DC current Gain

If my collector current VCC = 3.3V & Ic = 31.82uA then how to calculate minimum base current is required put transistor in saturation mode? Depending on it I will calculate base series resistor

As per my understanding hfe varies drastically in saturation region. Which hfe value to put into Ic = hfeIb formula?

Transistor Circuit

  • \$\begingroup\$ In saturation mode, \$h_{FE}\$ is completely useless as a parameter. It's better to just use \$V_{CE,sat}\$ and do your computation from there. It's generally bad practice to make circuits that depend on the precise value of \$h_{FE}\$ anyway as it's so poorly controlled. And as the datasheet shows, even in normal operation it varies a lot for this particular transistor. \$\endgroup\$ – Hearth Oct 30 '18 at 13:02
  • \$\begingroup\$ @Felthry But even after considering this parameter if only IC is known (e.g. 31.82uA); how to calculate Ib? Provided Datasheet values for VCE,sat are as follows: for (IC = 150 mAdc, IB = 15 mAdc) -> VCE sat = 0.4 Vdc. How to calculate base current on above information? \$\endgroup\$ – Akky Oct 30 '18 at 13:04
  • \$\begingroup\$ You can't. \$I_B\$ can be a wide range of values, pretty much anything you want as long as it's high enough to keep the transistor in saturation. \$\endgroup\$ – Hearth Oct 30 '18 at 13:07
  • \$\begingroup\$ @Felthry I would like to select resistor such that minimum current is drawn from 12V supply. I would like to know how to estimate minimum ib value so as to drive base? \$\endgroup\$ – Akky Oct 30 '18 at 13:10

The Vce(sat) curves on a typical datasheet for that part don't go under 1mA (see figures 2 and 6). Gain drops at lower current, especially on some older parts, but it's probably >> 20 at 31uA Ic. The MMBT4401, for example, lists minimum hFE as 20 at 100uA, but typically the gain does not drop off as rapidly as that would suggest (see figure 1).

You are probably safe using Ic/Ib = 10 (as in figure 2), but Ic/Ib = 5 would be even safer, since the datasheet is silent on lower collector currents.

The minimum is a bad number to use, it will vary from part to part. You want to use a safe number that will saturate any transistor you pull off any given reel of parts and will saturate it at the maximum Ic and over the entire operating temperature range.

  • 1
    \$\begingroup\$ +1; exactly what I was about to say re: not using the minimum. \$\endgroup\$ – Hearth Oct 30 '18 at 13:11

I'd be more inclined to begin by looking at this diagram in the data sheet: -

enter image description here

It only takes you down to a collector current of 1 mA but you can "sort of" interpolate to lower collector currents. At 1 mA collector current and 0.2 volts across collector and emitter, the base current is going to be about 7.5 uA but you are in a shady area where things are not well defined and I would expect that if the collector current was 0.1 mA the base current would be more like 10 uA. In other words, the hFE drops off as you start to reduce the collector current.

It's also very hit and miss trying to define the base current for a given Ic and saturation voltage. For the 1 mA example, I'd be considering a base current of more like 20 or 30 uA and expecting a saturation voltage of about 50 mV. The reason is because around the knee of the graphs you have little control and temperature effects can dramatically alter saturation voltage.

If you are looking for a defined saturation voltage then I think you should use other methods.


I like to start by figuring out how much collector current it would take to bring me to a saturated value of \$V_{CE}\$, and once there look at the data sheet to try to figure out how much base current I need to do it.

If you can't find nifty figures in the data sheet, like Andy points you to, An "OK" rule of thumb is to assume that \$\beta\$ is no larger than 10 in a transistor driven to saturation.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.