BJT voltage divider biasing - how to decide on the value of collector current?

audioguru said:

Obviously, the collector resistor value depends on the maximum load current and the power supply voltage.
Using a 9V supply, a transistor with a 1k collector resistor can drive a 100k load very well.

Click to expand...

atferrari said:

Hola audioguru

Thanks for replying.

Yes but, how to decide on the Ic to go ahead for my calculation? Guessing and rule of thumb seems a poor criterion for a proper design. And even then, I did not see any clearly stated for the four cases I mention.

Click to expand...

Nigel Goodwin said:

Because they can't be any clearly stated values for such vague cases.

As AG has already said, they depend mostly on external factors.

'Guessing and rule of thumb' are probably as good as anything else - there's no real 'optimum value', unless you VERY accurately specify everything around it, there's more of a 'large range' where performance is perfectly satisfactory.

Often in these types of 'calculations' you have no option but to decide on a particular value to start with for one component, you can then calculate the other values based on that starting point.

Click to expand...

atferrari said:

I see Nigel. I got your concrete reply but other than feeling free to try whatever comes to my mind, I am not any closer to choose a reasonable value.

Click to expand...

gophert said:

atferrari , ignore Nigel, he has trouble with abstract thoughts.

I think it is safe to say you should design collector current a minimum of 10x to (preferably) 100x ABOVE the transistor cutoff current (see "collector cutoff current" on datasheet - usually listed next to "base cutoff current"). An ON Semi datasheet for 2N3904 lists 50nA as the cutoff current for both cutoffs so I would design the minimum current at 5uA.

But... like said, it USUALLY depends on other components and you design amplifiers around them. But if you are trying to design for micro power you have to examine questions like you ask here, do the same for each component or module and design around your most pressing constraints.

Also, you need to look into how:
- downstream modules load (draw current) from that collector, and
- other current leakages, like reverse currents, or dark currents that can dilute your collector signal with noise.

I would recommend you trying to track down the "Art of Electronics" and review the section on low current circuits.

Click to expand...

throbscottle said:

Maybe I'm thinking of the wrong thing here, but when I went to college we had one class involving printed charts of collector currents, photocopied from datasheets I guess, so you could put your ruler across at the opposite angle so it crossed in the middle of your desired line - the spot where you get the most linearity - and the I and V on the graph's axes give you the value of collector load you need.

But I'm sure you already know that! Anyway in case you didn't, there it is.

Doesn't answer the question though, does it? I suppose the thing to do is think of the load resistor in terms of it being the input source for the next stage, and go from there.

Click to expand...

atferrari said:

It seems that I am still good at forgetting things.

Click to expand...

JimB said:

Now that is an expression worth remembering.

JimB

Click to expand...