# Saturating a NPN transistor

Discussion in 'General Electronics Chat' started by zoom, Jan 24, 2015.

1. ### zoom Thread Starter New Member

Jul 11, 2014
29
0
I'm a bit confused about the saturation of a transistor. Let's say I have a pn2222a NPN transistor. Its Hfe value is 100. My base voltage is 5 Volts and Vbe voltage drop is 0.6 volts. I want to draw 20mA from collector at most.

So far so good. Now I make the base resistor calculation simply by the following formula;

(5-0.6)/(20mA/100) = 22k ohm

Now, this is the maximum base resistor for putting 20mA to collector. Anything under this would put more current. Yet my load would draw the necessary amount of current given that it has a current limiting resistor. (Let's say an LED with a resistor)

The thing confusing me is, how I can determine whether my transistor is saturated.

Last edited: Jan 25, 2015
2. ### MrChips Moderator

Oct 2, 2009
12,212
3,276
When a BJT is in saturation mode it is turned on hard.
There will be very little voltage drop from collector to emitter, about 0.3V.

3. ### wayneh Expert

Sep 9, 2010
11,881
2,837
Setting the base current to 10% of the collector current ensures it is turned on hard. This is a very conservative rule-of-thumb and works for almost every NPN. It's often overkill, but if you want to reduce the base current, you have to test how low you can go without losing collector current. This will vary from transistor to transistor.

4. ### zoom Thread Starter New Member

Jul 11, 2014
29
0

Desired collector current : 0.02 Amps
10% of it is 0.002 Amps
(5-0.6)/0.002 = 2200 Ohm

So, my base resistor should be in the range of 2200 - 22000 Ohm to be able to ensure that it's hard on and supply my desired current at least. (For sure it will supply more than 0.02Amps as I get closer to 2200 Ohm, but since I have a current limiting resistor that's fine.)
Am I correct ?

Last edited: Jan 25, 2015
5. ### tracecom AAC Fanatic!

Apr 16, 2010
3,867
1,392
I think your calculations are probably correct, but I haven't actually checked the math. For me, the key to using a BJT as a switch is to think of it as a switch, i.e., either on or off. So, I just use a base resistor that allows a base current that is about 10% of the rated collector current of the transistor regardless of the actual amount of collector current the circuit needs, and depend upon the load circuitry to not exceed the rated collector current. I would just use a 1k base resistor, and be happy, but read my signature.

6. ### MrChips Moderator

Oct 2, 2009
12,212
3,276
That is accepted procedure. A base resistor of 1k to 4.7kΩ is typically used for a BJT for collector current of 20mA.

7. ### zoom Thread Starter New Member

Jul 11, 2014
29
0
I'm a bit confused again.
This was the initial calculation for Ic = 0.02A
(5-0.6)/(20mA/100) = 22k ohm

But here my Ic is 0.02Amps and my Ib = 0.0002Amps. Then my base current is 1% of my collector current.

So, calculating the base resistor is not sufficient alone. To be able to hard ON the transistor I need to supply more than the calculated base curernt, if I got it right.

If I do so and take the 10% of collector current, Rb has to be 2200 Ohm. Then may upper limit for base resistor is 2200 Ohm for saturating the transistor but this time my collector current will be more than 20mA. Looks like there is something wrong with me understanding.

I want to ask that is it enough to do the following calculation for base resistor (Rb) with desired collector current (Ic)

Rb = (Vcc - Vbe)/(Ic/Hfe)

Where Vcc is the base voltage and Hfe is the transistor gain.

8. ### MrChips Moderator

Oct 2, 2009
12,212
3,276
Firstly, the Hfe is never the same from transistor to transistor, even with the same part number.
While the quoted Hfe may be 100, the rule of thumb is to use a value of 10 for saturation mode.

The base current and Hfe alone do not determine the maximum collector current. It sets the capability of the BJT as a switch.
The maximum collector current will be limited by the collector series resistor you choose. So for example, if the supply voltage is 5V, selecting a collector resistor of 250Ω guarantees that the collector current will not exceed 20mA.

9. ### zoom Thread Starter New Member

Jul 11, 2014
29
0
That makes a bit more sense. However, you last calculation makes the assumption of Hfe = 1 and Vbe = 0 Volts , right ?
But your Hfe is most probably not 1, let's assume it is 50 for your operation conditions, and then you would get 1 Amps collector current with 250 Ohm base resistor, don't you.

10. ### MrChips Moderator

Oct 2, 2009
12,212
3,276
My 20mA calculation has nothing to do with Hbe and Vbe.

My calculation is for the collector resistor Rc.

Rc = Vcc/Ic = 5V/20mA = 250Ω

The base resistor Rb is

Rb = (Vcc - Vbe)/Ic/10 = (5V - 0.7V)/2mA = 2.15kΩ

Hence 2.2kΩ is a good choice.

Apr 5, 2008
15,517
2,298
cmartinez likes this.
12. ### zoom Thread Starter New Member

Jul 11, 2014
29
0
Oh sorry, I missed that part
Now it's clear, thank you

13. ### zoom Thread Starter New Member

Jul 11, 2014
29
0

Here one point drew my attention. I was not considering Vce voltage drop when calculating the collector resistor. So, for the datasheet of the transistor I have, which is PN2222A there are two max. values for Vce(sat)

Ic = 150mA , Ib = 15mA | Vce(sat) = 0.3 V
Ic = 500mA , Ib = 50mA | Vce(sat) = 1 V

Let's assume I have Ic = 200mA, then which value I have to pick up ? something closer to 150mA case, like 0.4 Volts. Also that's the maximum value.
Or should I ignore it to guarantee absolute 200mA with given Collector voltage supply ?

Apr 5, 2008
15,517
2,298
Hello,

Here I found a datasheet with a saturation curve:

On the corner there is stated they used Ic = 10 X Ib.

Bertus

File size:
33.4 KB
Views:
317
• ###### 2N2222_ME.pdf
File size:
308.1 KB
Views:
13
zoom and MrChips like this.
15. ### MrChips Moderator

Oct 2, 2009
12,212
3,276
So look at the curves bertus has provided.

At Ic = 200mA

Vbe = 0.9V
Vce = 0.2V
Ib = Ic/10 = 20mA

zoom likes this.
16. ### crutschow Expert

Mar 14, 2008
12,497
3,058
If it hasn't been mentioned, the Hfe is for use of the transistor in the linear region not as a switch.
If you look at the spec, the Hfe is specified with some significant Vce voltage, not Vce sat.
Look for the Vce(sat) value in the spec. and it will generally be given with a base current of 1/10 the collector current, as we all recommend.

zoom likes this.
17. ### zoom Thread Starter New Member

Jul 11, 2014
29
0
One more question on this;

200mA, would that lead too much heat on a small PN2222A ?
It stands up to 1Amps according to datasheet.

18. ### tracecom AAC Fanatic!

Apr 16, 2010
3,867
1,392
Good design practice is to not exceed 50% of the rating, so 200 mA should be fine.

19. ### crutschow Expert

Mar 14, 2008
12,497
3,058
As long as it's used as a saturated switch, it should be fine. For a high Vce= 0.5V saturation voltage the dissipation would still be only 100mW at 200mA current which the transistor package can readily dissipate, even at well above room ambient.