A J25 TIM (Ticket Issuing Machine)

Its microprocessor intelligence, printing and reading technologies and automatic features make it the most advanced terminal we've ever designed.

Built for full time continuous sell/cash service, it is completely human engineered for simple and fast operation. Your tellers can sit at the keyboard and have everything within easy reach. Errors due to fatigue are reduced and the terminal makes both selling and cashing transactions virtually automatic.

Concise Keyboard

Teller Display

J25 Tickets

Automatic Cancellation Option

Automatic Cashing

Betting Slips

Automatic Reading

Reliability Engineering

Backed By Experience

That is the end of the sales brochure. I remember the J25 well. I worked for ATL in Brisbane for something in the order of two decades continuing through several takeovers. Work commenced on developing the Brisbane VAX11 based totalisator system in 1986, to replace the PDP11 based system. The new TIMs for this project were the J25s. We still retained many of the J22s as well. This J25 image highlights a feature of all ATL electronic totes. The first computer based tote ATL developed which was for the New York Racing Association, is regarded as one of the first non-stop-systems. For the layman, that means disruption to service is intolerable. To satisfy this requirement, the systems have to be fault tolerant. In other words they have to continue to function normally, despite any equipment failures. The way this is done is to eliminate all single points of failure by providing redundant equipment. That is why every ATL tote system was a duplex or triplex system. In other words the central tote computers consisting of the transaction processors and front end systems were duplicated or triplicated. An example of this duplication can bee seen in this image. As the front end system is duplicated there are two paths of communication between the computer room and every TIM on the race track. That means every TIM is connected to two networks. On top of the operator screen in the image there are labels for annunciator lights. The fifth and sixth lights from the left, which are not visible as the machine is turned off in this image, are labelled Poll 1 and Poll 2. These are two communications status lights. During normal operation they should both be flashing which means the J25 is receiving polls from both of the front end processors in the central computer room. Should any fault condition arise, causing a failure of either network, or either front end processor, or either of the TIM's communication ports, all bet traffic will still operate on the other one. For interest, all the lights in sequence from the left hand side on the J25 are Power On, Ready, Bet Stored, Non d:s runner, Poll 1, Poll 2, Paper out and Fault.

An observation by Mick Gulovsen

In May 2015 Mick Gulovsen, who was an ATL engineer in Melbourne, working on the computer totes, made the following observation. I agree with him and I too have not seen a J25 without the ticket reader and or a card reader. I previously asked Neville Mitchell about this anomaly and Neville indicated that the J25 in the image was an early prototype. Mick's observation follows:

I saw your picture of the J25 and my first impressions were `That's not right!!!' I started on J25 and worked on them for 7-8 years before they were replaced in 1993 by the PWT (which is still going strong in Victoria).

Now, I expanded the J25 image and was writing to you about the screen as it appeared that the photo showed a LCD/Plasma/EL type screen instead of its 1 line by (32 I think, could have been 20 or 40) character Fluorescent display but then I looked more closely and realised that it was the polarised fascia and the writing was words like POWER POLL A B etc that had LEDs behind or under them. The screen itself is not visible in the picture. Indeed I suspect the screen is actually not present in that unit. Further, the reason it did not look right was there is no ticket reader on the right hand side. That is covered up by a blanking plate, I have never seen a J25 that wasn't fitted with a ticket reader.

In fact I have a little story about the ticket reader. There have been several times an operator was out by $321 or a combination of the 3 of those numbers, and one day I went to plug up a J25 to test it for the race day and I found a ticket for $321 in the chute. It transpired that a technician went to replace a faulty PWT with another unit and shut the keyboard flap just before he pulled the power cord. This causes the 3 bottom dollar value keys 3, 2, 1 and TOTAL key to be pressed by the edge of the ticket reader and produce a valid bet (not always a ticket as it depended on how quickly the technician pulled the power lead).

Webmaster's note: You can see the 3, 2, 1 value keys in the keyboard in the image, they are the three blue keys in the second row up from the bottom row. The TOTAL key, also mentioned, is the double width red key to the right of the three blue ones mentioned. When you close the lid of the machine for transportation, the keyboard which is hinged is pushed up and sits on top of the face of the machine. In this position the delicate parts of the machine are protected by the strong lid that the keyboard is mounted in and the handle that supports the keyboard when it is open, can be used for carrying it. The left and right arms of the handle can be seen either side of the keyboard projecting down from the body of the J25. As Mick indicates, in the closed position the keys he mentions, correspond to the position of the ticket reader chute, which is wide and squat and depresses the four of them. As Mick mentions, the reader chute is not present in the prototype machine in this image, but is installed in the production machines.

I like Mick's example, as it demonstrates how obscure electronic problems can be. This is the sort of problem that can take quite some time discovering what is happening. The first step to identifying a problem is recreating the symptoms which can be difficult with such an obscure circumstance of closing the lid then cutting the power. Closing the lid is not something that would have come to my mind in the initial list of most probable causes of the fault. That would come much later in the light of other evidence or as a result of scraping the barrel for less likely causes.

I recall an obscure keyboard problem, with a much later generation TIM in Queensland, called the PCTIM. It took quite some investigating, to discover the cause of the value of bets sometimes being twice the expected amount. As this was a rare event it did not receive a high priority and was regarded by other engineering staff as finger trouble (operator error). The first clue came when I realised that the operators complaining about it were amongst our elite. I went to some of these operators during meetings and asked if they could demonstrate the problem. Try as they might this was not possible. One thing I did glean from their attempts to demonstrate the problem is that it was probably a result of high speed entry of bet details and that it was a particular bet sequence involved. In July 2004, I sat down with an application that just echoed every key as it was depressed and went through the sequence identified by the operators. It took some time, but I eventually discovered that when entering the sequence in as rapidly as possible, I could change the sequence being displayed by the key echo program. It turned out that once the dollar key was depressed, indicating a value was to be entered next, the TIM did not wait until it was released and if a one key was depressed whilst the dollar key was still active, the first digit of the value would become two. Once identified, the permutations of this problem were more widespread, depending on what the first digit of the value was whilst the dollar key remained active and this changing of one to two, was the main problem focused on as it was the most commonly used. I will not engage in a full description of the problem but one perception of this fault was interesting. There are two value keys on the PCTIM and the operators felt that the use of the right hand dollar key was better than the left. The results of my testing confirmed their impression, showing that, if you hold the dollar key down whilst entering a number, 6 out of 9 combinations end up working out properly with the right hand dollar key whilst only 3 out of 9 combinations end up working out properly with the left dollar key. A memo to the operators warning them of this problem was the first action taken and ultimately a change to the keyboard firmware eliminated the problem.

Information from Kevin Johnston

My fellow colleague Mick Gulovsen has passed on the emails you have sent on the history of the tote and suggested that I may have a machine or two that might be added in picture form to your very comprehensive list. I've had to dig them out of hibernation from my fathers garage (Ray Johnston A.T.L.ian from 1968 - 1996 retired early due to health reasons but still going strong). They have been in storage for best part of 30 years and I'm quite surprised to see how heavy they actually were / are as I used to carry them a lot between racecourse to racecourse in my early days on the tote.

Unfortunately I don't have the whole collection as I'm missing a J18 which had lights under the keys which was modern for the time when it was used. Attached photos are of a J6 (I think, as I didn't have the pleasure of working on that model) which weighed in at 38.75kg, cast iron top on a wooden frame with a swing arm that the operator would push forward and down for a win bet and backwards and down for a place bet. This terminal was followed by the J8 which was the first machine that I worked on (when I say worked on that basically means I unwound the ribbons and had an ink pot and paint brush to re-ink the ribbons and wind them back up. A dirty job, but good for a young fella). The J8's weighed in at 38.25 and had nice stainless steel lids (no cast iron top) and a metal casing and were just an upgrade of the J6 with the similar action of a swinging arm, going from the numbers 1-24 with the same forward, backward betting style although the arm was now streamlined and covered in Bakelite. The third betting machine that we pulled out of the shed surprised me as I had no recollection of it. It is very similar to the doubles machine known as the J10, which has a set of red keys for the 1st leg and black keys for the 2nd leg and a larger button at the side to place the bet after selections were made. This machine was obviously an earlier version of the J10 as it had a smaller button positioned horizontally under the selection keys to place the bet. Although this was a slightly smaller machine than the J10 it weighed in at 47.65kgs (opposed to J10's 47.25kgs) and no handles!! This should be followed by the J18 (my missing piece) which was referred to as a TDQ (able to place a bet on trifectas, doubles & quinellas).

The following image of Kevin's shows a production J25 and this is the machine that Mick, Kevin and I remember working on. This one has both readers installed in it, instead of the blank panel to the right of the operator display above the keyboard, in the image at the top of this page. The upper reader is the ticket reader and the lower one with the wider ticket entry chute is the card reader.

Along came the blue J25a's ('79-'80) I noticed as did Mick G the picture you have on the site that it didn't have a ticket reader, hopefully our pictures will rectify that. There was another J25 model before the J25a which had an identical screen, keyboard and upper casing but all the internals were in a casing that went all the way from the bench level to the floor (quite monstrous and I believe were promptly shipped off to Sth Africa if I recall correctly). I apologise for waffling on a bit but there aren't many of us left that are involved with oncourse betting terminals. I can't see my children dragging out an ipad or mobile phone from under my garage bench in 30 years to reminisce about the glory years of betting on the tote.

Acknowledgements

• Thanks to Peter Collier for the copy of the company sales brochure titled The J25 is the Complete Sell-Cash Terminal
• Thanks to Neville Mitchell for the image of the J25.
• Thanks to Mick Gulovsen for his comments on the J25 image and his J25 anecdote.
• Thanks to Kevin Johnston for the image of the production J25 and the four other ATL terminals as well as his associated text descriging them.