This history page contains a photograph which is one of several belonging to the photo gallery pages which are part of several pages relating to the invention of the world's first automatic totalizator in 1913 by George Julius, and Automatic Totalisators Limited the Australian company founded by George Julius to develop manufacture and export these systems.

A Horse Adder with counter wheel GT display

This image shows a counter wheel adder. There are many of these in the machine room of a totalisator system. This one is similar to the one in a video clip on the the British Pathé Website to which there is a link in the Three more ATL systems in Asia/Links to other pages chapter of this website. The British Pathé Website indicates this type of adder was in use on a Sydney racetrack and indicates a year, 1927. An Automatic Totalisators Limited promotion document titled The PREMIER (JULIUS) AUTOMATIC TOTALISATOR, written in 1930, has an image of The Paddock at Randwick Racecourse. The annotation for this image reads: FIG. 3. "Paddock" Installation on the Randwick Racecourse, Sydney, N.S.W. This is the first all-electric "Premier" Totalisator built and installed in 1917, and still in operation.
More after the image...
Click here to go Back
Click on the image to go back to the Photo Gallery

There is no photographer's stamp on this photograph.


From the information above, I deduce that this Adder probably belongs to the 1917 Randwick system as its successor Julius Tote was installed in 1935 after this information was published. Other possible candidate tote systems for using this type of adder are Canterbury Park Racing Club, Rosehill Race Club, Victoria Park Race Club and AJC at Warwick Farm. As much was made of the 1917 Randwick system being the first totalisator to implement electrical elements, making it electromechanical, and giving it the name at the time of Electric Totalisator or as the British Pathé website calls it the wonderful electric Totalisator, I suspect it is highly probable to be the Randwick totalisator. The Randwick system installed in 1917 was a big system with 150 terminals and I think it is the world's first large-scale, on-line, real-time, multi-user system.

I have presented the following commendation from the secretary of the AJC as it refers to the 1917 Julius Tote at Randwick. He mentions the TIMs (Ticket Issuing Machines) as well as the selling staff operating those TIMs. He refers to the TIMS as Automatic Totalisator Machines. These TIMs were all attached to a machine-room full of adders of the type shown in the image above. As this system catered for a maximum of 42 runners in a race, there would have been 43 adders in the machine room, one for each runner and another for the Grand Total. As there were three central processing systems in this 1917 installation, there must have been three of these machine rooms. C.W.Cropper Secretary Australian Jockey Club wrote the following on 22 July 1922:

I have pleasure in stating that the "Premier" Automatic Totalisator Machines, which were installed at Randwick Racecourse and are operated by your Company on contract under the supervision of the Club, and which were first used at the Club's Spring Meeting, 1917, have, since their installation, given complete satisfaction. Their accuracy is unquestionable and the rapidity with which tickets are issued by the experienced operators leaves nothing to be desired.
The second image after this one in the Photo Gallery of this website, shows the type of TIM C.W.Cropper is referring to. To view this click on the image above and select the image thumbnail of the second image after this one.

Having mentioned the TIMs and their connection to the adders like the one above, it is worth mentioning that the 1917 Randwick system was the first of the electromechanical systems however it did not implement the front end systems utilising electromechanical TDMs (Time Division Multiplexers), which at the time were called Distributors. These were developed shortly after and mentioned in George Julius' Mechanical Aids to Calculation white paper written in a 1920, which was presented to the Institution of Engineers Australia as part of a demonstration that same year. Without the front end system, only one TIM could be attached to each solenoid in the adder. The demonstration system that was built and tested, which had the new front end system, was capable of supporting 1000 terminals and a sell rate of 250,000 per minute. Having mentioned the Institution of Engineers Australia, it is worth mentioning that George was an instrumental founder of that organisation. The following image in the Photo Gallery of this website shows the type of adder that was used in this demonstration system.

The Old Main Tote at Randwick 1917 Image of the old main tote at Randwick

The above image shows the old Main Tote House at Randwick which was custom built to house part of the 1917 Julius Tote. The adders like the one in the image above were housed on the upper central upper level under the central clock tower. The counter wheels of these adders were placed so they were visible from the outside in the long horizontal slot like windows in this building which are aligned in 3 rows of 14 adders. Three rows of fourteen means there are 42 adders one for each of the 42 possible runners in a race mentioned above. Below the clock in the central tower, there are two indicators. The upper one that looks like the taller of two black oblongs in a white frame, is a dividends indicator which is controlled manually. The lower one is the pool grand total indicator which will have the grand total adder sitting behind this window.

This Adder is an example of mechanical computing and is a small part of an electromechanical large scale real time multi user system. This is a four shaft adder as there are four main adding shafts. There are two shafts side by side that span the width of the adder at the bottom of the image. There are two more shafts between the right hand shaft of the two mentioned and the counter wheels. There are ten escapement wheels and associated solenoids per shaft giving a total of forty escapement wheels and solenoids. The number of teeth on each escapement wheel determines the value of the bet recorded. The more teeth the lower the value. The rotation of every escapement wheel is added by the epicyclic gear train connecting the escapement wheels such that the rotation of the adding shaft at any particular point along the gear train is the sum of the rotation of the escapement wheel at that point and every other escapement wheel upstream from that point, away from the counter display wheels.

In the previous paragraph I have deduced from the image that this adder has forty escapement wheels and solenoids. George Julius adds to the belief that this adder is the type used at Randwick by confirming that the adders in the 1917 Randwick Julius Tote indeed had forty escapements, which I have referred to as the escapement wheels and solenoids. George's reference to this appears in his Mechanical Aids to Calculation paper and this reference follows:

This modification has very greatly reduced the amount of adding gear required in the machine, as in the new type four, or at most six, escapements perform the same duty as was previously performed by the forty escapements in the Randwick type of machine.

The modification George mentions above, is the introduction of the distributors which were time division multiplexers as mentioned previously.

The Mechanical Aids to Calculation chapter of this website contains relevant extracts from George's white paper of the same name. To read this, click on the image at the top of the page to return to the Photo Gallery. Then Scroll to the bottom of the page and select the Go to the index button in the Navigation Bar at the bottom of the page. Finally, select the chapter titled Mechanical Aids to Calculation.

Note that there is a paddle wheel at the end of a shaft on the left hand side of the adder just below the counter wheels at the back. Mike Bell, an ex ATL Project Manager and Software Engineer informed me this is an inertia brake. The counter wheels are arranged as a decade counter each counter wheel rotates one digit after ten digits have been displayed in the counter wheel on it's less significant digit side. When they were operating the units counter wheel could have such rotational velocity that it appeared as a blur. This is well demonstrated in the British Pathé Website video footage. As betting is erratic there was significant stress placed on the mechanical parts. The worst case scenario occurs when the race begins. The betting activity goes from near its peak to nothing in an instant. Inertia prohibits machinery from stopping in an instant.This paddle, assisted braking by introducing aerodynamic drag, when the sell rate was rapidly declining. In addition, to control shock associated with inertia limited parts of the system, there is a device called a storage screw, which is well covered in other images of this photo gallery.

Briefly, the storage screw is a form of buffer memory which records transactions from the quick to accelerate adding shafts, and is read by the inertia limited parts of the adder which empty the buffer as they catch up.

Bob Moran has identified parts of this adder during his investigation into the 1917 Randwick system. Bob is a mechanical engineer who has done much restoration work on Julius Totes for exhibition purposes, long after they ceased operation. Additionally Bob has retrospectively created engineering drawings of parts of the Julius Totes, which are superior to the ones the company that manufactured the Julius Totes produced. Bob has also created Julius Tote demonstration systems, which can be hand operated and demonstrate how they used to work. Bob has also created what he calls the Discovery Shed to inspire young minds and generate interest in Science Mathematics and Engineering.

Bob discovered that the horizontal tubular section with large cog wheels at both ends of it, on the left hand half of the adder, immediately behind the left hand adding shaft at the front of the adder, is a storage screw. This buffered the transactions from all the adding shafts, until the units counter wheel could catch up. The storage screw had a device which today would be called a closed loop servo system associated with it, which ensured the units wheel did not suffer the shock associated with a sudden stop and instead was slowed before coming to a gentle halt.

Bob also identified the variable speed friction gear at the far left rear of the adder behind the paddle wheel brake. It consists of two wheels at 90 degrees to each other with the right hand wheel in contact with the left hand wheel and capable of moving in towards the centre of the other wheel or towards the rim of the other wheel. The right hand wheel is padded around the perimeter increasing friction between the two wheels, imparting motion from one to the other and the movement of the right hand wheel in and out effectively alters the gear ratio between them. This variable speed friction gear, is the control section of the closed loop control system.

Bob also found a very important part of this closed loop control system, which in electronics would be called the feedback path, which Bob calls the speed control lever. It is somewhat hidden and runs between the left hand end of the storage screw shaft, where the position of the storage screw within the body of this device is sensed, and runs to the variable speed friction gear. A segment of it is best seen near the bottom of the adder, between the second and third horizontal shafts rearward of the storage screw shaft. It runs beneath the second and third shafts mentioned. These three shafts have a unique appearance to further identify them. The storage screw shaft looks like it has a raised helix on the surface of it, giving it the appearance of being threaded, which it is on the inside. The second shaft looks square with longitudinal channels cut into it. The third shaft is a shorter, plain circular reduced diameter shaft.