555 waveform help

[pinchip]

Hi,

I am trying to make poor man's waveform generator. Few things I need help with:

1.  How to add sine wave to this model?
2. Triangle wave is not perfect triangle. Is there way to make it start from 0V and end to 5V?




EDIT: I was able to make triangle bigger by adjusting capacitor values to C1 2uF and C2 0.5uF. Now it is starting from 1V and ending to 4V. It takes some time to adjust....

[M0BSW]

Try this one

[pinchip]

Thanks! 

This helped to make it work. I need to adjust still some values to get it right

[Zero999]

You're better off using a Schmitt trigger and integrator to generate a triangle wave.

If you need a sinewave, add another integrator and an amplifier.
http://www.next.gr/oscillators/varius-circuits/function-generator-l12088.html

EDIT: rather than the TLC271 which is a single op-amp, use the TLC274 quad.

[pinchip]

Thx Hero999!

I was doing spice model accordingly and it seems to be real performance test for my computer. Simulation has been running 10 minutes and all four CPU cores near 100% utilization.

Let's see when simulation ends, if it ever ends....

Edit: simulation ended, but result was not as expected. I used LT2178 instead of TLC274 as I could not found those.

[Richard Crowley]

There is also the piecewise approximation conversion scheme to "round off" triangle-waves to sine-like waveforms...



And schemes using the non-linear IDSVDS transfer characteristic of a P-channel junction FET to shape the triangle waveform of <2% distortion...



Ref: http://www.circuitsproject.com/2014/03/simple-triangle-to-sine-converters.html

[Zero999]

Thx Hero999!

I was doing spice model accordingly and it seems to be real performance test for my computer. Simulation has been running 10 minutes and all four CPU cores near 100% utilization.

Let's see when simulation ends, if it ever ends....

Edit: simulation ended, but result was not as expected. I used LT2178 instead of TLC274 as I could not found those.

The schematic contained an error: the + inputs of some of the op-amps went to 0V, rather than half the supply voltage. I've corrected my schematic.

[128er]

Hi,
build the last shown circuit on my breadboard. Square and triangle works fine. But at the sinus output it's only DC. Changed the 10M resistor in the feedback path of the third opamp stage to 47k and it works, eeehhh, almost. Not a real sine, but better then DC.

[Zero999]

Here's another article I've found:
http://sound.westhost.com/articles/sinewave.htm

And this chematice. I can't remember the origional source though.

[ludzinc]

Thx Hero999!

I was doing spice model accordingly and it seems to be real performance test for my computer. Simulation has been running 10 minutes and all four CPU cores near 100% utilization.

Let's see when simulation ends, if it ever ends....

Edit: simulation ended, but result was not as expected. I used LT2178 instead of TLC274 as I could not found those.

The schematic contained an error: the + inputs of some of the op-amps went to 0V, rather than half the supply voltage. I've corrected my schematic.

FYI - Your R7 is set to "100mohm" - Spice interprets this as 100 milliohms.  Change to 10MEG and you'll be good.

[Martin.M]

in the bay you may find IC 8038 it contains a complete generator for sine, square and triangle

greetings
Martin

[Richard Crowley]

Remember that all of those "filter-based" solutions are good for only one fixed frequency.
If you need a range of frequencies, that is why the waveform-shaping solutions were created.

[oldway]

Some ideas 

[Zero999]

Remember that all of those "filter-based" solutions are good for only one fixed frequency.
If you need a range of frequencies, that is why the waveform-shaping solutions were created.

One way of getting round that is to use a dual ganged pot, so the filter cut-off is also adjusted when the triangle wave generator's frequency is altered.

[pinchip]

FYI - Your R7 is set to "100mohm" - Spice interprets this as 100 milliohms.  Change to 10MEG and you'll be good.

Good find! I will make correction accordingly. Thank You!

Some ideas 

WOW, I need some time to study this. Seems to be amazing stuff! 

[oldway]

Some explanations to help you:

1) use a symmetrical power supply to generate an ac signal without dc component.

2) choose a CMOS version of 555 as TLC555 or LMC555.
The output of the Cmos version can go near +Vcc and - Vcc and thus generates a square wave.
With a bipolar 555, the duty cycle is not 50% in this schematic.

3) The Cmos 555 generates a square voltage and an approximately triangular voltage which is sufficient for most of the hobby applications.
The maximum usable frequency is about 300kHz.

4) Triangular wave is high Z and must be amplified.
This is done by IC4A (1/2 TL082).

5) sine wave is generated from the triangular wave.
Harmonic distortion is about 2%

6) A very simple output amplifier with 50R output is provided.
Output amplitude and offset are ajustable.

NB: EDIT schematic.

[jlmoon]

Hi,

I am trying to make poor man's waveform generator. Few things I need help with:

Greetings,
If you need a simple function generator.. (poor mans style) have you considered a EXAR  XR-2206?  Now if this is a project (school, etc)  to help with the understanding of integration, filters and the sorts, by all means continue on your current path.  If you need a quick and dirty multi waveform generator the XR chip is a really quick and painless way to move forward. 

Regards,

JLM

[pinchip]

Some explanations to help you:

Thanks! this clarified a lot functionality

Greetings,
If you need a simple function generator.. (poor mans style) have you considered a EXAR  XR-2206? 

This is project to refresh old man's (me) electronics skills. I chose 555 initially as it is very cheap IC and I happen to have few of those around. EXAR XR-2206 seems offer better quality in all respects. I am glad that you proposed this chip. I will grab it next time I visit my part store. It will be fun to explore it after 555 experiments and maybe I could do some comparison, even we know results beforehand.

[oldway]

Do'nt forget the MAX038 ...
It's an absolete IC, expensive and hard to find but with very good performance.
Far better than XR-2206.

[robimarko]

Its like the king,but i ordered 2 XR2206 for 4.5 USD,and one MAX038 is at least over 10 USD