Wien bridge oscillator with lamp in amplitude control has irregular oscillation.

Thread Starter


Joined Dec 12, 2019
Hi all

I built wien bridge oscillator using a lamp to control amplitude, but I see that the oscillation is irregular.
One thing I noticed is that the amplitude is very sensitive to the variation of R1.
If I reduce R1 to 620 ohms the oscillation starts to swing until it dies close to zero. (printscreen 1).
If I increase R1 to 630 ohms, the oscillation increases to a maximum value, then starts to drop until it reaches zero. (printscreen 2).
Could someone give me a hint on how I can stabilize this amplitude?





Joined Jul 10, 2017
I am not able to open your file but if the circuit will not sustain oscillation, there is too much negative feedback. You stated that you built the circuit but only posted simulator waveforms. What does the real circuit do? The circuit below is a very generic Wien bridge oscillator. The variable resistor RV2 adjusts the amount of negative feedback. In your circuit, this value is not high enough if it will not oscillate.


Thread Starter


Joined Dec 12, 2019
In the breadboard circuit, the oscillation is almost perfect, different only in the simulation.
I would like to know the reason simulation dont work as circuit real.


Joined Jun 17, 2014
The pot and the bulb form a voltage divider which sets the gain. If either of these elements is not of the right value the circuit will not be able to sustain oscillations.

The pot resistance can be adjusted, but the bulb resistance is a function of the part number of the bulb. In general large bulbs will have low resistance, small bulbs higher resistance. If the resistance is not right though it will cause a gain that is too low and thus the circuit will not keep the oscillation going.

So the point is you must have the right bulb, the right part number. To determine this it would be good to have the curve of the bulb, the voltage vs current curve.

Another interesting thing about bulbs is that they get hot and when you heat up a metal it takes some time to cool down. While it is hot it has a higher resistance, and while it is cooling down, it retains some of it's higher resistance. That acts as a memory so it acts sort of like a short term memory bilateral memristor.
However, the oscillation period of the oscillator is usually much shorter than the cool down time, so the bulb acts as a resistor that can change resistance with current and remember that resistance over the entire cycle of the oscillator.

The resistance vs current though is very particular to the bulb part number, and hence the spice model that represents that bulb. You must have the spice model that matches the physical part.
If you dont have this, you can make one but you will need the curve of the bulb to do so. You could get that curve from direct measurements of the bulb using measuring equipment like a voltmeter and current meter. The timing factor is not as critical so you could estimate that in creating the model, as long as it holds its resistance for a while it should be good.

The mechanism by which the bulb adjusts the gain is simple. As the voltage of the output increases, the voltage across the bulb increases and so the current though the bulb increases. As the current increases, the bulb heats up more and so the resistance goes up. As the resistance goes up, the voltage across the bulb increases more and that means the circuit ends up having more negative feedback which reduces the gain. The reduction in gain lowers the output voltage, thus maintaining some degree of output voltage regulation.,
To get that to work though there must be more than enough gain to begin with, and that is what the pot does. Adjusting that gives the circuit enough gain to start oscillating and keep oscillating.
Last edited: