# Help with gain of the basic amplitude limiter

#### chihiro

Joined May 18, 2008
22
hello everyone! below there is a circuit that i took some help from Franco's operational amplifier design book, as u can see i got a gain around 3 from the 20k/(20+11k) resistors. but when i increase the 20k resistor to have a higher gain, i got my output sinusoidal waveform like noise signal. As i heard from the Franco's book, the RC values (158k, 1nF) are designed to have a correct sinusoidal waveform around a 3 gain circuit. i have posted the oscilloscope screenshots below also, anyway i need help on this circuit to set the max. gain around 50, when i got the 20k to 490k i got a distorted signal looks like nothing. I didnt have any knowledge on amplifiers without an ac input, as i understanded from the book, this circuit creates its own ac input and amplifies it. Thanks for the help from now.

NOTE: 2nd pic is the output signal V2, 3rd is the input V5... thanks

#### Attachments

• 96.9 KB Views: 50
• 77.7 KB Views: 39
• 76.7 KB Views: 38

#### chihiro

Joined May 18, 2008
22
by the way the 2V dc voltage and the 2M ohms resistor are for setting the VGS of the JFET to set the output peak limit which has almost no use for any misunderstandings.

#### beenthere

Joined Apr 20, 2004
15,819
Several questions come to mind - why a gain of 50? What does the JFET do for your purpose?

The circuit is built around Wien Bridge oscillator. Here is a link to a site that discusses the principle and does a simulation - http://www.ecircuitcenter.com/Circuits/opwien/opwien.htm

#### chihiro

Joined May 18, 2008
22
first of all thanks for the fast answer, the gain is 50 since our project requires that and i used jfet to have it work as a diode with turn on voltage, to limit the gain after a preset reference voltage. but the it's the same info about wien-bridge oscillators in the book, do i have to get a gain of 3 all the time to have this oscillator works ?

#### chihiro

Joined May 18, 2008
22
and with any change of 10k resistor to increase the gain, i got a square waveform from the output, and something totally unfamiliar from the input signal, so isnt there anyway to get a gain around 30-50 by changing the 10k resistor to a lower value, and setting the 158k, 1nF components to work with that gain properly ?

#### Attachments

• 61.1 KB Views: 25

#### beenthere

Joined Apr 20, 2004
15,819
The resistor - capacitor series and parallel arrangement are the frequency setting elements. You can adjust a resistance somewhat to set an exact frequency, but that is about the limit.

I might point out that a large increase in gain will give you a square wave output, and there are easier ways to make a square wave oscillator. If you need a larger sine wave, it would probably be simpler to add an op amp to give the extra signal excursion.

#### steinar96

Joined Apr 18, 2009
239
I would recommend a second amplifying stage as beenthere suggested. You can feed your sin wave to it and get the amplification you want.

#### chihiro

Joined May 18, 2008
22
yeah the second amplifier helped me with the issue now i can get a final output with a gain around 50, but is there anyway to decrease the amplitude of the input signal? i tried many combinations of RC values, and all those combinations resulted a distorted waveforms both in output and input, not even a square or something :S, i thought i could apply voltage division to the input, putted a resistor to the input line and again, waveform distorted :/

#### Audioguru

Joined Dec 20, 2007
11,249
The rectifier diode rectifies the output of the oscillator and produces a minimum of 0.5V peak. The 0.5V or more is filtered and used to control the FET attenuator so that the output of the generator is at a constant level of 0.5V peak or more. If you amplify it 50 times then the signal will try to be 25V peak.

Make an attenuator with two series resistors from the output to ground as a voltage divider for the output of the generator and feed the attenuated signal to the amplifier.

#### chihiro

Joined May 18, 2008
22
yeah if i get the output of the oscillator as the input of the gain calculations, that would solve my problem thanks audioguru, there is one more problem left, when i use a diode combination for both positive cycle and negative cycle instead of this circuit, i can get a pure sinusoidal, however when i use this jfet circuit i get a clipped negative cycle from the output, any idea of fixing that clipped part ? i tried to decrease the gain ~ 3 in the circuit but it has a too narrow range of changing with those values.

#### Audioguru

Joined Dec 20, 2007
11,249
yeah if i get the output of the oscillator as the input of the gain calculations, that would solve my problem thanks audioguru, there is one more problem left, when i use a diode combination for both positive cycle and negative cycle instead of this circuit, i can get a pure sinusoidal, however when i use this jfet circuit i get a clipped negative cycle from the output, any idea of fixing that clipped part ? i tried to decrease the gain ~ 3 in the circuit but it has a too narrow range of changing with those values.
The Jfet adjusts the gain to 3.0. You can adjust the output level by changing the gate bias voltage of the Jfet.

You have 1F (!) for C3 instead of a more reasonable 0.1uF or 1uF.
You should use a better opamp because the 741 opamp is 42 years old and has poor performance. Many do not work when the supply is only plus and minus 5V.

#### chihiro

Joined May 18, 2008
22
yeah 1F value seems a little irrational, but i roughly saw that the waveform became more "sinusoidal" when i increased the value of that capacitor. thanks for the advices, have a nice day

#### chihiro

Joined May 18, 2008
22
by the way, this circuit seems almost ok when i simulate it with ni multisim, but i have some doubts about the practical case, since its an oscillator without an external ac input, i thought there will be a tough noise effect when i construct the circuit on a breadboard, is it a possible concern? or are there easy ways to get rid of it ?

#### chihiro

Joined May 18, 2008
22
and one more question - sorry for seperate posts - can u suggest a better and easier to find commercial op-amp instead of 741 ? like LM series or others etc.?

#### Audioguru

Joined Dec 20, 2007
11,249
I use TL071 single, TL072 dual and TL074 quad low noise audio opamps. They have been used in hi-fi circuits for many years so they are available almost everywhere for a low price. I also used the LM833 dual audio opamp.