My old man, who was a electronics technician for most of his professional life, used to tell me that tube testers were tube salemen, because they were good at telling you that a perfectly good tube needed to be replaced.

The problem with most tube testers, as I've mentioned elsewhere, is that the commercially sold units (the ones you and I are likely to encounter) are designed to quickly test a huge variety of tubes which have a variety of characteristics; testers are a collection of engineering compromises used to accomplish this task. They had to offer enough options to be able to test a wide variety of tubes, yet be simple enough to be understood and, more imporantly, be used quickly. Professional service techs could not afford to spend a lot of time testing tubes. It was usually faster and just as conclusive to swap out tubes and see if the problem was solved.

The biggest problem with tube testers in this area was the Merit scale. Depending on the manufacturer, the meter face had a big band divided in half, with a green band on one side and a red band on the other, and the labels Good and Bad, or Good and Replace, or something similar. And there's usually a little yellow zone in the middle marked Weak, or just left to the user's discretion.

Hickok calls this the English scale, which seems reasonable to me. There's an adjustment on Hickok machines called the English dial; on other makes it'll be called other things. But they are all shunts—a method of adjusting the current level that the meter sees. This is how the same machine can test a power rectifier with a heavy current and a voltage amplifier with a relative trickle, and still use the same Good/Bad scale for both.

The shunt (or English) adjustment amount is provided by the tube-tester manufacturer, and it is found in the settings chart for each tube type. The way it was typically determined was by the manufacturer purchasing a representative sample of each type of tube and determining the actual performance of each; they then subjectively decided where that peformance would lie on the Good/Bad scale.

This is where the dragons live.

• Where do you draw the line between what's Good and what's Bad and what's middling? Few tubes that are "bad" are just barely bad or just barely good.

• a tube that tests Weak or Bad may be perfectly good for the application it is in; e.g. a "weak" audio output tube may still provide more power than needed to drive the speaker at realistic volume levels. Likewise, a tube that tests Good may still not perform adequately for its task, either because the engineer chose the wrong tube for application or maybe the tube-tester manufacturer determined the wrong setting for his English scale.

• the "English" shunt adjustment can be misused so that it unfairly biases in one direction or another, but most commonly toward Weak or Bad. Why? Because a customer who sees his tubes test "Weak" or "Bad" are likely to purchase a new one. The machine says it is "Bad," doesn't it? The machine is a machine—it is impartial, isn't it?

I don't mean to completely discount the merit scale. If you trust the chart, it is a quick and easy way to determine the performance of a tube. But the better testers have other scales with actual numbers (usually micro-mhos) or hash marks or other indicators that let you see how a tube performs on something other than the merit scale.

For more information on tube testers in general and manufacturers and models in particular, see Alan Douglas's book, Tube Testers and Classic Electronic Test Gear.

For more information on tube testers in general, I recommend Getting the Most Out of Vacuum Tubes by Robert B. Tomer.