Anyone have experience with a Shaffer automation system? I had an old one given to me but never put it into service. It was a rack of gear that could control network switching, R2R’s, carousel cart machines, etc. A large number of socketed ICs. Programed by keyboard and a seven segment Nixie tube display if I recall correctly.
Back in the day I used the MBN satellite relay pulses to trigger cart carousels to play local spots in the MBN breaks. The carousels were set up to auto advance to the next cart after airing. One carousel loaded with 30 second carts, another carousel with 60 second carts, and one cart machine to air hourly IDs. MBN also sent a relay to trigger the top of the hour ID cart. Built 555 timers to mute the network during the spot rather than mess with cart cue tones. All this without computer. Worked really nice except for when a carousel sometimes jammed.
The station had to migrate to computer automation when Moody began sending the same relay for 30 second, 60 second, and 2 minute breaks.
From: Allard, Mark G [mailto:MGAllard@unwsp.edu]
Sent: Saturday, February 03, 2018 1:52 PM
Subject: [CRTech] early automation [off topic]
I started my employment in ’79.
In reference to our early automation system, I cut my teeth on a Harris System 90 – about a dozen R2R’s, 4 carousels, 4 cassettes, and 4 time announce carts.
The system was installed a couple of years before I got there.
You haven’t lived until you’ve made time carts! Each 30-minute cart had 12 hours of minute-by-minute announcements. One cart had all even minutes. The other was odd minutes.
When all done, there could not be more than 2 minutes of recycle time or the announcements would get out of sequence. (Start OVER)
Battery backup was a car battery behind the racks with a trickle charger attached.
It felt like there were between 500 and 1000 IC’s in that thing and they were ALL socketed.
Most of the sockets had very little spring tension and on top of that, most of the IC’s pins had turned black from oxidation.
This made the system incredibly fickle. It would change time, change memory location, lose its mind, or get a corrupted character that would lock it up.
After a crash, I remember my supervisor would advise to power it down then just exercise all the IC’s and try again.
Over time, we removed almost every socket. Then the system was reliable.
Control and monitoring was done with a terminal – seven segment displays and keyboard.
Memory backup was on 1” wide paper tape. It would take about a ˝ hour to load.
If a tape got damaged, we’d get out the Scotch tape and use a pencil to punch holes.
That’s where I got ‘hands on’ experience at learning HEX code.
The timebase was controlled by a crystal in a temperature controlled oven.
During the day, the system was doing double duty. Half was AM, the other half FM.
At night, AM was daytime only so FM took advantage of all the R2R’s reducing the number of times automation needed to be fed.
Boy, am I glad those days are gone!