Voltage Controlled Sine Wave Oscillator

Thread Starter

C.Dawes

Joined Nov 14, 2020
1
Hi,

I'm having some trouble adding voltage control to an oscillator cct? For a group project we have been tasked with creating a sine wave output that's frequency must be adjusted between 1kHz to 100kHz with a 0 to 5 Volt input. Some circuits we have looked at are the Wein Bridge, Colpitts, Clap, and RC phase shift. Some searches have come up saying to implement varactor diodes in the Clapp or Colpitts circuits. However, I have been finding it easier to tune the frequency of the output with the RC phase shift network, but don't know how to implement voltage control to it. Any help would be greatly appreciated.

Cheers.
 

dendad

Joined Feb 20, 2016
4,451
You could use LEDs attached to LDRs (Light Dependent Resistors) using the LDRs as the frequency controls in a Wein Bridge circuit.
Control the LED brightness with your input voltage.
 

dendad

Joined Feb 20, 2016
4,451
I have used a 4046 PLL chip to generate a wide range VCO that will cover the range ok. Just using the VCO part will give over 1000:1 frequency range.
This is part of my breadboard, built around 40+ years ago. It has a multi turn pot as a frequency control.
In fact, I just went and measured it, and it ranges from 2.7Hz to a little over 2Mhz. Very handy in a test setup. If I remember correctly, the control voltage comes from the pot that has the low end lifted up one diode drop above 0V. It is doubtful I can find the circuit, but it may have come from a Tandy (Radio Shack) book originally.

The triangle wave from across the timing cap could be buffered and run through a triangle to sine converter, Maybe like...
http://www.falstad.com/circuit/e-sinediode.html
 

LvW

Joined Jun 13, 2013
1,752
Hi,

I'm having some trouble adding voltage control to an oscillator cct? For a group project we have been tasked with creating a sine wave output that's frequency must be adjusted between 1kHz to 100kHz with a 0 to 5 Volt input. Some circuits we have looked at are the Wein Bridge, Colpitts, Clap, and RC phase shift. Some searches have come up saying to implement varactor diodes in the Clapp or Colpitts circuits. However, I have been finding it easier to tune the frequency of the output with the RC phase shift network, but don't know how to implement voltage control to it. Any help would be greatly appreciated.
Cheers.
There is an oscillator circuit (GIC resonator) which offers the capability to tune the frequency with one single grounded resistor. This resistor can be replaced by a FET for voltage control.
 

dendad

Joined Feb 20, 2016
4,451
In the breadboard case was the 4046 VCO circuit.
4046VCO.jpg
It looks like just stray capacitance is used for the timing. Probably not ideal in real world applications, but good enough for this use.
This has proved to be quite handy.
The breadboard has a number of built in items, like the VCO. There is a debounced push button, audio logic probe, 7 seg incandescent display (do you remember those?), 8 stage divider, and a power supply.
BreadboardTop.jpg
Wiring.jpg
It is quite a while since I last had this open!
 

Ramussons

Joined May 3, 2013
1,404
How about Hetrodyning?
2 oscillators.
1 Fixed at 1 MHz.
2nd using a varactor tuning between 1.1 MHz and 1.201 MHz.
Hetrodyne, take the lower sideband - 1 to 100 KHz.
 

sparky 1

Joined Nov 3, 2018
756
@C.Dawes
What is a wide band oscillator and what is narrow band? Why would questions like these be helpful for a difficult group assignment ?
Does the criteria allow integrated circuit or discrete ? If an IC is fair play what are the methods they use to expand range using Voltage?
Can you have variable inductors and variable capacitors and how much closer does that get you to the wide band challenge ?
What are the typical frequencies of the various types of oscillators and their wave shape ?
What are some of filtering methods used in harmonic or divide by N and why it relates to broadband VCO ?
What are the various methods to expand the tuning ratio using voltage ?
Finally how do those methods work and which one would work for best for you ?
 
Last edited:

MrAl

Joined Jun 17, 2014
11,389
might be easier to generate a square wave VCO then pass it through integrator to change the waveform.
Just to note, an integrated square or triangle wave does not convert to an actual sine wave. The triangle wave converts to a v^2 wave which is close to a sine, but there will be distortion. So the distortion aspect of it becomes an issue. Same with analog wave shaping circuits like with diodes or transistors.

The nice thing is, triangle waves are sort of easy to generate so if the distortion is acceptable then that's a good way to go. We can actually calculate the THD.
 

MrAl

Joined Jun 17, 2014
11,389
Yes thanks for clarifying, the wein bridge is the best way to create pure sine waves but very difficult to make them variable.
What we could do is analyze the effects of having an imbalance in the tuning resistors. If the effect isnt too bad or we can correct it somehow then maybe that would be a starting point.

There is also the square wave and switched capacitor filter approach. The square wave is easy to get variable, and switched capacitor filters are inherently variable. The side effect is some switching noise which is then filtered with a higher frequency low pass filter.
 

LvW

Joined Jun 13, 2013
1,752
Yes thanks for clarifying, the wein bridge is the best way to create pure sine waves but very difficult to make them variable.
I think - it is the most simple way, not the "best". As I have mentioned, for single-element control you need two opamps. Even for WIEN-type oscillators there is an alternative (using a second opamp) with single-element control.

Another hint: A good and frequency-independent method for form shaping (triangle into sinus) is to use the tanh-function of a two-transistor diff. amplifier. Works good for input levels of 20...50 mV.
 

PinkMoors

Joined Nov 28, 2016
2
I have used a 4046 PLL chip to generate a wide range VCO that will cover the range ok. Just using the VCO part will give over 1000:1 frequency range.
This is part of my breadboard, built around 40+ years ago. It has a multi turn pot as a frequency control.
In fact, I just went and measured it, and it ranges from 2.7Hz to a little over 2Mhz. Very handy in a test setup. If I remember correctly, the control voltage comes from the pot that has the low end lifted up one diode drop above 0V. It is doubtful I can find the circuit, but it may have come from a Tandy (Radio Shack) book originally.

The triangle wave from across the timing cap could be buffered and run through a triangle to sine converter, Maybe like...
http://www.falstad.com/circuit/e-sinediode.html
This is undoubtedly the best way to go. I have built a similar generator, used for comms testing on custom cables.
 

RPLaJeunesse

Joined Jul 29, 2018
252
Look at using a transconductance amplifier to control an integrator, followed by a triangle to sine converter. Or a simple digital VCO feeding a switched capacitor filter that cleans up a divider also clocked by the VCO.
 

jps_hmw

Joined Feb 18, 2016
1
I have used the XR-2206,m it is a monolithic function generator integrated circuit capable of producing high quality sine, square, triangle, ramp
 
In the breadboard case was the 4046 VCO circuit.
View attachment 222801
It looks like just stray capacitance is used for the timing. Probably not ideal in real world applications, but good enough for this use.
This has proved to be quite handy.
The breadboard has a number of built in items, like the VCO. There is a debounced push button, audio logic probe, 7 seg incandescent display (do you remember those?), 8 stage divider, and a power supply.
View attachment 222802
View attachment 222803
It is quite a while since I last had this open!
Hello, and thanks for digging this up and posting it. The schematic is clear with only one exception. Can you tell us what the blob is, just below the .047 cap and connecting to GND? It's labeled WOB? IN? Thanks, Aaron
 
Top