COMMENT ⊗   VALID 00011 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00003 00002			MOTOR CURRENT SPECIFICATIONS FOR STANFORD ARMS
C00008 00003			MOTOR SPECIFICATIONS FOR STANFORD ARM
C00013 00004		GEAR RATIOS AND GEARING DETAILS FOR STANFORD ARM
C00015 00005	LITERATURE ON THE STANFORD ARM
C00016 00006	MOTORS
C00018 00007	ARM PROGRAMS
C00019 00008	
C00030 00009	Bearings required on arm
C00031 00010			SPECIFICATIONS FOR THE MODEL "STANFORD" ARM
C00036 00011			THE CARE AND FEEDING OF THE STANFORD ARM
C00053 ENDMK
C⊗;
		MOTOR CURRENT SPECIFICATIONS FOR STANFORD ARMS


	This file lists the specifications for the motors on the 
	mechanical arms. Arms are named by color or proposed color
	and joints are numbered starting at the base and working out
	to the hand.. 

	NOTE: When it says max. current limit it means that the motor
	must NEVER see  a current exceeding  this value even  for one
	millisecond!
	Also,  for sufficient  maximum arm  velocity, the  power amps
	should put out a maximum 24-28 volts d.c., bipolar. 

	

GOLD ARM  (built in 1969)

Joint #1- U9M4 motor (Photocircuits Corp.)- .84ohms, no max current limit.
	design for 15-20 amps pulse (100-1000 ms.) and 6 amps cont.
Joint #2- U9M4 motor- same as Joint #1.
Joint #3- E-576-01 motor(Electrocraft)-1.55ohms, 24 amps max.current limit
	design for 10-15 amps pulse and 4 amps cont.
Joint #4
←oint #5- same as Joint #4
Joint #6- 
Hand- NT-0739-C motor(Inland Motor Corp.)-16.7ohms, 1.55 amps max.current limit.


BLUE ARM   (built in 1972)

Joint #1-U9M4 motor- see above
Joint #2-U12M4 motor- .75 ohms, no max current limit, but can design for
	10-15 amps pulse(100-1000ms) and 7.5 amps continuous (5-10 secs.)
Jount #3- 3069-255 motor(Magnetic Technology)- 5 ohms,5.06 amps max. current limit.
Joint #4- BYLM-
Joint #5 BYLM-
Joint #6 Globe -
Hand- NT-0739-C See above


RED ARM  (built in 1974)

Joint #1- U9M4- See Above
Joint #2- U12M4- See Above
Joint #3- 402-14(Magnedyne)- 2.6ohms, 7.5 amps max. current limit.
Joint #4- 1937D-100(Magnetic Technology)-16.7ohms, 1.4 amps. max. current limit.
Joint #5- 1937D-100-  See above
Joint #6- NT-0741-C   See above.
Hand NT-0741-C       See Above.

		MOTOR SPECIFICATIONS FOR STANFORD ARM
			Feb. 22, 1975


Joint 1


	Printed Motors Inc.- Model U9M4T ( Motor-Tachometer in single unit)

	Cont Torque- 20 inoz
	Max. Cont. Stall Current- 6.2 Amps
	Terminal Res. .84ohms @ 25Deg. C, 1.21 ohms @ 150 deg. C.
	Torque Constant- 6.1 oz-in/amp
	EMF Const.- 4.5 V/1000rpm
	Damping Const.- 1.2 oz-in/1000 rpm
	Total Inertia- .008 oz-in sec sec.
	Regulation@ const voltage- 41.6 rpm/in-oz.
	Armature Inductance- <100micro-henries.
	Average Friction torque- 6 oz-in.
	Mecchanical Time Const.- .035 sec.

	Tachometer Constants
		
	Output Gradient- 2.25v/1000rpm
	Output impedance 1 ohm.

Joint 2

	Printed Motors Inc.- Model U12M4T ( Motor-Tachometer in single unit)

	Cont Torque- 66 in-oz
	Max. Cont. Stall Current- 7.6 Amps
	Terminal Res. .75ohms @ 25Deg. C, 1.07 ohms @ 150 deg. C.
	Torque Constant-14.4 oz-in/amp
	EMF Const.-10.6 V/1000rpm
	Damping Const.- 4.5 oz-in/1000 rpm
	Total Inertia- .033 oz-in sec sec.
	Regulation@ const voltage- 6.77 rpm/in-oz.
	Armature Inductance- <100micro-henries.
	Average Friction torque- 6 oz-in.
	Mechanical Time Const.- .023 sec.

	Tachometer Constants
		
	Output Gradient- 5.30v/1000rpm
	Output impedance 1 ohm.


Joint 3

	Magnedyne Inc.- Model 402-14 (20vdc winding)

        Peak Torque-300 in-oz
	Max. Stall Current- 7.5 Amps
	Terminal Res.2.75ohms @ 25Deg. C.
	Torque Constant-40 oz-in/amp
	EMF Const.-.3 V/rad/sec.
	Damping Const.- 4.4 ozin/rad/sec.
	Total Inertia- .029 oz-in sec sec.
	Average Friction torque- 9 oz-in.
	Mechanical Time Const.- .007 sec.
	Electrical Time Const.- .0016 secs.
	Max. continuous stalled current- 3.5 amps.

	Tachometer Constants   (Separate Unit) Servo-Tek, Inc.
		
	Output Gradient- 7.0v/1000rpm
	Output impedance 100 ohm.


Joints 4 and 5

	Magnetic Technology Inc.  Model 1937D-100-043 

        Peak Torque-20 in-oz
	Max. Stall Current- 2.8 Amps
	Terminal Res.4.3ohms @ 25Deg. C.
	Torque Constant-7.1oz-in/amp
	EMF Const.-.05 V/rad/sec.
	Damping Const.- .093 ozin/rad/sec.
	Total Inertia- .0015 oz-in sec sec.
	Average Friction torque- 1.5oz-in.
	Electrical Time Const.- .0005 secs.
	Max. continuous stalled current- 1.5 amps.

Joint 6 and Hand
	
	Inland Motors, Inc.  Model NT-0741-C 

        Peak Torque-6.6 in-oz
	Max. Stall Current- 1.55 Amps
	Terminal Res.16.7 ohms @ 25Deg. C.
	Torque Constant-4.5 oz-in/amp
	EMF Const.-.03 V/rad/sec.
	Damping Const.- .0086 ozin/rad/sec.(zero Z source)
	Total Inertia- .00013 oz-in sec sec.
	Average Friction torque- .35 oz-in.
	Electrical Time Const.- .0004 secs.
	Motor Constant- 1.10 in-oz/(watts↑.5)
	Max. continuous stalled current- .7 amps.


	GEAR RATIOS AND GEARING DETAILS FOR STANFORD ARM
			Feb. 22, 1975

Joints 1 and 2

	USM Corp. Harmonic Drives - Model 5C-100-2-BL

	Reduction Ratio- 100/1
	Moment of Inertia (Seen at input)-.007 oz-in sec sec
	Approx. Average Spring Const.- 130,000 in-lbs/radian
	Maximum Backlash- 3 minutes of arc .
	Maximum output torque- 800 in-lbs.(Harmonic Drive limit)

Joint 3

	Motor Pinion gear drives rack- 3 inches/turn.


Joints 4 and 5

	USM Corp. Harmonic Drives- Model 1C-72-2

	Reduction Ratio 72/1.
	Moment of Inertia seen at input end-.00047 oz-in sec.sec.
	Approx. Average Spring Const- 19,000 in-lbs/radian
	Maximum Backlash- 5 minutes of arc.
	Maximum output torque-100 in-lbs.(Drive limit)

Joint 6

	Two stages of spur gear reduction- 80/1 overall

Hand

	One stage of spur gears, and rack and pinion on output.

	
LITERATURE ON THE STANFORD ARM

MSPECS.S  Motor current specs for the motors on all the Stanford arms.

SPECS.S   General Specifications sheet on the current model Stanford arm.

ARM1.S    The Care and Feeding of the Stanford Arm (Installation and early operation).

AL Report- An Automation Language.

Lou Pauls Report- Arm Trajectories, and Servo Equations- 1972-1973 stuff.

ARM.PAL [11,BES] - Shimano's PDP 11/45 arm servo routines.

REFS.DOC  - Useful References on control and operation of the arms.
MOTORS

Globe motor details.  
Typical demagnetization results

As delivered - back emf equals 9.2
Take out magnets and replace- back emf goes to 8.2
Take out magnets and press like poles together until they touch-
back emf drops to 6.2 or so.
So we see the potential torque loss due to dissassembly.
Careful disassembly gives only a 157 torque loss, but fooling 
around results in up to 50% torque loss.

A good test for fully charged globe motor magnets is to see if it cogs.  If it
does, then the magnets are at least 75% charged.  If it doesnt, then you
have weak magnets.

Placing the poles of a globe motor between the poles of a large alnico magnet
seems to properly recharge the magnets.  As an example- recharging the motor above 
in this manner yeilded a back emf of 9.6.  Higher than the original number- but no
doubt, when put to use, the magnets can be expected to discharge a bit in the
presence of demagentizing armature current induced fields.

As a furthur note- testing the motor with the large charging magnet between the
poles gives an emf of over 15.  
ARM PROGRAMS
TO create a macro type
DEFINE  MACRONAME

THEN THE INSTRUCTIONS


TERMINATE WITH A BLANK LINE

TO EXECUTE, TYPE
MACRONAME

DO

TO SAVE A FILE TYPE 

BEGIN FILENAME
MACRONAME
MACROCNAME

END

TO EXECUTE A FILE TYPE
DO FILENAME



		Parts list for Stanford Arm

Joint #1

U9M4 Motor- Photocircuits Corp.  Glen Cove, N.Y.
FS11-00-04-5-014 Brake.  General Time, Torrington Conn.  
HDC-5C-100-2-BL (2 tooth difference model) Harmonic Drive-USM Corp.,Wakefield,Mass.
030/105 Tachometer- Micro Mo Electronics, Cleveland, Ohio.
KA 040 XP3 Bearing-Prec. 3-  Kaydon Bearing- Muskegon, Mich. 
KB 040 XP3 Prec. 3.  Bearing-  Kaydon.
Potentiometer Element Material- order from New England Instrument Co.
	Natick, Mass. (One sheet 3" x 18" of 100 ohms per square Resistofilm
	material will make all the pot elements.)

Joint #2

U12M4 Motor- Photocircuits Corp.
FS11-00-04-5-014 Brake-  General Time.
HDC-5C-100-2-BL- (2 tooth difference model) Harmonic Drive,   USM Corp.
030/105 Tachometer- Micro Mo Electronics
KA 040 XP3 Prec. 3 Bearings  2 Required-  Kaydon.

Joint #3

402-14 (20 vdc winding) Motor , Magnedyne Inc.  Carlsbad, Ca. 
FS17-00-06-5-014 Brake  - General Time Co.
8103-R5K-L.10-LT Potentiometer- Beckman Helipot.  Palo Alto, Ca.
477501 Bearing,  New Departure
MFS3KDD Bearing- Fafnir
2112N913 N- Tachometer- HICO. , Menlo Park, Ca.
G176 Spur Gear- Boston Gear.
G583-4 Rack-  Boston Gear.
34KDD Bearing- 16 reqd. -Fafnir.

Joint #4

1937D-100-12v. Motor- Magnetic Technology, Canoga Park, Ca.  
HDC-1C-72-2 Harmonic Drive.  USM Corp.
2112N913 N - Tachometer-  HICO
FG2-1078-902 Brake Armature-  General Time
3TKR-10-14U Bearing- Split Ball Bearing Div., MPB Bearing Corp. Lebanon, N. H. 
B544DD Bearing- Fafnir
B542DD Bearing- Fafnir
P120A5-150 Gear- Winfred M. Berg Inc.  499 Ocean Av.  East Rockaway, N.Y.
P120A7-52 Gear- Winfred M. Berg.

Joint #5

1937D-100-12v Motor- Magnetic Technology
2112N913 N Tachometer- HICO
FS08-00-04-5-014 Brake- General Time
HDC-1C-72-2 Harmonic Drive-  USM Corp.
KAA10XL0 Bearing- Kaydon
MFS3KDD Bearing- Fafnir (.3750x.8750 flanged,double shields, prec. 3 or better)
MFS1KDD7 Bearing-  Fafnir (.2500 x .6250 flanged, double shields-prec. 3 or better)
B541DD Bearing- Fafnir

Joint #6

NT-0741-C Motor- Inland Motor Corp.  Radford, Va.
050/004  Tachometer- Micro-Mo Electronics
FS08-00-02-5-014 Brake - General Time
B538DD Bearing- Fafnir
B539DD Bearing - Fafnir
AVF8K20  Bearing 2 reqd.-Fafnir (or S518FC-MPB)
AVF8K16  Bearing 1 reqd.-Fafnir (or S418FC- MPB)
AVF12K20 Bearing 1 reqd.-Fafnir (or S5632FC- MPB)
P64A21-100 Gear- Winfred M. Berg.
YWS6412 Pinion Wire- Boston Gear (1 foot req'd)
P120A5-126 Gear- Berg
PS120S2-12 Gear- Berg

Hand

NT-0741-C Motor-  Inland Motor Corp.
FS08-00-02-5-014 Brake- General Time
78CSB502 Potentiometer- New England Instruments
P120A10-130 Gear- Winfred M. Berg
P120S9-32   Gear- Berg
P96A5-90    Gear- Berg
PS96S4-16 s.s. pinion shaft- Berg (96dp, 16teeth) 
R2-14 Rack-Berg, 1 piece reqd., 64 dp, 11 inches long.
S1-10 precision rod- Berg (.1562 dia., 24 inches long, 303 stainless steel rod)
YWS6410 Pinion Wire- 6 inches-  Boston Gear (64dp, 10 teeth)
B25-4 Oilite bushing- Boston- 1 piece reqd.
S6316FCHH Bearing- MPB Corp.- 1 reqd.
S418FC- Bearing- MPB Corp. - 1 reqd.

Electrical Components Required (preliminary)

Tape Cable- Burndy or Hughes
Main Connector -Amphenol 57-40500 and 57-30500 -one each
Tape Cable Connectors- Cinch- 50-30-C10 - Four req'd.
#4 brake-  #40 gauge nyclad wire- 1/4 lb. spool is enough.
Hand connector- Augat 14 pin 3m cable connector and low profile dip socket
	to match- one each req'd.
3M Flex Cable
Shielded Cable-  Micro Cable Corp.

Notes: Most bearing brands are interchangeable- substitutions are OK.
	Gears from PIC, Sterling, Berg are interchangeable, sometimes
	different bore sizes or hub styles are acceptable in case of
	a particular number being out of stock.

All parts (or equivalents) listed above are  supplied in the standard
proposed kit of parts for the arm.  Some assemblies will be included,
where necessary.  Encoder version has encoders substituting  for pots
on the first three joints. 
Bearings required on arm

MPB
S418FC - 2 ea
S518FC - 2 ea
S5632FC- 1 ea
S6316FCHH- 1 ea
SR6FRHH -2 ea (or Fafnir MFS3KDD .375x.875 FDD, prec. 3)
SR4FCHH - 1 ea ( or Fafnir MFS1KDD7  .250x.625 FDD, prec. 3)
3TKR 10-14U (.625x.875 x.156, prec. 1 or better)
   

FAFNIR

B538DD - 1 EA
B539DD -1 EA
B541DD - 1 EA
B542DD - 1EA
B544DD - 1EA
34KDD - 16 EA


NEW DEPARTURE

477501 - 1 EA

KAYDON

KA040XP0 - 3 EA
KB040XP0 - 1 EA
KAA010XL0 - 1 EA

		SPECIFICATIONS FOR THE MODEL "STANFORD" ARM

LIMITS OF MOTION-EACH JOINT
#1- 300degrees total rotation
#2- 200 degrees (optional 300 degrees)
#3- 75 cm. total linear motion 
#4- 330 degrees- (optional 600 degrees)
#5- 220 degrees
#6- 330 degrees- (optional 600 degrees)
Hand- 10.5 cm. maximum opening.

MAX SPEED EACH JOINT
JOINT #      UNLOADED      LOADED WITH 2.2 KG.- AGAINST GRAVITY
1		4 rad/sec	2 rad/sec
2		2 rad/sec 	1 rad/sec
3		1.0 m/sec	.5  m/sec
4		5 rad/sec	2 rad/sec
5		5 rad/sec	2 rad/sec
6		8 rad/sec	3 rad/sec
Hand 		8 cm/sec	3 cm/sec

MAXIMUM LIFT
	Within  the  defined  workspace, which  is  a  50  centimeter
diameter hemisphere centered on a point on the same plane as  the arm
base and 40 centimeters from the centerline of the  base  plate,  the
arm will lift a rough steel block having a mass of 4.5 kg.

MAXIMUM REACH
	Maximum reach of the arm is about 1 meter.  Thus the arm  can
access points within a 2 meter diameter hemisphere.

ACCURACY AND PRECISION
	Within the defined  workspace, the arm  has an accuracy  of 5
millimeters  and a  repeatability of 2.5 mm when  using potentiometer
calibration tables and  12 bit resolution a/d.  With encoders on  the
first three joints, the accuracy is 3 mm and the repeatability is 1.5
mm. Resolution is equal to the repeatability.

OTHER DETAILS
	The arm runs on 120vac supply voltage.A power supply provides
+ and-  30 vdc for the servo motors  and brakes.The seven d.c. motors
are controlled through linear  current amplifiers which require a  +-
10 volt signal range from a DAC (8  bits min.). A cable  from the arm
provides   the  feedback  signals  to   the  computer  and/or  manual
controller. This cable  also powers all the  motors and brakes.   For
the  potentiometers and tachometers  a reference  power supply  and a
minimum of a 16 channel, 12 bit a/d is required. For the encoders,  a
quadrature  decoder   and  up-down  counter   for  each   encoder  is
required.The encoder electronics interface directly to the i/o buss.

TORQUE, FORCE AND DUTY CYCLE LIMITS
JOINT#		MAX.CONTINUOUS TORQUE	MAX.TORQUE(10%DUTY CYCLE-MAX. 10 SECS)
1		200 kg-cm		400 kg-cm
2		400 kg-cm		800 kg-cm
3		3.2 kg			7 kg
4		35 kg-cm		70 kg-cm    (1 kg-cm approx. equals 1 in-lb)
5		35 kg-cm		70 kg-cm             
6		10 kg-cm		20 kg-cm
Hand		2.3 kg			4.5 kg
		THE CARE AND FEEDING OF THE STANFORD ARM

			BY VIC SCHEINMAN

	This  paper  explains  some of the operating and maintainence
details of the Stanford Arm.

	INSTALLING THE ARM

	The arm must be bolted to a  solid  table  surface  or  other
suitable  mounting plate. The 1/2 inch screw threads on the bottom of
the base plate are for this purpose. Use them all! The wires  running
down  the  side  of the main column indicate the out of range area of
motion for joint #1, thus, these  should  be  placed  away  from  the
workspace.  The supply cable for joint #1 can exit either thru a hole
cut in the table surface, or thru the slot cut in  the  base  of  the
arm. The two wide cables running to the other joints should be strain
relieved in such a way that they do not get in the  way  of  the  arm
when   it   is   operating   in   its   normal  workspace.  A  little
experimentation will easily show  where  a  suitable  clamping  point
should be.

	Place  the  power amplifier and control box such that all the
cables from the arm will reach the box. Do not add extender cables to
the  arm,  as  this will increase the overall resistance of the motor
drive cables and will result in slower motions and increased response
times.  A  typical  location for the amp. box is under the table with
the cables being fed thru the table surface. Plug the amp.  box  into
the  power  supply. Again, do not attempt to extend the cable length.
The power supply plugs into 117 v.a.c. and is fused for  8  amps.  An
extension  cord  can  be  used  here  if necessary. For semi-portable
applications, where the arm is mounted on a dolly or cart,  the  amp.
box, and the power supply should be mounted on the same device.

	The  manual control box plugs into the front of the amplifier
box, with the cable orientation colorcoded as is the  case  with  the
cables  to  the  arm.  DO NOT PLUG THE CABLES IN BACKWARDS OR CONFUSE
THEIR ORDER! If the hand held control box is not plugged in, the  arm
will  not operate as the "OFF" mode is automatically selected in this
case.

	Plug the computer into the computer plug using a  50  pin  3M
ribbon  connector  wired  to  the  A-D  channels  and DAC channels as
described later. This cable need not be plugged in if the arm  is  to
be used in manual mode only.


OPERATION IN MANUAL MODE

	Set the manual control switch to OFF,  either  one  will  do.
Turn on the power supply, indicated by the pilot light. Place all the
brake switches in the ON or LOCK position. To grab and place the  arm
somewhere,  release the brakes on the proper joints, grab the arm and
move it to where you want it. Then LOCK the brakes. To move  the  arm
remotely,  Put  all  the  brakes in the LOCK position and then select
which joint you want to move with the  joint  select  switch  on  the
manual  controller. Now turn the speed and direction control knob and
the selected joint will move slowly. If anything goes wrong,  release
the  knob  immediately  and it will return to center, turning off the
servo and locking the joint. DO NOT ATTEMPT TO INCREASE THIS  MAXIMUM
VELOCITY-IT  IS  SET LOW FOR YOUR OWN PROTECTION. If the joint stalls
under too great a load or because it has hit its own stop (joints 3-7
only)  do not hold the knob on any longer than necessary, as this may
cause excessive motor heating and possible motor damage.

COMPUTER CONTROL

	To operate the arm in computer mode, the arm  must  first  be
properly  interfaced  with the computer. Thirteen A/D channels, 7 DAC
channels , 7 brake bit outputs and 7 enable channel outputs  are  the
minimum  interface requirements. For more than 300 degree rotation of
joints #4 and #6 you must have two more A/D channels. A potentiometer
element  power supply is also necessary. The paralleled resistance of
all the pot elements is about 200 ohms, so a 10 volt supply  must  be
capable of supplying at least 50 ma. To reduce precision requirements
of this supply, it helps to use an extra  A/D  channel  to  read  the
supply  voltage. The tachometers have bi-polar outputs, with one side
common. Should your A/D be single ended you will have to  provide  an
offset  voltage  to  keep them within A/D range. You may also want to
install external tach op. amps. to set the tach gain to provide  full
scale A/D signals(see table of tach maximum output signals).

	The  output from the DAC must be limited to less than + and -
15vdc. If you have a single ended output, an offset must be provided.
It  is  best  to do this in an output op. amp. Some means of clamping
the output to less than 15 vdc. should be installed  to  insure  that
the motor current limits are never exceeded, even in the event of DAC
amplifier saturation or catastrophic  failure.  The  power  amplifier
input  impedance  is  10k ohms. Full scale current is 15 volts input,
for each joint.

	The brake drivers require a TTL driver output. A  low  signal
turns the brakes off. To enable  the  power  amplifiers,  FET  switch
gates  are provided. These, too, require TTL high level logic signals
from the computer.

	To  operate  the arm in computer mode, the manual control box
must be plugged in and the mode selector knob set in "COMPUTER" mode.
On  the  present  model, the only built in way of stopping the arm in
emergency is to turn the mode select knob to OFF.  The  computer  and
manual  brake  switches  are  ORed together. Thus the manual switches
should be in LOCK position when operating the arm in  computer  mode.
Likewise,  the  computer gates should be low when operating in manual
mode.

	The arm should only be  operated  in  computer  mode  with  a
carefully  debugged  program. Some sort of duty cycle protection must
be included in the program to prevent overheating of the motors. This
will  normally  not  be a worrysome problem, but if the arm stalls up
against a surface, or else holds a large load against gravity for too
long a time, motor heating can be damaging. Prevent this by putting a
timeout in the control routine. Experience  has  shown  that  no  one
trajectory should take longer than 5 seconds.

	The  power  amplifiers  are  current drivers. This means that
they provide a current proportional to DAC voltage. The servo  motors
are very sensitive to overcurrents. Thus it is imperative that the 15
volt dac output level never be exceeded, otherwise demagnetization of
the  field  magnets  will  result  with  an associated reduced torque
constant (torque/current). Because of the freeness of all the joints,
current  is  proportional to joint torque. Thus, the computer command
can be interpreted as a joint torque command. This should be kept  in
mind when developing the servo routines.

	There  are  no  stops on several of the joints. Thus, various
protection features must be built  into  the  software.  It  is  also
suggested  that  one hand always be kept on the mode select knob when
debugging programs, to permit almost instant emergency switch off.  A
separate  emergency  stop  button  connected  to  the  I-O bus of the
computer is a valuable accessory, as the mode select switch will only
turn  the  power  drivers off. It will not insure that the brakes are
switched to LOCK position. This can only be done in the  computer  on
the present version of the hand controller.


TINKERING

	No doubt there will come a time when  you  will  want  to  do
something physical to the arm. Resist this temptation mightily!! But,
if the poor arm requires maintenance, and  no  one  in  the  know  is
around,  proceed  with  great  caution. What follows are some general
guidelines. Sometime in the great future, a service manual  of  sorts
will be issued. No promises as to when!

	The  first  point to remember is to keep your eyes open. Look
over the situation very carefully and try to  diagnose  the  possible
problem  before  opening  things up or removing anything. Look at the
layout drawings carefully.

	The second point to remember is that everything  should  come
apart  easily-  it  went together that way! If you have to use force,
you probably haven't removed all the screws,  or  else  you  are  not
supposed  to be taking it apart there. The motors must never be taken
apart. This means that you must not remove the armatures from  within
the  fields  of  the  open  motors,  or  open the cases of the housed
motors.  To  do  this  will  result  in  instant  demagnetization,and
resulting  torque  constant  reduction. Don't open the arm up just to
see how it works- you don't do it on  your  own  arms,  so  take  the
suggestion.

	The  third  point to remember is that there are lots of wires
running around the arm. Be careful not to break  too  many  of  these
when  taking things apart or you'll really have a mess on your hands.
Oh yes, if you must fool with the pots, keep your cottonpicking hands
off  the  elements unless you have some lilly white cotton gloves on.
And do things gently, the wiper elements are fragile and bend out  of
shape easily-especially during assembly or disassembly.

	Fourth-   you  probably  will  have  no  difficulty  assuring
yourself that you can maintain the arm. In case you didn't measure it
when  you took it apart, the brake armature spacing is about.010-.020
inches. Also, gears run smoother if there is a little bit of backlash
(free  play)  rather  than  none. Harmonic drives can accidentally be
installed anodal. This means that the flexible inner gear  which  has
two  less  teeth than the outer ring gear has been installed with all
the difference on one side, rather than one tooth difference on  each
side. You can tell that something is wrong because it will be hard to
push the wave generator (the ball bearing like  thing  on  the  motor
shaft) into place, and then the drive will be hard to back drive.

	That's about it for now, I hope you have read this far before
doing anything important. Actually, if you did read all the way  thru
to  here-congratulations,  you  are  one  of  the few people who ever
bothers to completely read anybody's instructions before  plugging in
a new "toy".