电视发射器麻烦

Article By : Robert Yankowitz

高功率的电视发射器并不是一件令人兴奋的事情，只是一个大金属盒（6英尺高，12英尺宽，4英尺深），里面装有电子设备，但大部分在外面都毫无功能。当前的型号是固态的，充满了数百或数千个高功率晶体管，以约50伏的速度运行。它们成为2010年代中期最常见的类型。

The previous generation of UHF transmitters (starting at 470 MHz, channel 14), were just as plain to look at, but could be much more interesting. They used vacuum tubes known as inductive output tubes (IOTs)—successors of klystrons. The tube is about 32 inches tall and weighs 64 pounds. It sits in a “circuit assembly” containing tuned input and output cavities and weighs almost 350 pounds. Each IOT amplifier, of which there were usually two, has a power supply generating 35,000 volts DC and draws up to two amps, producing around 25,000W of average RF output power. Yes, the efficiency is low, but that’s a result of a number of factors, including the need to operate in class AB mode.

是什么使它变得有趣？一方面，物联网上的35公斤，这确实将使发射器工程师专注。正如同事曾经警告过我的那样：“鲍勃（Bob）要小心，鲍勃（Bob）可能会杀死您很长一段时间。”发射器具有许多电子和物理的互锁，以防止这种情况发生。

对于另一个人来说，尽管物联网管非常坚固，并且可以在适当的护理和喂养下连续运行超过60,000小时（大约七年），但如果不适当保护，它也可能会瞬间被破坏。实际上，物联网放大器中的电路比例很高，只是为了保护管子。

One of the ways it does this is to monitor a large number of parameters and remove the high-voltage from the tube in an instant, in case any parameters are out of range. It does this by firing a crowbar, usually a thyratron tube, similar to an SCR, shorting out the high voltage which bypasses the high voltage supply’s stored energy away from the tube until the power supply’s circuit breaker opens.

A crowbar event is an exciting thing to be around. By exciting, I mean frightening. Relays click, contactors bang, really big circuit breakers trip, and sometimes the lights in the building flicker. Then the transmitter resets itself and starts up again. If a real fault remains, as opposed to a transient overload, this happens up to three times in a ten-second period before the transmitter gives up and stays off.

这就是这个故事开始的地方。我们的发射器中的一个放大器开始断断续续地扭曲。有时，每天会发生两到三次，而另一次只会每隔几天一次。LED的许多故障均未显示出问题根源的任何线索。

撬棍旅行的间歇性和随机性具有机械问题的感觉，所有冷却风扇，高压鼓风机和液体冷却泵都会振动一切。因此，我和我的助手从发射机中击败了活着的日光，竭尽所能试图激发它的一切。这非常愉快。不幸的是，这无济于事，因为每次我们认为我们找到合适的命中位置时，事实证明这是一个错误的警报。

As I mentioned, an awful lot of things have to be just right for the transmitter to be happy and allow the amplifier to be energized. There are circuits continuously monitoring each of these things. During operation, if any of these parameters falls out of range, even for a microsecond, the amplifier will shut down and recycle. Each of these circuits sends a TTL-level good/bad signal to the amplifier’s control board, where they are summed through an AND gate. If the output of the AND gate is high, the amplifier is allowed to operate, and a front-panel LED called “Amplifier Ready” is lit. Should any of the inputs drop out, the LED and amplifier turn off. Our problem was that one of these signals was dropping out momentarily, and so fast we couldn’t even see the LED flicker.

As it happened, we had a logic analyzer, not a common piece of equipment for a television station. In order to catch the fault in action, we hooked it up to monitor all of the AND gate’s inputs and set it to trigger when the gate’s output went low. Then we could discover which protection circuit had fired. Naturally the transmitter was wise to what we were up to, so it behaved itself for the next three days. Finally, we caught it in the act, and found the “Magnet Current in Range” signal was the culprit.

我的助手开始将布线追溯到采样电路。通向控制板的电线是通过机柜底部的弹簧端子带连接的。他用手指从电线上数来，当他触摸正确的那条电线时，它向上抬起。事实证明，工厂的汇编器并没有一路打开弹簧触点，而电线只是触摸了触点，仅通过重力和电线的刚度固定在那里。随着时间的流逝，电气接触的质量降低，直到最终开始失败。它坐在那里，玩负鼠的十五年，然后才终于醒来。

So, in the end it turned out we were right – it was mechanical. We just didn’t beat up the right part of the transmitter.

This article was originally published onEDN.

罗伯特·扬科维茨（Robert Yankowitz）最近在马萨诸塞州波士顿的一家电视台退休，在马萨诸塞州的一家电视台工作了23年。在此之前，他在罗德岛普罗维登斯的一个车站工作了15年。