function generator

[soundman]

Hi,

My professor wanted me to try to make a Function generator by not using does ready function generator chips.

Internet has a bunch of that kinds of function generators, But Can you guys tell me the best one's in your opinion or maybe that one's which you have made and are satisfied with?

Thank a bunch.

 

[Hero999]

It depends on what functions you need at what frequency range?

Here's a couple of ideas.

The circuit for the triangle wave generator is the site linked below.

I can't remember where I found the wave shaper from, I think it was EDA?

The resistor values in the wave shaper are critical and should be 1% tolerance of better resistors, 5% isn't good enough.

The filter has the disadvantage of requiring a dual ganged pot and having to vary both C1 and C3 when changing the frequency range.

The wave shaper based function generator, only needs P1 to be adjusted and C1 to be changed when switching ranges.

 

[soundman]

Thanks,
My professor wants me to try to get at least 1kHz (the Max is not important) and an amplitude of 10V when I try to make a function generator, Any idea please?

What's your idea about this: https://www.electro-tech-online.com/custompdfs/2010/05/k23.pdf

Wave shapes, the minimum values of 1kHz and the amplitude of at least 10V and finally make the circuit on a PCB are the whole things which my professor takes into account.

Thanks

 

[soundman]

2 other questions regarding to the K23.pdf link,
1: What are those 4 diodes in the schematic?
2: Cam I incarse the power supp;y to 18V by no change in the circuit irself?

Thanks

 

[Hero999]

Aren't you supposed to design it yourself?

If it's a soldering test then fair enough, get the kit but if you actually need to design it yourself, it won't do.

EDIT:
The purpose of the diodes is explained in the PDF.

For more information research the diode formula in Wikipedia.

 

[soundman]

Thanks,

How can I get a domain of 10V? Can I adjust the output amplitude somehow to?

 

[audioguru]

Do you want the output to be 10V peak-to-peak or 10V RMS?

In the kit:
Do you know what is the level at the 4 diodes?
Do you know that the ratio of R9 and R10 sets the output level of the sine-wave?

 

[ericgibbs]

Thanks,

How can I get a domain of 10V? Can I adjust the output amplitude somehow to?

hi Soundman,
Its a quiet day, so here is your circuit running in a LTspice simulation, at approx 1kHz.

As you can see its a rough Sine wave.

EDIT:
This 2nd plot shows the actual voltage levels.

 

[soundman]

Do you want the output to be 10V peak-to-peak or 10V RMS?

In the kit:
Do you know what is the level at the 4 diodes?
Do you know that the ratio of R9 and R10 sets the output level of the sine-wave?

Hi, Well I want it to be at least 10V Peak to peak, I want to use TL074 Instead the LM348 Chip And use a 18V power supply if possible.

About your second & third questions, Are you trying to find out my knowledge in electronics?
The level at diodes should be (1K/(1k+820R))x 9V.
yea I know how an op-amp does work and Know the formulas for an inverting/non inverting confuguration

 

[soundman]

hi Soundman,
Its a quiet day, so here is your circuit running in a LTspice simulation, at approx 1kHz.

As you can see its a rough Sine wave.

EDIT:
This 2nd plot shows the actual voltage levels.

Hi,

Thanks a lot for your time.

Yea, I have simulated it by Proteus software and noticed that the Sine wave is not so good, I could improve it a bit by adding a 10nF cap across the R10. How can I improve it yet?

Can I add another op-amp somehow to increase the output values? I want it to be adjustable so If I could use an op-amp at the output just a pot instead of the feedback resistor could be used to change the output voltage values.

 

[soundman]

By the way how could you get 9V for the square output At the right pic??!
There is a big difference between the amplitudes at the left and the right pics!!!!

 

[ericgibbs]

By the way how could you get 9V for the square output At the right pic??!
There is a big difference between the amplitudes at the left and the right pics!!!!

hi,
The left plot uses that 'dview' option which allows the plot to show the outputs on separate sections of the plot, one above the other.
The right plot is showing the actual voltage levels as measured, the reason for the 9V square wave is that the OPA's are set for rail to rail.

The form factor [quality] of the 'sine wave' output will change as the frequency changes, you may have to add a 'form factor' pot so that you can trim for the best sinewave at different frequencies.

If you want 10Vppk signals I would suggest at least a 12V supply [ideally I would go for +15v].
Use a decent OPA that will allow swings close to 0V on the output.


EDIT:
This is a plot using the LM348.
Compare this to the previous OPA, it shows the difference.

 

[Hero999]

Yuck, that's a poor excuse for a sinewave.

Both the circuits I posted make a better sinewave than that.

The simple RC filter shaper gives a reasonable sinewave.
**broken link removed**

Here's what the waveform looks like with the component values given above:


See, it isn't perfect but it's better than the kit.

It can be improved by increasing C3 but the gain needs to be increased (increase R5) which will reduce the maximum usable frequency.

As you seem to be pretty new to this, I'd recommend going for the basic RC filter wave shaper.

Here's a site which gives the formula to calculate the frequency of the triangle wave generator I used in my circuit:


Here's another site with a more detailed explanation:
Op Amp Triangle-Wave Generator

C2 is a DC blocking capacitor so it needs to be high enough to pass the lowest frequency of interest.

The minimum value can be calculated using the following formula:
[latex]C = \frac{1}{RF}[/latex]

R is the value of the load impedance, in this case R1 + R4

F is the frequency when the signal will be reduced to 70.7% of the maximum value, so it's probably a good idea to set this variable to halve the minimum frequency of interest.

C is the required capacitance in Farads.

 

[soundman]

ok THANKS,

Can you guys tell me how to make the output so that it to be adjustable? Does a simple pot at the final outputs will do the job??

 

[Hero999]

Guess.

If you understand how the circuit works, you should be able to figure that out for yourself.

Hint: adding it directly to the output will increase the output impedance, which is not good.

 

[soundman]

Hi Guys,

I want to be able to change the duty cycle in this circuit too:
https://www.electro-tech-online.com/custompdfs/2010/06/k23.pdf ]
how Can I do so?

By the way Is there any simple way to generate smooth wave by the said circuit too?

Thanks a bunch

 

[soundman]

I need to change at least the duty cycle of the said circuit, But do not know where to put a pot? Any idea where to put it?

Thanks

 

[audioguru]

I think the duty-cycle is changed if you change the ratio of R1 and R2. They can be a pot.

 

[soundman]

Are you sure about that? I suspect that changing those 2 resistors effectively change the duty cycle, I guess It change the freq too?

 

[ericgibbs]

Are you sure about that? I suspect that changing those 2 resistors effectively change the duty cycle, I guess It change the freq too?

hi S,
The original circuit was designed to give an approx 50/50 square wave so that the 'sine' wave could be approximately shaped, if you change the mark/space ratio the sine wave will be badly distorted.