5
\$\begingroup\$

Can someone explain what this extra LED between the bases of the transistors in an astable multivibrator does?

I built the prototype and I see that without this LED there is no blinking. Both sides of the multivibrator light up constantly.

When this extra LED is added, the multivibator starts blinking as it ought to. What is interesting is that this extra LED also blinks, but with very short pulses, unlike the LEDs which are connected to the collector circuits.

The circuit is from here:

https://whadda.com/product/blue-led-christmas-tree-wssa100b/ https://cdn.velleman.eu/downloads/0/user/online_manual_mk100b.pdf

enter image description here

\$\endgroup\$
2
  • 4
    \$\begingroup\$ Please link the source of the circuit in question. Maybe the LEDs adds asymmetry so that the oscillator starts. There are other ways of doing this of course. \$\endgroup\$
    – Andy aka
    Commented Jan 12 at 9:25
  • 1
    \$\begingroup\$ Astable multivibrator: what starts the first cycle \$\endgroup\$
    – Andy aka
    Commented Jan 12 at 11:36

2 Answers 2

4
\$\begingroup\$

With some tinkering with a bespoken polarized-cap model parameters I've managed to reproduce the behavior the OP described for their circuit. In this simulation, reverse-voltage current leakage is modeled with standard default diodes D16 and D17 from LTspice's standard.dio, without any adjustment made to saturation current or other diode model parameters. The current leakage should be included in the model: the circuit with non-polarized capacitors C1, C2 would oscillate unconditionally, no "extra LED" is required to be added in the transistor base networks.

Now, adding reverse-voltage current alone into the model is not sufficient to suppress oscillations. The oscillations can be suppressed with the coarse ideal-diode model of leakage current; however, with voltage polarity reversed, the real valve effect turns off not abruptly, as if with ideal diode switch-on, but softly, as if with real diodes. Now, in the real electrolytic capacitor, the transition to reverse polarity is also accompanied with capacitance value variation. I have no idea of how much this capacitance value changes; this parameter is not readily available in datasheets. If we adopt the rather drastic plummeting to a value of 1 uF at reverse voltages, as in the formula for capacitance values for C1,C2 capacitors of my simulation, Q = if(x>-0., x*22µ, x*1u), the simulation plot shows that oscillations do not start and diodes in both multivibrator legs light up constantly (Ic(Q1)=4mA, Ic(Q2)=2.9mA):

no-extra-LED

The upper plot pane shows reverse polarity voltages of -0.56V across capacitors C1,C2.

With the "extra LED" D15 added, the oscillations start:

extra-LED-added

the "extra LED" is blinking exactly like it is described in the OP post:

blinking-extra-LED

Disclaimer: The assumptions about the electrolytic capacitor model and model parameter values in this post are rather arbitrary. I am introducing the model and posting the simulation plots here with the only purpose to indicate the possible explanation of the OP's observations. No sources I've consulted support the validity of this model for simulating electrolytic capacitors in circuits with SPICE simulators.

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

As already pointed out by @Andy aka, diode (my D3) introduces a dissymetry in the behavior.

This diode is lighted only for a very short time as simulated.
This behavior ... because the inverse voltage on (my) Q1 base is just limited to ~ -1.2 V)

enter image description here

\$\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.