By the spring of 1956, the team working on the Suppressed Aerial System was being wound down and team members being re-allocated into other projects. In consequence the team leader of Suppressed Aerials - Ken Coppin was looking around for another project, which would allow him to use his considerable knowledge of 'Wheat-stone Bridge' principles.
Recognising that these principles could be applied to automating tool-room and workshop equipment such as lathes and mills, he was able to persuade the Chief Engineer at the time that it would be advantageous to the business if he were allowed to research and develop this to see if it resulted in a workable system.
Ken was obviously able to put forward a good argument that if he could develop automatic machine control, this could be not only be applied to improving productivity within EKCO but could also be a product, which would be saleable and give the business, diversification from radar and communications equipment.
Ken was obviously a good salesman as well as a good engineer since he was not only given the go ahead but was provided with funding to begin development work albeit with only George Dring to support him.
In simple terms a Wheat-stone Bridge is a balancing system, where Ken Coppin used a sensitive centre stable relay to carry out the balance function, which initially gave three levels of control best described as coarse, medium and fine accuracy.
In the system developed by the Machine Tool Team, they used a series of wheat-stone bridges, the 'Master Potentiometers' of which could be set up to the desired measurement. The drive unit turned slave potentiometers in the servo head until voltages were balanced out.
The first three digits of a co-ordinate were selected through counters coupled to the slave potentiometers, which count the number of turns made by the lead screw. The last three digits were set up on separate potentiometers but resolved by a single slave potentiometer.
The drive unit (fixed to the X and Y adjustment machine lead-screws of the machine tool) was part of what was described as a 'bang-bang' servo system and stability was achieved by the use of a slow final approach speed to the desired position. Note: the first three decades operated a large motor and the last three a small motor.
Pre Production Control Unit
Photograph Courtesy of George Dring
The above photo shows a Pre-Production drive unit fitted to a Jig-borer at Malmesbury during tool room trials and giving of demonstrations to possible customers. Note X Axis not shown.
To demonstrate the Machine Tool control versatility, an early trial was made using a engraving machine since this is very demanding in terms of moving the engraving tool tip in a the X / Y axis. The photo below shows the set up.
Engraving Machine
Photograph Courtesy of George Dring
The machine tool system was designed to be fitted to existing machinery where it could automatically re-set a 10 T.P.I lead screw to + 1/500 of a single turn and could control up to 1,000 turns. The final approach was uni-directional allowing the affects of free backlash to be avoided.
Gearing was used in the system to allow use of 4, 5, 6, 8 T.P.I lead screws in addition to 10 T.P.I lead screws and the same basic accuracy applied to these various alternatives.
The re-set accuracy of 1/500 of a turn represented 0.0002 in linear movement of the worktable and a table speed of 20 inches per minute was normal although it was possible to obtain higher table speeds with lower repeat accuracies or visa versa. Both traverse movements operated together and the system was designed to be also suitable for radial and circular tables etc.
General Description of system.
As offered to clients, the system comprised an E118 combined control unit for both traverses and a traverse unit (E119) for each lead screw movement. Single control units were also available calibrated for circular measure. The system incorporated safety and interlock circuits to help prevent incorrect operation.
The E118 Control Unit was normally mounted on a stand, which was orientated to a convenient position for operators.
Ekco Drawing Office Print
Drawing by Colin Forward
Shown are the 'Decade dials used to set up the 6 digits of each ordinate, the number chosen showing in the window, which was illuminated from behind. The lamps also being used to show the correct functioning of the circuits.
The system had an 'auto/manual/zero' switch which when set to 'manual' enabled each table to be moved in either direction at full or slow speed.
E118 Control Cabinet Front
Photograph Courtesy of George Dring
E118 Control Cabinet Rear
Photograph Courtesy of George Dring
The control cabinet had 4 relay decks, 2 of which are shown removed in the rear view. There were 2 rows of 'Yardley switches, only the top row of which can be seen here, these called up the dimensions required i.e. the X and Y settings, converted to drive 4 motors, 2 each per head.
The E119 Traverse Unit was the second component and comprised a drive unit and a servo head coupled by suitable gearing to a slotted output shaft. This output shaft fitted to the existing lead screw either directly or through a suitable coupling.
The Traverse unit incorporated an AC motor, which allowed movement at a rate of 20 inches per minute unit the units was within 0.1 of an inch when it switched to a slow speed mode for the final approach to correct position.
The Servo head incorporated some clever features one of which was the ability for the user to datum shift to a new zero position.
Once all the basic design work had been completed by 1958 trials were undertaken with various customers principal of which was Atlantic Cables of Southampton although AWRE Harwell and a company called Henry Simons were also early customers.
One of the interesting aspects of adopting Machine Tool Control was that this showed up faults in existing lead screws, either caused by wear or in some cases by faulty new manufacture since up to the advent of M/C control errors in the 'lead thread' could be manually compensated for.
Where this was evident, EKCO recommended users fitted 'lead screws' manufactured by 'Beavers' in Canada since these were produced with a precision ground thread fitted with a re-circulating ball nut and Beavers could supply them to customers requirements.
Where customers required more accuracy, the 'lead screws' could be calibrated (the Beaver one were) and a follower covering the full table movement fitted to the table. This follower operated a potentiometer supplying an extraneous voltage to the last bridge, thus errors could be eliminated.
In 1958 development work also started on 'Punch Card Control', which was perhaps one of the very early applications for this method of recording and storing information although each hole had to be individually punched – a somewhat laborious process.
Despite the system being technically advanced for its day, it was under-funded and sales were slow not helped by the fact that there was no engineering support available to the customers unless George Dring himself visited the client. This situation was not rectified until circa 1961 when a support engineer – on the road – was employed.
With the move of EKCO to Southend, Ken Coppin brought the machine tool system to Southend on his own and low level development work continued until the mid 60's when all further work ceased.
With hindsight, the machine tool control system if properly developed and funded could have been an outstanding system but this was not to be.
As it was, the first effective system was promoted and sold by EMI-MECH albeit at considerably higher price than the EKCO system.
