BJT transistor base current calculation

Other Hardware General Electronics
1

Hey

I have a BJT NPN transistor, and i would like to know how do i calculate the maximum current that transistor "lets through" (collector to emitter) given the current to the base of transistor. I have a 1K resistor wired to the base of transistor, 5V Vcc. I don't quite understand the electrical characteristics in the datasheet. Something about "Collector cut-off current", then there's also Base-Emitter On Voltage, and Base-Emitter Saturation Voltage. I have a vague idea of what that means, but could someone pls explain..

So what I really need to know is, how to calculate the max value of the series resistor that I can wire to the base of transistor, to still allow maximum current through collector-emitter?

I know the question is a bit vague, sorry about that, I can't explain it much better.. Thank you for any help you can give! Cheers, Val

2

I have found this page to be helpful in understanding transistors:

3

Hm, ok, thanks, that helped. :slight_smile:

I have sort of a followup question, hope this can go into the same thread. So if I know the dc current gain (designated as beta i think), could i potentially use a transistor to limit current that goes on through the emitter to the device? I mean, by placing a correctly valued resistor in series with the base of transistor, to control the saturation current?

For example, take the pic If I control the value of Rb to control the current the transistor lets through when saturated, could i omit the 62 Ohm resistor wired in series with the LED?

Maybe a stupid question, sorry :smiley: Cheers, Val

4

So what I really need to know is, how to calculate the max value of the series resistor that I can wire to the base of transistor, to still allow maximum current through collector-emitter?

We can make some appoximations & assumptions to simplify the calculations. You need to allow for tolerances anyway, so there is no exact answer. ( A lot of engineering has to do with knowing what's critical and what can be assumed, ignored, or approximated. :wink: )

We do need to know the approximate collector current. Or the "worst case" or maximum for your application. (You don't need the transistor's maximum current rating for the calculation... But, you need to make sure your application doesn't exceed the device's maximum rating.)

You shouldn't be thinking in "maximum" values... You need to find an approximate value that will work with the actual current in your circuit.

First, I'll assume this is a switching application, and that the emitter is grounded.

Next, I'll assume the transistor beta (current gain) is around 100. The data-sheet will give you the minimum value (and sometimes a "typical" value). But since this is a switching application we dont need to know the exact value. So, we can design a circuit that works with betas between 20 & 50 and we can be sure it will always work. (Assuming a low beta insures that we alway have enough base current to saturate the transistor.)

The base-emitter voltage is around 1V with the transistor turned-on. The exact voltage (usually) isn't important, since (in most applications) most of the voltage is dropped across the resistor... A rather large-percentage change in B-E voltage won't change the voltage across the resistor that much. In fact, you can assume zero B-E voltage, and most of the time it will work, since we are assuming a low beta and providing plenty of base current.

Now, you can calculate a resistor value that gives you a base current that's about 1/50th (to 1/20th) of your collector current.

Example - If you need 1 Amp, and you have 5V into the resistor (4V across the resistor). Ohms' Law says you need a ~200 Ohm resistor.

5

could i potentially use a transistor to limit current that goes on through the emitter to the device? I mean, by placing a correctly valued resistor in series with the base of transistor, to control the saturation current?

Yes... You can use a resistor (or something else) to control the base current, and as long as you are not in saturation, the collector current is proportional to base current (multiplied by the beta). In that case, the transistor is operating "linearly", not as a switch. You can use someting to "linearly" dim an LED.

When you are in saturation the load device (and voltage) determines the current. If you reduce base current to the point where you are no longer in saturation, the transistor begins to limit the current.

When you are operating lineraly, the transistor will "see" voltage and current at the same time. It will dissipate power, and it will heat-up. So, you have to be aware of the transitor's power rating as well as it's current rating, and in some cases you need a heatsink. That is, you can burn-up a 1 Amp transistor with much less than 1A if you are dissipating power. With an LED, this is usually not a problem.

6

The problem with current gain is that its "poorly characterised" - which is gobbledygook for "varies a lot between devices". So any circuit design that relies on it having a given value is a weak design. Typically you design for the minimum value of beta in the datasheet (note that it varies with current too...).

Since the variation between devices can be as big a factor as 3 or 4 IIRC, this means your current limiting would also vary greatly.

7

MarkT:
The problem with current gain is that its "poorly characterised" - which is gobbledygook for "varies a lot between devices".

That's a good point! It might OK if you are dimming one LED for a hobby project, or if you are just experimenting. But if you are building a product on an assembly line, or if there are several transistor/LED dimmers in your project, every LED would have different brightness with the same "settings".

Typically, you'll design the circuit so that resistor values control everything. For example, if you study op-amp based amplifiers, you will see that the large amounts of negative feedback are used, and the resistor values determine gain. (Or you do it digitally, such as PWM dimming. :wink: )

8

A! Thank you for your exhaustive explanation DVDdoug! Great, my calculations seem to pan out.

Also great point MarkT! I was just thinking of possible applications, but you're quite correct, that would be a poor design, haven't really thought of that :slight_smile:

Thanks everyone for your help. Cheers, Val

9

For swotching applications, beta has no meaning: you should provide sufficient Ib to saturate the switch. Usually Ib is at least 1/10 that of Ic.

In your case, Ic is determined by that resistor (together with supply + the led's forward voltage drop).

So the calculationb is backwards: select your resistor to get the desired Ic and from there, get the required Ib and calculate the corresponding base resistor.