# Output characteristics curves do not merge in LTspice

I wanted to plot the output characteristics curves and Early effect of a transistor in CB arrangement. In LTspice I made a DC sweep along with parametric sweep as follows where I sweep the Vcb from -24V to 5V and for each fixed emitter current from 0 to 100mA with 10mA increments:

In the plots below -I(V1) is the collector current and -V(C) is the collector-base voltage.

Even at around -24V these curves do not merge. Is the simulation setup wrong?

• common-base is very stiff output, as you see. The only effect will be the slight variation of BETA. Edit the model param for BETA, from 100? to 5, and rerun. May 10 '19 at 15:41
• Your BJT model is too simplistic I reckon. May 10 '19 at 17:17
• Think about the polarity of V1 and the collector-base junction May 10 '19 at 22:26

You should use the DC sweep capability in Spice. Set up two sources, one a current (for the base recombination current) and the other a voltage (for $$\V_\text{CE}\$$.) As follows:

Once you have the above circuit in place, you can use the .DC card to perform the sweeps. Note that I first sweep $$\V_\text{CE}\$$ over some range I want and then I sweep the base current.

These sweeps can be linear, octave, decade, etc. You can create custom sweeps using the .STEP card and arranging to provide a list of values for some parameter, too. It's your call. Standard sweeps are convenient. But custom ones are achievable with a little more work. In the above .DC sweep I set up, I let Spice do the sweeps.

Here's the result. And you can see the Early Effect, quite easily:

• Can you please do the same thing with Common Base configuration? Im curious what you will obtain. There I get those results. Yours is not CB. May 10 '19 at 21:50
• @user16307 I thought you just wanted to see the Early Effect. The arrangement I used is designed to explore it, as it usually is explored. You want to use a fixed base current and sweep the VCE. I don't see the meaning of common base in this context.
– jonk
May 10 '19 at 21:51
• I asked because in common base the curves do not merge. Please do that so we can see where the problem is. And if you use 2N3906 would be great, May 10 '19 at 21:52
• @user16307 I'm suggesting the idea that the Early Effect is inside the BJT. You either set up an experiment to expose it to view, or you do not. There is no meaning to CB, CE, CC here. Those are external circuit modalities. They have NO meaning at all when trying to examine the Early Effect in a BJT. For that, you design the right experiment and perform it. That's all. I now suspect your question is misguided by some pre-concieved notions you have that are getting in the way. Perhaps you are really asking if the Early Effect is important in certain circuit arrangements?? (Or it's me.)
– jonk
May 10 '19 at 21:58
• @user16307 Ah. I see. I'd need to read the context. Not just look at the labeled pictures. The Early Effect has a model parameter, VA, which provides it to Spice. This parameter is NOT broken down into CB, CE, or CC modes of operation and the BJT frankly has no idea which mode it is in, anyway. It's just a BJT. So all I can say at this time is that I think you are asking about the impact of the Early Effect, when it is applied in various circumstances. (For example, the cascode arrangement is designed to mitigate it dramatically and it does so very well, so almost invisible then.)
– jonk
May 10 '19 at 22:08