Difference between revisions of "Virtual spring test"
From DIDEAS Wiki
m |
m |
||
(4 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
+ | {{pf_top_nav}} | ||
= For SC 751 series code = | = For SC 751 series code = | ||
− | + | The purpose of this page is to discuss how to run the virtual spring test of the robot w/the testjig. Specifically, setting of the testjig and robot. | |
− | |||
− | |||
− | |||
− | |||
+ | ==initial robot configuration== | ||
+ | * disable SM updates to the MC w/command 127 0 | ||
+ | * w/testjig, position robot to the zero position | ||
+ | * (optional) run command 104 w/repeat to confirm that the robot position is near zero (|3rd number| <100), and that the SM state (2nd number) is as commanded (3 initially, 13 later) | ||
+ | * zero (adjust) the K and B for the VS w/command 131 0 0 | ||
+ | * increase the overload limits w/command: 1122 30000 30000 9000 9000 (last number is continuous ~amps*1000) | ||
+ | * command SM to enter the virtual spring state w/command 170 13 (confirm w/cmd 104) | ||
+ | * enable SM updates to the MC w/command 127 5 | ||
− | ==other | + | |
− | *130 can be used to read the VS parameters | + | ==testjig config== |
− | *1120 can be used to read the electronic fuse parameters | + | *range 0 to 64000 (this is inital and final) |
− | *1121 can be used to monitor the electronic fuse | + | **because jig doesn't return to zero, for now run w/ plantarflexion and a range of 0 to 64000 (where zero is near ramp engagement, and is also the zero position for the ankle [where it was boot]) |
+ | * set (de)acceleration, ~10 rads s-2 (or maybe larger? 20 rads s-2) | ||
+ | * RPM 1111 for stiffness test, 4444 for damping test | ||
+ | * set passes to a multiple of 2 (VERY IMPORTANT, otherwise, might stop at 64000) | ||
+ | * 1 second delay between passes (will allow fuse to re-zero) | ||
+ | |||
+ | ==some experiments at 1111 rpm == | ||
+ | *low stiffness w/command : 131 1000 0 (appx 100 nM/rad) | ||
+ | *med stiffness w/command : 131 2000 0 (appx 200 nM/rad) | ||
+ | *high stiffness w/command : 131 4000 0 (appx 400 nM/rad) | ||
+ | |||
+ | ==some experiments at 4444 rpm == | ||
+ | *moderate damping w/command : 131 0 400 (appx 40 nM/rad sec) | ||
+ | *moderate stiffness and damping w/command : 131 4000 400 | ||
+ | |||
+ | |||
+ | ==results== | ||
+ | *K value is approximate 12% small, or appx 20% greater when the effect of K2 is considered | ||
+ | *B value is approximate 20% larger, or +-10% when estimated robot screw friction is removed. | ||
+ | *RE40 to EC30 ratio is appx 3.3 to 1 (based on moment arms of 112mm vs 37 mm, and RE40 ratio of 22/20) | ||
+ | |||
+ | |||
+ | ==other== | ||
+ | *130 can be used to read the VS parameters (the first 2 are relevant in this experiment) | ||
+ | *131 sets K and B : eg 131 K B | ||
+ | *1120 can be used to read the electronic fuse parameters (first two values are maximum sums, the 2nd two values are the decay rates in q15) | ||
+ | *1121 can be used to monitor the electronic fuse (first 3 values are the present sums, 2nd three values are actual current in q15) | ||
+ | ==limits== | ||
+ | *testjig : use 1111 rpm, ~10rad s-2 acceleration, position of 0 to 64000, 2 passes, 1 sec between passes | ||
+ | *set fuse to ~9 amps w/command 1122 30000 30000 9000 9000 | ||
+ | *VS K: don't exceed 400 nm/rad (eg 131 4000 0) - can result in oscillation or overload. | ||
+ | *VS B: don't exceed 40 nm/rad s (eg 131 0 400) - can result in oscillation |
Latest revision as of 19:15, 22 August 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
Contents
For SC 751 series code
The purpose of this page is to discuss how to run the virtual spring test of the robot w/the testjig. Specifically, setting of the testjig and robot.
initial robot configuration
- disable SM updates to the MC w/command 127 0
- w/testjig, position robot to the zero position
- (optional) run command 104 w/repeat to confirm that the robot position is near zero (|3rd number| <100), and that the SM state (2nd number) is as commanded (3 initially, 13 later)
- zero (adjust) the K and B for the VS w/command 131 0 0
- increase the overload limits w/command: 1122 30000 30000 9000 9000 (last number is continuous ~amps*1000)
- command SM to enter the virtual spring state w/command 170 13 (confirm w/cmd 104)
- enable SM updates to the MC w/command 127 5
testjig config
- range 0 to 64000 (this is inital and final)
- because jig doesn't return to zero, for now run w/ plantarflexion and a range of 0 to 64000 (where zero is near ramp engagement, and is also the zero position for the ankle [where it was boot])
- set (de)acceleration, ~10 rads s-2 (or maybe larger? 20 rads s-2)
- RPM 1111 for stiffness test, 4444 for damping test
- set passes to a multiple of 2 (VERY IMPORTANT, otherwise, might stop at 64000)
- 1 second delay between passes (will allow fuse to re-zero)
some experiments at 1111 rpm
- low stiffness w/command : 131 1000 0 (appx 100 nM/rad)
- med stiffness w/command : 131 2000 0 (appx 200 nM/rad)
- high stiffness w/command : 131 4000 0 (appx 400 nM/rad)
some experiments at 4444 rpm
- moderate damping w/command : 131 0 400 (appx 40 nM/rad sec)
- moderate stiffness and damping w/command : 131 4000 400
results
- K value is approximate 12% small, or appx 20% greater when the effect of K2 is considered
- B value is approximate 20% larger, or +-10% when estimated robot screw friction is removed.
- RE40 to EC30 ratio is appx 3.3 to 1 (based on moment arms of 112mm vs 37 mm, and RE40 ratio of 22/20)
other
- 130 can be used to read the VS parameters (the first 2 are relevant in this experiment)
- 131 sets K and B : eg 131 K B
- 1120 can be used to read the electronic fuse parameters (first two values are maximum sums, the 2nd two values are the decay rates in q15)
- 1121 can be used to monitor the electronic fuse (first 3 values are the present sums, 2nd three values are actual current in q15)
limits
- testjig : use 1111 rpm, ~10rad s-2 acceleration, position of 0 to 64000, 2 passes, 1 sec between passes
- set fuse to ~9 amps w/command 1122 30000 30000 9000 9000
- VS K: don't exceed 400 nm/rad (eg 131 4000 0) - can result in oscillation or overload.
- VS B: don't exceed 40 nm/rad s (eg 131 0 400) - can result in oscillation