COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\|\\M1BDR30\M2SIGN57\M3NGR25\M4NGR20\F2\CVICARM
C00011 ENDMK
C⊗;
\|\\M1BDR30;\M2SIGN57;\M3NGR25;\M4NGR20;\F2\CVICARM
\F3\C154 EAST DANA STREET
\CMOUNTAIN VIEW, CA. 94041
\F4\←L\-R\/'7;\+R\→.\→S   Telephone:
\←S\→.415-965-0557
\F1\COctober 8, 1975

Professor George N. Saridis
Department of Electrical Engineering
Purdue University
West Lafayette, Ind. 47907

Dear Professor Saridis:
\J	A couple of  your  students have  contacted  me  regarding  some
problems  associated  with interfacing  and  controlling  your M.I.T.
model arm.   In this letter I  will attempt to  answer some of  their
questions.  I hope you will forward the letter on to them.

	The arm,  as delivered is  equipped with a velocity  servo in
manual control mode only.  In computer control mode there is no servo
of any sort supplied.  What happens when you switch from manual mode
to computer  mode is that  the velocity servo  is turned off  and the
is computer attached to the  current amplifiers for  each joint.   These
amplifiers are  different  from most  servo amplifiers  in that  they
produce  a current output  proportional to voltage  input rather than
voltage output.   This means  that motor torque  is commanded  rather
than motor voltage.  From servo theory, the motor velocity damping is
very  low in  this mode.   This means  that if  the commanded current
(motor  torque)  is below  the  level  required  to  overcome  static
friction and gravity, the arm  will not move at all.  If this current
is greater than  this friction  and gravity,  the arm  will move  and
continue to accelerate  until the current  is limited by  the maximum
supply voltage.   Thus the arm either  doesn't run, or it accelerates
to full speed.  To properly operate the arm, you MUST have  some sort
of velocity feedback.  You should also have position feedback too.

	The  reason  we use  current  amplifiers  instead of  voltage
amplifiers (which have higher natural damping),  is that the current
amplifiers allow control over joint torque and forces.  This provides
an extra  degree of control which  can not easily be obtained with voltage
amplifiers.  I  have enclosed  a reference  which (among other things)
gives  some of  the
reasoning behind the choice of current amplifiers.

	Your students  have also asked  for some more  information on
the  electronic circuitry.   The electronics  are made by  a separate
contractor.  He provided the original schematics, and has  gone over
them again.   Enclosed  is a more  recent version of  them.   I hope
they  include the details you are after.   I made a couple of on the
spot modifications to adapt the signal levels to your  system, so the
circuits  should be reasonably  correct, with  the execption  of some
resistor values.  In any event, you should be able to follow your way
through the circuit with a little patience.

	I hope this information helps you out.   Feel free to contact
me should you have any furthur questions.\.
\←L\→S\←R\-L\/'2;\+L\→L
Yours sincerely,


Vic Scheinman
\←S\→L
VDS:pdp10