Maker Pro
Maker Pro

Oscillator circuit - possibly Wien Bridge Oscillator?

TFahey

Mar 30, 2014
5
Joined
Mar 30, 2014
Messages
5
Hello all,

As part of a first-year electronics project, I decided to do a voltage controlled oscillator. After looking at a few schematics online for ideas, my lab partner and I settled on two designs, the first of which didn't work. After building the second design, it worked perfectly...only one problem - we don't know why! :eek:

Here is the circuit as we constructed it:

Circuit Diagram wout Capacitor.jpg

It was based on a drawing we found on the internet that was similar, but there was a 10nF capacitor in parallel with resistor R1 in that design, also wired up to the ground. In terms of our circuit, it would have looked like this:

Circuit Diagram with Capacitor.jpg

Unfortunately, the lowest capacitance capacitor we had access to was a 100nF, which didn't work (it just outputted a constant DC voltage of around 15mV), so we took it out to get the first circuit shown, and lo and behold, we got a lovely oscillating waveform.

The interesting thing about this oscillator is that there's no DC input into the OP amp. The way my lab partner described it was that it must "feedback off of itself", possibly using the power supply for the OP amp.

I did a bit of research to try and figure out how exactly the circuit worked, and came across a type of oscillator called a Wien Bridge oscillator that looked very similar to our circuit, except they all had capacitors going into the OP amp +ve input (like the original schematic, which we ignored because it didn't work). I think I understand the basic concept of how the Wien Bridge oscillator works - the oscillating waveform is formed by the charge and discharge of a capacitor? What I don't understand though, is, in our circuit, the only capacitor (C2) is one that's not part of the feedback loop - it's only part of the circuit that goes to the output. So how does it know when to charge/discharge? It's really weird - the capacitor that's meant to cause the oscillation in theory, is the one that stopped the oscillation occurring in practice! :confused:

I tried plotting the output in LTSPICE, which came up with even more weird results:

Graph 1.jpg

...and a close up of the signal:

Graph 2.jpg

The waveform we got looked like this:

2014-03-25 12.37.08.jpg

So. in summary, I'm pretty confused. One possible explanation is that the wires themselves caused a capacitance to occur, providing the 10nF from the original schematic. However, I don't really like this explanation, as it's making a pretty large assumption. Another thought that occurred to me was that the comparator that's meant to cause the capacitor to discharge when it reaches the correct voltage isn't able to keep up with the speed of the charge discharge, and so there's periods of time when the voltage is at a fixed level, before the comparator "catches up", and causes the capacitor to discharge. This would explain the waveform on the oscilloscope, and go part way to explain the results in LTSPICE. Is there something I'm missing?

Thanks very much guys. :)
 
Last edited:

Arouse1973

Adam
Dec 18, 2013
5,178
Joined
Dec 18, 2013
Messages
5,178
Look at the position of C1, clue is it is in the wrong place. Make sure you are using the correct type of OP-amps, look up single supply and rail to rail OP-amps.
Adam
 

LvW

Apr 12, 2014
604
Joined
Apr 12, 2014
Messages
604
@ TFahey: There are several reasons your circuit cannot work as desired.
At first, your power supply has no ground reference - thus, there is no proper operational point. You should use a split power supply.
Secondly, each linear oscillator needs a second-order feedback path.
Therefore: Do you intend to create a linear oscillator with a sinusoidal output (because you are referring to the WIEN type) ?
 

TFahey

Mar 30, 2014
5
Joined
Mar 30, 2014
Messages
5
Hello,

Sorry for the lengthy gap, I've been spending time doing revision for my upcoming Easter exams!

@Arouse1973
You're right it is in the wrong place - but somehow the circuit produced the desired effect anyway! I just can't figure out how :/
Also, the power supplies are labelled wrong - that was simply a graphical mistake on my part not fully understanding how to represent the op amp in Ltspice. The circuit we built was rail-to-rail, with a +9V to -9V, dual power supply setup.

@LvW
As above.
A second-order feedback path? I think that terminology may be beyond my fairly limited knowledge at present...I do know that there isn't a capacitor in the feedback loop, where there should be one - I'm just wondering if the capacitance of the wires themselves is taking the place of the 1nF capacitor that was meant to be in the feedback loop.
The original intention was to create something along the lines of a voltage controlled oscillator - just a basic synthesiser that would taken a linear voltage input from a function generator or battery, and output a sinusoidal output whose frequency would change with the input voltage. However, by changing tack to a circuit that didn't take an input as originally specified, I've got rather muddled as to how the circuit actually works.
 

LvW

Apr 12, 2014
604
Joined
Apr 12, 2014
Messages
604
@TFahey: In case you need further help it would be good (in fact: necessary) to tell us what you really want to design.
I assume, you want to create a linear oscillator with a sinusoidal output? Correct ? Which frequency? Tunable about what range? Amplitude?
 

TFahey

Mar 30, 2014
5
Joined
Mar 30, 2014
Messages
5
@LvW
Apologies for not being clear, obviously, I'm quite new to electronics, so I'm not used to getting the vital information across immediately :/

I set out with the aim to build a linear oscillator with a sinusoidal output - I should emphasise, this was all as part of an electronics project, so there was a bit of experimentation involved with the circuit. We ended up with the circuit you see in my original post.
There was no particular frequency specification, although one that operated in the audible range was preferable, so we could play it through a speaker. We ended up with a frequency range between around 5250Hz to 5900Hz, simply by changing the values of the resistors (as they were all variable resistors). The amplitude was 1V peak to peak.
 

LvW

Apr 12, 2014
604
Joined
Apr 12, 2014
Messages
604
@TFahey
OK - you need an oscillator tunable between 5.25 and 5.9 kHz with an amplitude of 1Vpp.
Several circuit topologies are possible and extensively described in the literature. The circuit you have shown looks a bit "uncommon" (In fact: I have never seen it and I doubt if it will oscillate.)
Where did you find this circuit? And why did you choose it?
 

TFahey

Mar 30, 2014
5
Joined
Mar 30, 2014
Messages
5
@LvW
This circuit was a modification of a circuit that looked more like this:
Circuit Diagram with Capacitor.jpg

We found that when the capacitor C1 was included in the circuit, a constant DC signal was measured at the output. However, removing C1 from the circuit, gave an oscillating waveform like the one shown in the original post. I should add, C1 should say 1nF, not 10nF. Unfortunately, we did not have any 1nF capacitors, and thus the capacitor we did use, which caused the circuit to give the constant DC signal when wired up, was around 100nF, as these were the smallest available in the lab.
 

LvW

Apr 12, 2014
604
Joined
Apr 12, 2014
Messages
604
The circuit as shown is very far from being a harmonic oscillator. More than that, the components R7, R6, C2, R8 do not contribute to the desired dynamic behaviour because they act as a load only (with a poor high pass characteristic).
Why don´t you try to use one of the well-known circuits (WIEN, BUBBA, phase-shift, double-integrator,..)?
 
Top