John Loughmiller, Contributing Writer
Several years ago I worked for a company that manufactured professional grade videotape recorders used by television stations throughout the U.S. The recorders used an arrangement of four heads positioned every 90 degrees on a disk. A 2-inch-wide videotape was pulled past the rapidly rotating disk and each head recorded about 17 lines of video. The whole thing worked very well most of the time because the basic engineering was settled, having been around since 1956, albeit vastly improved by the 1980s when the problem described below occurred.
One particular model used a vacuum loading system to position the tape in the transport and there were two transports, side-by-side. Used for playing commercials, the normal procedure was to initiate a dual transport load, shortly before the commercials were scheduled to play, via a large rotary storage drum that held 24 cassettes. Additional cassettes would be loaded as the preceding ones were played, rewound and placed back in the rotary drum.
Late one afternoon, I was the senior engineer on duty providing telephone support and received a call from a station complaining that their machine would be happily playing a string of commercials and then suddenly stop in the middle of a sequence. When it stopped, it would ignore the rest of the sequence, rewind the two cassettes previously loaded in the transports and place them back in the drum. The problem only happened during the station’s six o’clock news program which was really bad because, like most TV stations, the newscast’s commercial slots were premium placements and the loss of an entire two minutes of commercials was a very bad thing. The station management was not happy, which meant the chief engineer was really unhappy and threatening to drop the VTR off his loading dock into the Dumpster.
I poured over the logic diagrams trying to find some failure sequence that could cause such strange behavior but came up dry. The next day found me on an airplane and by 6 p.m. I was perched on a stool in front of the machine.
The first commercial break went fine, but the next one dumped everything midway through the second cassette. There was no warning, it just dumped. Diagnostics found nothing wrong.
The next day’s six o’clock newscast found me with a four-channel storage scope probing the logic bay and, once again, the machine dropped the sequence, this time a bit later in the newscast. Again, no clues were present. It was time to recalibrate my approach.
The third day I positioned myself between Master Control and the room where the VTR lived. I could see both the person initiating the commercial playbacks and the machine. I watched closely as each playback was initiated.
This time, the sequences played fine until the last one with the machine hiccupping at the precise moment a buzzer sounded indicating the arrival of the UPS man at the loading dock. Buzzer sound = machine dump. But why?
As it turned out, the station personnel ran an unshielded twisted pair remote start cable from Master Control to the machine and placed it in the same cable tray with an equally unshielded twisted pair from a push button on the shipping dock to a high-intensity buzzer. The crosstalk between the two cables was enough to reach the machine’s logic circuits and trip the cancel function that shared a multi-function IC with the remote start function.
Had I not gone outside of the VTR room and observed the machine dump when the buzzer sounded, I’d probably still be on site.
About the Author
John Loughmiller is the owner of Technical Support Group, a consulting company in Kentucky. Previously, he worked as an engineer for Ampex Corp., Sony Corp. and TV One Inc. A suspected Luddite and know curmudgeon, his spare time is spent terrifying hapless students who want to learn how to fly airplanes.Get Published in Sherlock Ohms!
Have you applied your deductive reasoning and technical prowess to troubleshoot and solve an engineering mystery that even the fictional Sherlock would find most perplexing? Tell us about it in 600 words and we’ll pay you $100 if we publish your case.
Email Karen Field at: email@example.com