0
\$\begingroup\$

I have a source (J1) that varies from 2.0 V to 8.0 V. I'm trying to switch on a NPN transistor (Q2) when the source is above 5.1 V such that it can supply a control line (J3) via a PNP transistor (Q1).

In simulation Q2 does switch on and it does supply the control line, but in practice Q2 does not switch on. I can provide more detailed info on measurements if needed.

What could be wrong with the design?

enter image description here

\$\endgroup\$
12
  • 3
    \$\begingroup\$ What voltage did you test the circuit at? I'd expect the real-life version to turn on at about 6V, not 5.1V, due to the base-emitter voltage drop. \$\endgroup\$ Jan 15 at 22:21
  • \$\begingroup\$ Does control go high at all or is it consistently low? Keep in mind that BJT transistors don't turn on as soon as you apply a slight voltage to them, they have the equivalent of a diode drop across them from base to emitter for NPNs. So for Q2 to turn on, its base needs to be at ~.7 volts which means it won't try turning on until the input is actually at 5.8 volts with that Zener. \$\endgroup\$
    – cEEa
    Jan 15 at 22:21
  • 4
    \$\begingroup\$ Oh, and your circuit will also short out the supply voltage through Q1's base-emitter diode when it does turn on. That might have just killed your transistors when you tested the circuit. Q1 needs a base resistor, i.e. 1k Ohm. Remove the transistors from the circuit and test them, they might be dead. \$\endgroup\$ Jan 15 at 22:22
  • \$\begingroup\$ @JonathanS. I'm aware the Q2 will only turn on fully at 5.1V + 0.7V.That might be it. I never saw that path from Q1 through Q2!! \$\endgroup\$
    – bluscape
    Jan 15 at 22:25
  • \$\begingroup\$ I agree with Jonathan - check Q1 and Q2 for death. \$\endgroup\$
    – AnalogKid
    Jan 15 at 22:44

2 Answers 2

1
\$\begingroup\$

So the solution was to add a resistor to the base of Q1 as pointed by Jonathan S in the comments. See the updated schematic. But I now have a new problem. With a 5.1V zener Q1 will switch on at 3.9V and with a lower voltage zener like 4.3V, it will switch on at 1.9V. I was not expecting this to happen. With a 5.1V zener I would expect a 5.1V voltage drop and you should only see any current or voltage at the base of Q2 around 5.1V (the breakdown voltage). I know it will conduct a little earlier but not that early. When I use an 8.2V zener I see the expected response. Q1 will switch on at 8.8V. What could be causing this?

enter image description here

An update.

As pointed by td127, I needed more current through the zener. I added a low value resistor at the zener anode to ground. This solved the problem.

enter image description here

enter image description here

\$\endgroup\$
3
  • \$\begingroup\$ You're relying on a trickle current through the zener to generate 5.1V but a zener typically needs on the order of a mA to comfortably hit its breakdown point. \$\endgroup\$
    – td127
    Jan 23 at 19:40
  • \$\begingroup\$ You got it right. There was not enough current for the zener to fully turn on. But to make a sharp turn on of Q1 and Q2, you can eliminate R5 and change R3 to 1K. \$\endgroup\$
    – VictorTito
    Jan 23 at 21:03
  • \$\begingroup\$ I was waiting for someone to mention that. I cannot reduce R3 since another line is also controlling the base of Q2. Otherwise I would have done that. Thanks \$\endgroup\$
    – bluscape
    Jan 24 at 5:30
1
\$\begingroup\$

You will be able to get this to work provided the ambient temperature varies only over a limited range and after carefully hand-selecting the resistors to compensate for the transistors' gain and the Zener diodes relatively soft knee (both of which vary with temperature). That approach may be acceptable for your application.

A robust solution would be to use a band-gap reference that is invariant to temperature. Consider using a TL431 or equivalent. If you have never used that part before, it basically works like a "precision NPN transistor" in which the equivalent "BE junction" is precisely fixed at 2.5V and (mostly) independent of temperature. You will need the TL431, a PNP transistor, and several resistors for bias and setting the trigger voltage.

enter image description here

The advantage here is you get a well defined trigger point (5V in this case but you can bump that up to 5.1V if you desire) that is no longer sensitive to transistor gain or the effects of temperature. Hope this helps.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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