BJT biasing q - resistorfrom the base to +9v instead of to the collector

[midwayfair]

On a mobile, sorry about the lack of a scheme.

I'm talking about a situation where the base's positive bias is referenced to the 9v rail instead of to the collector or a 4.5v reference voltage.

This works just fine for the most part, because, well, I've done it a lot and haven't really encountered any problems. My question is:what does biasing in such a way actually, well, do? Clearly it's not the 'right' way because very few circuits do it that way.

[PRR]

The bias resistor "looks at" the transistor _collector_. This is negative feedback.

[R.G.]

Paul, I think he meant that he took a resistor to the positive rail, not the collector.

@MWF: if that's the case, what you did used to be called "fixed bias" back in the early Japanese-radio era of the early 1960s. It works, but it's tricky to get just right if you need a big signal swing. It tends to need a big emitter resistor if it's going to be stable.

If you think about it, the current through that one resistor is just the power rail voltage minus the base voltage divided by the resistor. That pulls the base up and turns on the transistor. If the emitter is grounded, the base goes to about one diode-drop - and stays there. The base current is fixed, and never changes, but the collector current is just the base current times the (variable!) hfe. The transistor biases somewhere that depends on the resistor and the value of the transistor's hfe. The result is anything but predictable, and varies with temperature and phase of the moon.

A big emitter resistor helps, by providing voltage feedback to the base, but it's not as effective as the four-resistor stabilized bias circuit, for some math and other technical reasons.

As a result, you tend to have to match the base resistor to the situation, including the individual transistor, to get much signal out.

It's usable, and there were zillions of Japanese pocket radios sold that way. But there are better and more predictable ways, so it's hardly ever used any more.

[midwayfair]

It works, but it's tricky to get just right if you need a big signal swing.

Huh, that probably explains why it freaks out a little when I feed it a boosted signal ...

Thanks for the explanation. I originally used it because I didn't hear a difference when using a guitar signal and it tended to be really convenient on a layout and then kept using it because it worked, but it sounds like I shouldn't use it anymore if I can help it. The circuits did have a pretty big emitter resistor, bigger than the collector in one case, and the base resistors were really big, too (2.2M was the smallest).