Hi there!!

I have a question about hfe of transistors. Lets take for example a low power npn transistor, 2N2222A. On 2nd page, there is a section of table where various conditions are stated and various hfe values are presented. On the right, there is also a max and min values of hfe.

I know the single formula (\(\beta\) = \(\frac{Ic}{Ib}\)) that is used to property use the transistor in combination with ohm's law. Can someone please explain the meaning behind the minimum hfe and maximum as well as those test conditions and how they can be useful.

Thanks much!!!

 

It means it's a random variable and the highs and lows tell you something about the range of parts' values you can expect (either due to process variations, inspection culling, or both). hfe is a notoriously poorly-controlled quality parameter.

 

Because it varies so much, circuits are usually designed so that they will work OK with a range of hfe.
Have a look here for an explanation.
The whole chapter is worth a read.
http://www.allaboutcircuits.com/vol_3/chpt_4/10.html

 

IIRC, Wikipedia's entry on transistor biasing has a lot about this too.

 

Way back when I took transistor theory our book was "Wave generation and shaping" by Leonard Strauss, and our professor (who by a curious coincidence was also named Leonard Strauss) taught us to just assume the beta of a transistor was 30 minimum, as that was the criteria for an IC wafer acceptance.

 

The values in the datasheet set upper and lower limits, transistors are notoriously fickle. Depending on how they biased, the exact value can vary even on the same transistor.

There are many ways to overcome this, mostly through various biasing techniques that set the gain of the circuit, which will always be less than the gain of the transistor. If not, then the transistors gain is the limiting factor.

 

And perhaps one of us should note that while:

\( \beta = \fra {I_B}{I_C}\)

is a gross oversimplification it is a quite useful simplification when used correctly.

 

And perhaps one of us should note that while:

\( \beta = \fra {I_B}{I_C}\)

is a gross oversimplification it is a quite useful simplification when used correctly.

I guess you meant to write

\( \beta = \frac {I_C}{I_B}\)

 

And perhaps one of us should note that while:

\( \beta = \fra {I_B}{I_C}\)

is a gross oversimplification it is a quite useful simplification when used correctly.

\( \beta = \fra {I_C}{I_B}\)
is the exact definition. No oversimplification. As others have mentioned, beta varies from one device to the next, and with current and voltage. But the ratio Ic/Ib is always beta.

 

Well then, when I insert my PNP into a multi meter, I get a single value, not mins or max. Is that the value that I should use in my calculations? I have got to read that chapter... Thanks you all!

 

Well then, when I insert my PNP into a multi meter, I get a single value, not mins or max. Is that the value that I should use in my calculations?

That's true but if you were designing something that was going to be produced on a mass scale you'd want to be sure your design worked for the range of β that might be expected for that device type - i.e. from minimum to maximum.

 

Well then, when I insert my PNP into a multi meter, I get a single value, not mins or max. Is that the value that I should use in my calculations? I have got to read that chapter... Thanks you all!

You should use the minimum in the datasheet.
This would be the worst case.

 

Your meter just gives an hFE for a fixed low current value; perhaps a milliampere or few.

If you are using a transistor as a saturated switch, then use:
Ib = Ic/10
where Ic is the desired collector current.
If you try to use the minimum hFE in the datasheet, the transistor may not be in saturation, with resulting power dissipation in the transistor due to a high Vce.