# Making DC current gain assumptions when solvin BJT transistor circuits

I spent an hour trying to figure out why my math was different than a simulation on a particular circuit. It turned out my DC current gain I chose, 100 as per the minimum value on the datasheet, was incorrect; It was approximately 200. It got me thinking, is there something I should be looking at that can help determine the actual DC current gain value prior to solving a transistor circuit beside the min and max values on a datasheet? Is it intuition I'm lacking?

• Ideally, you should design your circuits so that the $\beta$ of the transistor doesn't change the overall behavior very much. Other than that, a "typical" value on the datasheet, or the value from a vendor-supplied SPICE model is the only thing that can tell you what value might be more likely. Real samples are likely to vary quite a bit from part to part and over temperature, so trusting the SPICE model isn't safe either. Mar 4 '20 at 21:56
• Discrete transistor current gain varies greatly. However, inside a chip, $\beta$ variations transistor-to-transistor might be reasonably small, making them well-matched. Mar 4 '20 at 22:02
• Your experience is lacking the use of negative feedback to make use of high gain and resistor ratios to stabilize Q point and AC gain. your H bias or other will suffer from non-linearity and offset for max swing. Mar 4 '20 at 23:25
• Jaywalk, why don't you share with us the circuit and your calculations and the simulation results? I think it would go a long way in helping us answer your deeper questions about self. Give us a chance to see what you see. And I think we may be able to help find the right keys to unlock that intuition you wonder about. But we need a peak inside to do that. Help us, eh?
– jonk
Mar 5 '20 at 6:36
• It's one of our standard circuits, I honestly can't let anyone take a look at it. I can draw up a similar circuit after work but FYI it's a solenoid driver, the command bit for it takes a lot more power than expected to run. The circuit itself works for any arbitrary beta value within spec but it depending on the beta value it consumes a lot more power. Mar 5 '20 at 14:00