Revision as of 06:41, 19 October 2009

PF Users Navigation:

• PCB; Lifefix_users; Pf_users; benchtest_users; CalibFix Users; Hardware; Assembly ; iochan ;
• PCA : PCB ; AKENC SC/MC 218  ; FET 217 ; IMU219 ; Swifi ; Rev200_mods  ; PCA Inventory
• Special Commands to the Ankle (PFCMD) : State Controller Commands; Motor Controller Commands; Python Examples; IMU Commands ;PFCMD_PY; Pf_calb_table_py; Virtual spring test; PF EEPROM
• DOC: Pf_users; Powerfoot Keyboard User Interface; Steps for Manual Tuning; "Dashboard" Program For Assisting with Tuning
• NEW (CEB) WIKI
• Torque Feedback Controller Guide

Edit

• LF: Lifefix_users; Ltf_science; lifefix_log ; Loadwasher ; Lifefix;
• TESTS: LF_TEST100; LF_TEST_101; LF_TEST102; LF_TEST_103; LF_TEST_104

safe startup

• make sure copley is 'phased' and that the zero point is set

• power the AK, wait for the high pitch 'beep beep'
  • if a there is a low humm, then likely the commutation failed. The AK will retry. Hand adjust the back drive screw between attempts to help.

• ideally be running the matlab plotter on CB's machine (This has to be running for now. TBF ASAP)

• log on to the linux box as root
• from /root cd to pfcmd (eg cd pfcmd) - this is a soft link to the correct directory
  • /home/cbarn/src/pfcmd_python (local copy of the program)
  • /mnt/ceb_svn_share/pfcmd_python (remote mounted copy of program)

• Test that the app will run (-n 0 = do zero passes)
  • ./run_lifefix_v2.py -i sw8 -c 0 -p 2 -T 100 -F 1 -E 1 -n 0 -S
  • ideally there will be no error messages and some matlab graphic output will appear
  • if the swifi isn't working an error message will be output and the process stopped.

• Run the gentle cycle (-F 1) for 1 pass (-n 1) with 100N of AK output (-T 100):
  • ./run_lifefix_v2.py -i sw8 -c 0 -p 2 -T 100 -F 1 -E 1 -n 1 -S
  • there should be no python errors, however there will be an envelope error

• Run the full cycle (-F 0) for 1 pass (-n 1) with 100N of AK output (-T 100):
  • ./run_lifefix_v2.py -i sw8 -c 0 -p 2 -T 100 -F 0 -E 1 -n 1 -S
  • there should be no python errors, however there will be an envelope error

• Run the full cycle (-F 0) for 1 pass (-n 1) with 2000N full thrust (-T 2000):
  • ./run_lifefix_v2.py -i sw8 -c 0 -p 2 -T 2000 -F 0 -E 1 -n 1 -S
  • there should be no python errors, and no envelope errors.

• proceed with a short run (100 passes) to verify operation
  • ./run_lifefix_v2.py -i sw8 -c 0 -p 2 -T 2000 -F 0 -E 1 -n 100 -S

stopping

• ESC terminates the program in the correct manner.
• Do not use Ctrl-C

issues

• SWIFI problems will abort the program. Need to do a re-connect.
• sometimes the SWIFI outputs garbage upon power up. SWIFI needs to be power cycled to correct.
  • the lifefix program is now testing for valid SWIFI output.

changes

0) The command line options are somewhat different.

1) The move_ecat program stays running for the entire run. python and move_ecat communicate via signals.  we're doing a 88k cycle run - i hope there aren't memory issues!

2) We're no longer using motor position as a trigger to start the thrust playback.  Instead, after python sees the ACK from move_ecat, it forces the playback trigger.

3) Belt testing is passive.   The range of motion turning a PVT playback is check against a range.

4) If the motor temperature exceeds 35C, then the cycle slows based on temperature.   

5) The relationship between the start of PVT playback and the thrust table payback has changed from previous experiments - but should be more constant.

6) The program monitors the SWIFI and will extend the cycle to read more SWIFI data.  IF the SWIFI stops transmission, the program should stop.  I'll soon add re-connect.  If the app stopps due to SWIFI connection, then just restart the application.

known issues

Pf_insects