davidbuckley.net
home  >  Timbug II < Simplhex > Victoria 15 February 2015

Simplex and me, December 1994
Simplhex by David Buckley
September 1993

  • Simplhex photos 2000
  • Simplhex photos 2006
  • Basic Stamp-1 program
  • HexDump for PIC16C84
    Simplhex - Retired
    Design and building started September 1993.
    A self contained six legged walking vehicle using 12 model-control servos, it has an onboard control computer enabling it to walk whilst avoiding obstacles.
    Size - 13 inch long, 12 inches wide over wiskers, 6 inches high
    [33cm x 30cm x 15cm]
    Operational area - 3ft * 3ft minimum.

    In Sepember 1993 the University of East London, where I was working towards an MSc in Robotics, had an Open Day and I made Simplhex to demonstrate on that day in the Mobile Robot Lab.
    Simplhex uses twelve Hitec HS300 servos and was controlled at that time via an umbilical from an Acorn A3000 running !Servo. Unfortunately !Servo only allowed a maximum of four servos and so the twelve were connected as four logical servos. two control the up/down of two alternating tripods, one controls the clockwise/anticlockwise swing of the four corner legs whilst the other controls the clockwise/anticlockwise swing of the two centre legs.
    Power was supplied by four AA cells.

    In 1994 I obtained my first Parallax Basic Stamp-1 and in December 1994 decided to use it to control Simplhex. Since the Stamp didn't have twelve I/O lines, one for each servo, I used the same arrangement of four logical servos. Each logal servo was pulsed, directly, by a Stamp I/O pin. There was just enough room in the Stamp for a program to make Simplhex walk forwards until the one sensor, a microswitch on the nose, touched something then Simplhex would sit down and turn by holding the two centre legs straight out and rowing round using the four corner legs until it had turned sufficiently then stand up and continue walking.
    The photograph shows Simplhex with the nose sensor and the Stamp-1 board mounted on a piece of mdf held onto Simplhex by ? - was it 'blue tak' or rubber bands?

    Later I added two whiskers and changed the Stamp-1 for a PIC16C84. The Stamp program was compiled using Micro-Engineering Labs Basic Compiler.
    The circuit board in the centre of the back routed the signals from the Stamp board to the servos.

    Simplhex was voted the best hexapod at the Robot Olympics in 1996 and won a Plaque.
    [Actually the plaques were very disappointing, all of them handed out at the event were exactly the same with no inscriptions.]
    The best of the other walking robots were:

  • Reading University's hexapod which actually walked, sort of, although I don't think they had managed to get it to turn;
  • Kent University's hexapod, created by David Bissett, which used 1/4 scale servos (ie big) and could stand up and wave its legs about but didn't walk;
  • Manchester University's hexapod with three (or was it five) Transputers to control it - all it did was twitch a bit, they said they had problems with the processor comunication!
    Simplhex was beaten in speed only by a Meccano walker with only one motor (so it couldn't even turn). The power to the Meccano walker was supplied through an umbilical so when its handlers found Simplhex walked faster they just turned up the voltage from their power supply until it beat Simplhex at which stage the motor was working on about three times its rated voltage!

    Simplhex was used continually for several days at another exhibition, I calculated it must have walked about 15Km, wearing flats on the ends of the legs and wearing out the pots in the servos in the process, so servo operation is now jerky.
    I could have changed all the servos but I didn't have the heart to rip Simplhex apart and so although it worked very well it has now been retired.
    However on the 18th July 2015 when I tried Simplhex the servos on the front left leg almost didn't move so I decided to replace all five defective servos with the 'new' HS-300 servos Martin Smith (of East London University) bought as replacements nearly twenty years ago. Great, Simplhex now strides about like new.

    more photos of Simplhex in 2000 more photos of Simplhex in 2006

    5th June 2006
    Although Simplhex is now retired that doesn't mean it no longer works. Periodically it gets checked over and demonstrated. One thing that has puzzled me over the years is that originally Simplhex was very sprightly but in later years has been working as if it was very tired and could hardly stand up. Originally it worked off 4 AA NiCad or even 4 AA Alkaline cells and some of the NiCads were 10 years old! I have bought new Nicads from various sources, used high power Alkaline but nothing made any difference. The servos turn easily and nothing is binding so what could it be? Has Gravity increased? Unlikely! Even though some of the pots are suspect the servo mechanics are not worn. So I tried Simplhex from a bench power supply. At 5v and 3A Simplhex worked fine. Can it be that AA cells have changed and are now designed to limit their output current to 'safe' levels? Anyway I decide to put on an extra two cells and use a 5v regulator for the PIC and drop the voltage to the servos by running them through two 3A Si diodes in series. Success, Simplhex now works properly again.

    9 March 2007
    I went to a talk by Rex Garard at Sheffield University and my friend John Fox persuaded me to take a robot along and suggested Simplhex because he likes it. Last night I put in new AA Energiszr Lithium cells and tested Simplhex. Today in John's workshop I demonstrated Simplhex for Andy, another of John's friends who was going to the talk, and Simplhex was very tired dragging the rear feet. I tested the cells; all good voltage, but the supplied voltage of the battery after the diodes was only 4v! Testing each cell by connecting it in series with an analog test meter set to 10A indicated 2 to 3A was about all they could manage. So I rewired Simplhex to omit the two 3A diodes. Now Simplhex walks properly again. [But hmmmm the cells get warm!]

    18 July 2015
    I replaced the five worn out servos with 'new' HS-300 servos Martin Smith (of East London University) bought as replacements nearly twenty years ago.


    home   <   top   >