |
-240 |
151.726073 |
-480 |
151.711762 |
-480 |
151.744123 |
-1000 |
151.739643 |
60 |
164.541673 |
reversed direction to back up |
-120 |
154.102809 |
|
-120 |
148.349958 |
got passed! |
-120 |
147.748588 |
|
-120 |
143.744394 |
|
-120 |
135.842827 |
|
120 |
162.382413 |
backing up again |
-60 |
155.459455 |
|
-60 |
148.915499 |
|
-60 |
142.829186 |
|
-60 |
136.959795 |
|
|
|
|