Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Common emitter configuration of BJT

  1. Sep 9, 2012 #1
    in common emitter configuration of BJT , why do we have linear relationship between output current (collector current ) & output voltage ( collector emitter voltage) ? Has it anything to do with early effect ? If yes then why don't we have linear relationship in case of common base configuration ?
  2. jcsd
  3. Sep 9, 2012 #2
    There is no relation between output current and output voltage. It only has linear relation IF you use a linear load at the collector. Typical example is a resistor and if you have one with NO voltage coef. Then the voltage is V=IR. You put a non linear load, the voltage is absolutely non linear. Case in point, I can make the output voltage logarithmic by using a forward biased diode as the collector load and make it a logarithmic amplifier. People in Guitar amp design use carbon composite resistors that has voltage coef to introduce some even harmonics.
  4. Sep 10, 2012 #3

    Yes, the variation of Ic vs. Vce in a common emitter configuration is caused by the Early effect. Notice that Early is capitalized, because it was named after James Early. In a common base configuration, one sees flat lines on the Ic vs. Vcb curve. Notice that the x-axis of the CE is Vce and on the CB is Vcb. The Early effect acts like a shunt resistor across the current generator of the collector. For a CB, the Early resistance is Re, and for the CE configuration, the Early resistance is Re/(β+1). This can be verified by a study of transistor models. Since the Early resistance looks smaller in a CE configuration, the Ic lines start to show a slope.

  5. Sep 10, 2012 #4
    I don't know whether OP meant linear relation of Ic vs Vce when using input signal or output resistance as you describe.

    Do you have any article of early resistance?
  6. Sep 10, 2012 #5

    Ideally, the Ic should not change with Vce in the active region of a BJT. In the active region, the BJT is a voltage controlled current source (VCCC).

    The OP asked why the Early effect does not show up as much in the CB configuration. As stated in my previous post, it is because the shunt resistance is "betatized" downward in a CE configuration.

    The Early effect causes more Ic to exist than would be present without the Early effect. That is equivalent to a shunt resistor across the almost infinite resistance of the collector current generator.


    See the "Small-signal model" paragraph in this link below

    Last edited: Sep 10, 2012
  7. Sep 10, 2012 #6
    I know about early effect, ro and Va, I just have not heard of Early Resistance. Do you have formulas on this Early Resistance?
  8. Sep 10, 2012 #7

    You can get the resistance from the slope of the Ic vs. Vce curve.

  9. Sep 10, 2012 #8
    re is conventionally is emitter resistance and is 1/gm!! Unless you call output resistance re. What you are talking is output resistance using early voltage to calculate ro≈Va/Ic.

    I am not sure what the OP was asking, I take that he was asking about change of Ic with change of Vce with input voltage like any ordinary amplifier. What you are talking is mainly change of Ic with change of Vce with Ib or Vbe keeping constant. These are two totally different thing depend on what the OP meant.
  10. Sep 10, 2012 #9

    I said Re, not re. I was using the "e" subscript to represent "Early".

    Nothing of the sort. I am assuming the output resistance "ro" is very high in a CB configuration.

    Yes, that is what the second link I posted says it is.

    He was asking why the Ic current lines have a slope in the CE configuration, and are almost horizontal in the CB configuration.

    Not really. If Ic changes with Vce, the Early effect is in play. Otherwise not.

  11. Sep 10, 2012 #10
    Good evening Ratch,

    The slight difficulty with this approach is that Vce is not something you can force onto a transistor, unlike Ic (without lots of blue smoke that is).

    The mode of action of a CE circuit is that the transistor adopts a Vce compatible with the set Ic, not the other way around.

    This is why transistor manufacturers publish curves of the variation of beta with Ic, not curves of the variation of Ic with Vce.
  12. Sep 10, 2012 #11

    Of course you can. See this link. http://www.ecelab.com/vce-ic-curves.htm The Vce voltage varies from 5 to 15 volts while Ic hardly changes at all.

    In the curve shown in the above link, Vce changes much and Ic changes little. Even less if a CB configuration and Ic is plotted agains Vcb. They are mostly independent of each other unless the Early effect comes into play.

    Someone did not tell the manufacturer in the link of that stipulation.

  13. Sep 10, 2012 #12
    OP has to specify the condition. If you specified everything else remain constant and Ic change with Vce, that would be very obvious you are talking about early voltage. We all know that.

    Read back the OP, the Va statement is a separate statement. Particular if he was asking about output resistance like you suggested, CB should absolutely be a linear relation also, it's just has much higher output resistance, It is still linear. OP need to clarify this.
    Last edited: Sep 10, 2012
  14. Sep 10, 2012 #13

    Now is your chance to clarify your question.

  15. Sep 10, 2012 #14
    Yep, because it is day and night difference!!!
  16. Sep 11, 2012 #15
    sorry for the late reply
    i was studying from microelectronics by donald neaman
    According to the book, in CE configuration for a constant Vbe , if Vce increases then Vcb also increases thus making more reverse bias. NOW the effective base width will decrease and gradient changes will come into the picture ultimately increasing the Ic. That was the EARLY EFFECT. So i think this whole thing is applicable to CB config also because in this case increase in Vcb for constant Vbe will increase the Ic for the same reasons. So, i want to know that why in books they always show horizontal lines for CB config. ?
  17. Sep 11, 2012 #16


    User Avatar
    Science Advisor

    The Early effect is the increase in base transport factor (and hence increase in both [itex]\alpha = I_C/I_E[/itex] and [itex]\beta = I_C/I_B[/itex]) due to the base narrowing as [itex]V_{CE}[/itex] increases.

    This effect is exactly the same for both the CE and the CB configuration. However, since only tiny changes are required in [itex]\alpha[/itex] to give a significant change in [itex]\beta[/itex], the effect is simply not very noticeable in the CB case (since you are plotting Ic for a given Ie in this case, rather than Ic for a given Ib as in the CE case).

    Take a numerical example. Say that the base transport factor is 0.995 and the emitter injection efficiency is also 0.995 then the overall [itex]\alpha[/itex] will be approximately their product, giving [itex]\alpha \simeq 0.990[/itex] and [itex]\beta = \frac{\alpha}{1-\alpha} \simeq 99[/itex].

    Now imagine that we increase Vce so that the base narrows and the base transport factor increases by just 0.1% to 0.996. The overall [itex]\alpha[/itex] will now increase to about [itex]\alpha \simeq 0.991[/itex], however [itex]\beta = \frac{\alpha}{1-\alpha}[/itex] increases to approximately 110, an 11% increase!
    Last edited: Sep 11, 2012
  18. Sep 11, 2012 #17
  19. Sep 11, 2012 #18
    erece & uart,

    I believe I offered an explanation of why the Ic current lines are horizontal in a CB configuration and have a slope in the CE configuration. See post #3 of this thread. The Early effect effectively puts a shunt resistor across the collector current generator and reduces the transistor's performance. The slope of Ic shows the resistance of the collector circuit. I said that the more vertical slope of the CE configuration was because the transistor see its output resistance as rc/(β+1), which is smaller than the CB impedance of just rc. The lower resistance is why the lines show a more vertical slope. Any good book on transistor circuits will show that CE and CB have a different output impedance.

    I don't think that the smaller value of α compared to β makes any difference. They both have a one to one correspondance to each other. The output resistance of CE can be expressed as also as rc*(1-α), which is identical to the value above.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook