History of the Telstra Research Laboratories

 

The Junction Cable Tester

When I was a Trainee Tech, 1969 onward, the PMG was deep into its rollout of ARF and ARK crossbar telephone exchanges per the 1960 Community Telephone Plan to provide automatic telephone service to pretty much anyone who wanted a phone. This meant a huge installation programme of junction cables interconnecting at the telephone exchanges in a sort of hierarchical/mesh network. A typical junction cable would have 1200 pairs. Each time Lines Installers installed a cable, it would have to be tested - this would tie up a senior tech and two helpers for two to three weeks in an extremely boring job.

You'd use an MDF "test shoe" connector to connect an oscillator to each pair in turn, while at the other end, your mate, in telephone contact, would connect a termination and level measuring set. A third chap would look at the measuring set meter, calling out the reading each time. All pairs in a 100 pair unit of the cable had to be checked for crosstalk to all other pairs in the unit. That's 12 x 0.5 x 100 x 99 = 59,400 checks per typical cable. When a cross-talking pair worse than 69 dB was found, we were supposed to put a shrink plastic sleeve on the MDF block so that the pair could not be used. After a while, you could literally fall asleep on the MDF ladder while doing this work. All day long, every day in exchange MDF rooms, you could hear us going "90, 92, 98, 91, 94, 99, 99, 96, 96, 98, 68, 99, 96, 94, 99, aww shit was that a bad pair back there?" And we had to do some of it over again.

After doing this on about 5 cables, I decided there HAS to be a better way, and I'm not doing it any more. Back at the depot I thought out a method using uniselectors (a sort of stepper motor that could select one out of 25 sets of contacts - used in English Step-By-Step exchanges) pulsed with power transistors interfaced to the meter circuit of a Siemens TM set. A tech volunteered to make up a special connector that connected to all 25 pairs in an MDF block vertical at once. He came up with the goods - a real rough hand-made affair made out of brass shim, string, and leftovers. I built my transistor & uniselector unit, and we demonstrated to the depot supervisor that it worked - and all you had to do was connect it up, connect to the MDF block, sit back, and watch while it found all the faulty pairs, in five minutes, eliminating hours of drudgery.

Well, it worked great on the bench. In the field it was not very good. This was because MDF blocks were designed as solder tag blocks, not as a push-on male connector. The tags had flux and oxidation on them - you couldn't get a reliable connection. Back at the depot, the tech pulled apart his hand-made connector female, and manually punched a ridge on all the contacts with a sharpened up screwdriver, then re-assembled it. We figured that with a ridge on each contact, providing a pressure point, the act of pushing it on would cut through/clean the MDF tag and we'd get a good contact. Back to the exchange - it worked! We got reliable low resistance connections every time on every pair.

My supposed genius came to the attention of the Divisional Engineer. He was impressed with the immense manhour savings, but not with the rough handmade appearance. He decided to send me (still just a trainee tech) to Melbourne to find out how to do it properly. He temporarily made me a Senior Tech as the Professional Officers Association had some kind of industrial dispute on - they were making their point by only "going thru channels" this meant they would not talk to any tech - only the supervisor.

I arrived at Research Labs and met folk who seemed not to have heard about the industrial action. Right smartly I found myself surrounded by some Engineers, other sorts of whizzos, and a metallurgist who was expert on relays, switches and connectors. The metallurgist said he would design a proper tag/contact, vastly superior to my mate's "home-made" roughie. He burbled on at length about alloys, contact resistance, tensions, and tolerances. I realised that he was proposing a flat slide-on contact, and objected, pointing out we had tried that, and found you need a sharp pressure "knife" to cut thru the flux and oxide layer. All the TRL guys then ganged up on me, told me I knew nothing, and that I should leave it to the experts - i.e., them.

I returned to Perth, and after some months, a pair of TRL-built connectors arrived. They were beautifully made, very neat, with a professional style and finish way beyond what me and my mate could ever hope to come up with. They had custom plastic parts and metal work obviously made with precision on a many-tonne press. They must have cost quite a lot in tooling to make. But they had flat wiping contacts, not knife-edge oxide-cutting contacts.

We never did get to see if the TRL metallurgist's tags/contacts gave reliable low resistance connections though. The whole thing could not be fitted onto an MDF block. Apparently the TRL draftsman had got an MDF block from somewhere and took measurements off it. It happened that the PMG purchased its' MDF blocks from three different suppliers, and each made them to slightly different dimensions - about 3 mm variation top to bottom - as any decent field tech would know. And each manufacturer had fairly broad tolerances. Didn't matter a bit for the intended application, having wires hand soldered to them. But the TRL connector, most elegant as it was, could not be adapted - at best a different one would be needed for each of the three MDF block types. We stayed with our own "homemade" connector, as it's flexible design fitted all MDF blocks.

After testing about 20 or 30 junction cables without a hitch, some senior chap told me I must write it up as an official suggestion and send it to the Staff Suggestions Board. A year or so later I got back a letter. In it the Board said that they had in accordance with procedure had my suggestion evaluated by an appropriately qualified group (TRL). The letter then went on to say that "... my idea was not of any practical importance, but attached is a cheque for $10 as we want to reward and encourage young lads with initiative such as yourself." In fact my idea saved the PMG/Telecom millions. In subsequent years as I transferred to other postings, as one did, I got called back a couple of times to repair my uniselector beast.