perm filename NRL2.VL[1,VDS] blob sn#243735
filedate 1976-10-23 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00002 PAGES
C REC PAGE DESCRIPTION
C00002 00002 \|\\M1BDR30\M2SIGN57\M3NGR25\M4NGR20\F2\CVICARM
\F3\C154 EAST DANA STREET
\CMOUNTAIN VIEW, CA. 94041
\F1\COctober 21, 1976
Dr. Gerald Wilson
Naval Research Laboratory
4555 Overlook Avenue S.W.
Washington D. C. 20375
Enclosed is the repaired interface. We have added all the
multiplexers and the extra DAC. We have also changed the DAC for joint
#4. I have enclosed a list of corrections you should make to your
program, to reflect this new calibration. Just type DONE and open the
proper locations and rewrite them as specified. When finsihed, 1000G will
get you going again.
Also enclosed are 8 resistor-diode assemblies (one spare in case you
lose one). These are to be inserted on each servo card in the position
indicated between the 7411 and HI1-201-5. Follow the annotated sketch.
First remove these ic's. Then you must carefully cut the trace with the
xacto blade which is also enclosed. This is the thin trace running from
the feedthru hole to pin 8 of the left HI1-201-5. Now insert the assembly
in the hole and carefully solder it in place as shown. Use a long thin tip
soldering iron and be careful not to slop solder on any other traces. Now
put the ic's back being careful not to bend any pins under. This completes
That fix was for the case of the arm glitching after an OPEN command.
You may also get some glitching after a PAUSE or GRASP which immediately
follows a COARSE function. This is because the joint positions are read
after these two functions and if the arm is not where it is supposed to be a
correction motion is executed at the start of this next required motion.
This appears as a glitch. COARSE modified functions don't null out and
thus the initial error on the motion immediately following the PAUSE or
OPEN. All other instructions are OK, as the joint angles are not read in
these cases. Moral, don't use a COARSE motion just before a PAUSE or
I have also included some more documentation. First, the details on
how the interface works. You should also refer to the DRV-11 notes in the
DEC handbook. To make things even simpler, I have included a tape of our
interface test program. Load this like the arm program, and follow the
instructions. It is restarted each time by typing 1000G, and stopped by
hitting any key. It allows reading of individual or multiple channels.
It allows you to drive individual arm joints in Position or Current mode.
NOTE: THIS PROGRAM DOES NOT HAVE PROTECTION FEATURES LIKE AUTOMATIC
TIMEOUT, SO DONT LEAVE IT DRIVING JOINTS. IN CURRENT MODE, THERE IS
ALMOST NO VELOCITY DAMPING, SO JOINT MOTION IS VERY UNSTABLE...AND THERE
ARE NO COMPUTER OR ELECTRICAL LIMIT STOPS. PICK VALUES NEAR 000 TO START
WITH. A HIGH CURRENT WILL DESTROY THE ARM IF A MECHANICAL LIMIT STOP IS
HIT. I have supplied you this tape as a convenience... please dont wreck
the arm with it.
Other documentation includes a schematic of the clock and front panel
switch circuitry. Also included are a couple of arm drawings showing
internal details...they are not complete but this is all we have. I have
also included a couple of revised interface schematics showing pin numbers
for your reference. Please replace your drawings with these.
I feel that we have now completed our part of the contract, so
please authorize payment...furthur delays will hurt us both.\.
Enclosures: Interface test tape, Absolute loader tape, Servo card mod.
dwg., Updated program documentation showing construction of Transform,
Interface connector wiring documentation, Bit assignments
sheet, Instructions for modifying arm program becasue of change in DAC
#4, Revised interface dwg.,arm layout dwgs., DAC, A/D and MUX spec.
sheets., Resistor-diode servo card parts, and repaired interface.\.