4
\$\begingroup\$

I'm looking for timer circuits without 555 IC and trying to understand them. There are some examples that seem easy to understand, but cannot wrap my head around the others.

schematic

simulate this circuit – Schematic created using CircuitLab

Schematic from: Circuits DIY - Top 3 Simple Timer Circuits - Single Transistor Timer Circuit

So there it looks to me that when SW1 is ON, C1 capacitor is getting charged, base of Q1is active and relay is active, switching to normally closed. When SW1 is OFF, capacitor discharges through 2N3904 base terminal to ground and allowing the current to flow through the relay, keeping that whole circuit ON for some time.

schematic

simulate this circuit

Schematic from: Homemade Circuit Projects - 1 to 10 minute timer using just two transistors

Similarly here, but with added potentiometer. I think how this works is that when S1 is ON, current flows through 1M POT to the base of BC547, consequently enabling current to also flow to the base of BC557 and from emitter to the relay coil, activating the circuit. When S1 is OFF, capacitor discharges and time the circuit will be ON depends on the potentiometer value, right?

What I can't understand is how this circuit would work.

schematic

simulate this circuit

Schematic from: Digital Lab - Simple 2 minute Timer Circuit for your DIY

When S1 is ON, it turns on T2 and then T1, which allows the current to flow, lighting up the LED and charging the capacitor. But after the capacitor is charged, no more current flows, and it cannot discharge back through the transistors which will be OFF either way. What am I missing here?

Also, why does adding a second transistor in this circuit double the ON time?

schematic

simulate this circuit

Schematic from: Circuits DIY - Top 3 Simple Timer Circuits - 2 Transistor Timer Circuit

Thank you for reading this far, I would appreciate any feedback on my analysis.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Welcome. Perhaps I'll be able to chime in later. But for now, congratulations on writing one of the best formatted and clearly expressed questions to grace EESE in a long time. Please do stick around. \$\endgroup\$ Feb 8 at 20:16
  • \$\begingroup\$ I agree, the third schematic has too many bugs and cannot work. \$\endgroup\$
    – Jens
    Feb 8 at 20:58

3 Answers 3

4
\$\begingroup\$

I have fixed the third schematic:

  1. Ambivalent wire crossing R4/R5/C1
  2. Polarity of C1 corrected
  3. Added GND connection to emitter of Q1

I assume R4 was intended as inrush current limiter for the C1 charge phase. R3 is now directly connected to C1, which will have a "fade in" effect at the LED.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
1
  • \$\begingroup\$ @Kuba hasn't forgotten Monica Thank you for editing \$\endgroup\$
    – Jens
    Feb 9 at 4:25
4
\$\begingroup\$

Jens has fixed up the 3rd schematic so it should work.

As to your question as to why adding the second transistor increases the time, have a look at this simulation- I've used a 400Ω (30mA) relay coil and assumed a 5mA drop-out for the relay (pull in will be much higher) - the traces show the current through the resistor masquerading as the coil (R3/R6):

enter image description here

enter image description here

This agrees (approximately) with the doubling of time. The magenta trace shows the single transistor circuit.

The Darlington configuration (two transistors) has higher gain and requires less base current to maintain the minimum current through the relay coil. You'll also notice that the initial current is a bit less due to the increased voltage drop of the Darlington. Also the single transistor version has a very slow drop of current so it will drop out more lethargically.

It should be mentioned that none of these circuits are very stable, accurate or even very predictable even compared to the relatively sloppy 555.

\$\endgroup\$
1
\$\begingroup\$

The circuits you show are called single-shot timers, or monostable multivibrators. The latter means that they change state (vibrate - they used to be made with mechanical contacts), but always end up in a fixed state (monostable) - either ON or OFF after the delay is over.

Their main problem is strong dependence of timing on both supply voltage and temperature. The supply voltage value is a time scale factor, albeit a nonlinear one. The absolute temperature has an exponential influence on time - not great.

To make it practical, we'd want a circuit where the absolute supply voltage range is large while the time delay remains constant. We'd also want the absolute temperature to have at most a proportional influence on the time delay, rather than exponential.

To do that, we want a "cut down" or minimal comparator designed specifically for this application. It will have a much easier job than a general-purpose comparator would, since we have control over the voltages being compared.

Just like in a 555, we'll be comparing the capacitor voltage in a discharging R-C circuit to a fraction of the supply voltage. This makes the timing largely independent of supply voltage. The charging is accomplished by the pushbutton/switch, so no circuitry is needed for that, and we don't need the 2nd comparator that the astable 555 requires.

schematic

simulate this circuit – Schematic created using CircuitLab

The circuit is made up of classical analog building blocks: an RC exponential voltage vs. time source, a fixed reference source, a differential pair acting as a comparator, and a simple inverting output stage.

The reference circuit generates about 1/3rd of the supply voltage as a voltage to which the RC circuit voltage is compared. Once the RC circuit voltage drops below this reference level, Q1 starts conducting, Q3 conducts, brings Q4's base to ground, and turns Q4 off.

C2 reduces the positive feedback from the load to the reference input due to non-zero supply voltage source impedance. When the load is activated, the supply voltage drops a bit. How much depends on the circuit layout, the type of power source, distance to it, etc. No matter, though, as it is a positive feedback: load turns on, supply voltage drops a bit, reference voltage drops a bit, RC circuit's voltage is now a bit above the reference level, output turns off, supply voltage rises and reference crosses RC level, output turns on, and so on. This would cause "load chatter" as the RC circuit crosses the reference voltage level.

C2 keeps the reference voltage stable enough that the load ordinarily won't influence the comparator state, and no chatter feedback will occur.

This timer is designed to drive light loads, such as the coil of a small 12V relay may be - not less than a couple hundred ohms. To drive heavier loads, Q4 needs to be a Darlington or Sziklai pair. The transistors can be whatever you want more-or-less. 2N3904/3906 or 2N2222/2907 will work just fine, as will any other general purpose small signal transistor. Q4 may need to be beefier, depending on what load it's switching, but not for typical small relay loads.

The voltage from the RC circuit and the voltage from the reference divider both scale with supply voltage. Thus comparing them (which one is bigger) will yield same results no matter what the supply voltage is, approximately.

And that's indeed what happens. Here's a plot of the load currents vs time, at supply voltages from 3 to 23V, with the switch being released at 10ms into the simulation. The delay is fairly independent from the supply voltage, and it similarly doesn't depend much on the temperature. As far as timer circuits go, this is about the simplest one. It's possible to reduce it to 3 transistors for very light loads.

enter image description here

To change the timing, leave R1 unchanged, and adjust the value of C1.

With more transistors it is possible to design a much lower power version of this circuit. As it is, it consumes about 0.08mA per each volt of supply voltage. At 12V, that amounts to about 0.9mA. The additional transistors would act as a stabilized current source for the long-tailed differential pair, and provide additional gain in the output stage, so that the differential pair idle current could be kept much lower.

\$\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.