Magnifying glass to the rescue

This is one of my favorite success stories from the late 1970s.

I was working at KGGM TV as a studio engineer at the time. When you work at a TV station, election day is a very big deal, and 1978 was no exception. Every piece of equipment that the station has gets pressed into service, and then some. (I brought in a couple of personal power supplies to operate a few more gadgets.) Everyone is expected to be "on duty" on election night. I'd drawn the job of being the studio engineer that night. (The others were better able to lift things, so were better suited for the remotes.) Two of our four studio cameras were at remote sites, so I was down to just two cameras.

The big studio cameras have a lot of controls, and these ones were about 10 years old at the time, and so didn't really have that much in the way of automatic controls. The big studio cameras are the ones you sometimes see at professional sporting events, and in the late 1970s, new ones were going for around $95,000 apeice. (BTW, the "camera operator" only has responsibility for where the camera is pointing, how far in or out it is zoomed, and focus. Everything else about the picture is up to the engineer.) A good engineer always tries to keep one step ahead of the director, looking at the camera the director is going to go to next, to assure that it has the best possible picture.

A little after 7:00 PM, I punched up Camera 1, and noticed that it needed a little more red, so I reached over to the controls and turned up the red. A couple of minutes later, the director was about to go to Camera 1 again, and I noted that it needed a bit more red, so I turned it up again. The next time I punched up Camera 1, it was, incredibly, still in need of more red, but there wasn't any more to give it, so I said to the director (through the headset intercom) "Camera 1 just died". That was not a fun report to have to give, but it was the truth. The chief engineer happened to be at the studio right then, and since I couldn't leave my post, he went out onto the "sound stage" and fiddled with it for about 20 minutes, to no avail. We rolled Camera 1 over into the corner and "flew" with just one camera.

A couple of days later, after the "dust settled", the engineer who normally worked on cameras started working on Camera 1. After about two weeks, he gave up. Then the guy who normally worked on the transmitters spent about a week on it, with no success. Next, the chief engineer took a shot at it for about 4 days, and at the end of that time, Camera 1 still wasn't working. Finally, the other engineer who worked "sign-off" shift (from about 4:00 PM to when the station went off the air for the night, usually between midnight and 1:00 AM) took the camera apart not once, but twice, and couldn't find the problem.

About mid December, I was working on a Saturday night, with nothing much going on, but by law I had to be there. I decided that since everyone else had tried and failed to fix the camera, I had little to lose by trying. I did have two clues: First was that when you first turned the camera on, it would work for a couple of minutes. Second, through all of the testing, it had been determined that the problem was actually out in the camera "head", and not in the roughly three foot high rack of "support electronics" that lived in the control room.

So, I gathered up all of the "extension boards" (which would get the various circuit boards sticking out of the rack so you would work on them one at a time), a few cans of "freeze mist" (used to cool components to around 0° F), and a heat gun, drug Camera 1 out into the middle of the news studio, turned on the lights (to the camera head up) and started working on it. I proceded to put the cards, one by one, on their extension board, turn on the camera, wait for it to fail, then freeze the entire board and see if the camera "came back". When I finally got to the "red channel driver", I was rewarded with a (relatively) nice picture when I froze the board, and when I heated it up with the heat gun, the nice picture wasn't so nice.

So, the next step was to freeze only about half the board, to figure out which half the problem was in. Once I had isolated the problem to half the board, I then halved that, and got it down to just a quarter of the board. When I'd gotten it down to being about a square inch, I changed my approach. I powered down the camera, took out the board, and very closely inspected the offending area using my magnifying glass. After a few minutes, I noticed a cracked solder joint on a 5 watt resistor.

I took the board into the shop, plugged in the soldering iron, and resoldered that joint. When I took the board back out to the camera, everything worked fine, though it took me the rest of that evening, as well as the whole of the next evening, to get all of the adjustments back where they should be (as almost all of them had been fiddled with during the preceding weeks).

The real moral of the story is to try to approach troubleshooting from a logical perspective, and be patient.

This page last updated: 01-Oct-2010

Copyright © 2010 by Clark Jones