signal converter

I have to do a signal converter circuit for coursework.
This circuit should transform an sine wave input in a square wave, sawtooth wave and in triangular wave with same frequency, amplitude and Dc offset from the input

The sine wave input could has amplitude varing from o to 2, and it could have a +/- 5 DC offset.

To build this circuit I can use just Opamps, Transistors(Mosfet), and diodes.

In thought passing my input wave in a half-wave rectifier circuit and then pass it output in low-pass filter to get it offset. it works fine, for a 2V amplitude wave, but it fails for other amplitudes.

To transform the wave i use a high-pass filter to get the sine ana make it pass for a comparator, then i use a sommator and a inverter.It works ok for an 2V amplitude input.
To the triangular output I'm doing something similar( after the input I pass it throught a OpAmp integrator.

I haven`t tried the sawtoooth yet.
Hope someone can help me

Welcome, BXL!

Sounds like you've gotten a good start, except for this part:

...In thought passing my input wave in a half-wave rectifier circuit and then pass it output in low-pass filter to get it offset...

Click to expand...

Not sure why you're doing this. Simply passing your sine wave through a capacitor will eliminate the DC offset. Or are you required to introduce an offset?

Anyway, can you supply us with schematics of your circuit designs?

cowboybob said:

Welcome, BXL!

Sounds like you've gotten a good start, except for this part:

Not sure why you're doing this. Simply passing your sine wave through a capacitor will eliminate the DC offset. Or are you required to introduce an offset?

Anyway, can you supply us with schematics of your circuit designs?

Click to expand...

Thanks for reply, cowboybob

You mean a capacitor in series with the sine input?
I have to save the sine wave input offset and introduce it in the ouput. I think a simple capacitor don`t do that.

The first opamp has almost no gain and the 500k resistor does nothing. If you want the first opamp to square the sinewave then the 1k and 500k resistors should be swapped.

audioguru said:

The first opamp has almost no gain and the 500k resistor does nothing. If you want the first opamp to square the sinewave then the 1k and 500k resistors should be swapped.

Click to expand...

But that the idea. The first Opamp is just part of the high-pass filter. To square the wave I use a comparator(second opamp).

The first opamp is just a high input impedance follower so the 500k and 1k resistors are not needed. Simply connect the output to the (-) input.
The squared output of the second opamp will probably have an output level higher than the sinewave. How will you reduce its level and how will you match its level to the level of the sinewave?

That Opamp was really useless.

audioguru said:

How will you reduce its level and how will you match its level to the level of the sinewave?

Click to expand...

I wasn`t able doing it. To turn around this problem, I`m feeding the comparator with the same input amplitude voltage. Ex:if my input is a 2V sine , I feed my comparator with +/-2v

You do not have a comparator. Instead you are using an opamp.
Most opamps do not work from a power supply as low as +/-2v.
The input signal level for most opamps must be LESS than their supply voltage.

If an ordinary opamp has a power supply that is +/-5v then its output will be about +/-4v.
If a "rail-to-rail" opamp has a power supply that is +/-5v then its output will be +/-5v if it has a high impedance load.

I see. Is there a not very difficult way to turn the amplitude in 2V after passing throught a comparator?Or is there a way to make the wake peek amplitude being exactly the opamp limit?

Two resistors make an attenuator. They can reduce the level of a signal to any lower level.

audioguru said:

Two resistors make an attenuator. They can reduce the level of a signal to any lower level.

Click to expand...

But attenuate my signal won`t work. I signal is already smalleer than the opamp limit. Ineed force the opamp to saturate to square it.
Is there a way to make the wake peek amplitude being exactly the opamp limit?

Your input sinewave is 2V peak-to-peak.
The squaring opamp is powered by +/-5V and produces a squarewave of about 8V peak-to-peak.
Two resistors attenuate the squarewave to 2V peak-to-peak.
Isn't that what you want?

Actually, my input amplitude is some value between 0V-2V.And the hard is to find how much I need to attenuate.

That is correct.
The squaring opamp always has the same output amplitude even when the input sinewave level is tiny.
Then the levels will not be the same.

I cannot think of an application for this silly project.

AG, I think this is a course assignment and as such probably has no useful application beyond stetching the student's thinking.

This is the problem as you stated it:

I have to do a signal converter circuit for coursework.
This circuit should transform an sine wave input in a square wave, sawtooth wave and in triangular wave with same frequency, amplitude and Dc offset from the input
The sine wave input could has amplitude varing from o to 2, and it could have a +/- 5 DC offset.

Click to expand...

If I am reading this correctly, I suggest that by maintaining a gain 0f 1 (one) for each stage of the circuit (while, of course, using dual supply OpAmps) you can obtain both signal AND offset levels at the outputs that mimic those of the input. These may require AGC circuits, but I'd try it without them first.

And, as AG suggests, a simple resistive attentuator at the output of each stage would handle unwanted amplification.

There are formulae for determining the necessary OpAmp designs and component values to do this, as well as the signal shape conversions. I'll admit, they are not simple, but they do work. Only real problem I ever had with them was adjusting for common component value variations and OpAmp response realities so as to get as close as I could to formula values (that and getting my head around "i" ).

As your assigment is for a school course, I'll assume that something short of perfect will be sufficient.

I would also suggest that you design and build this device one stage at a time, only adding the next stage AFTER confirming the proper functioning (on paper and then for real) of the previous stage. A circuit simulator would be very useful (there a number of free ones, like LTSpice, or Texas Instruments' TINA).

Of course, all of these suggestions assume that you drop the input cap from your current schematic, so as to maintain the offset value.

BXL said:

But that the idea. The first Opamp is just part of the high-pass filter. To square the wave I use a comparator(second opamp).

Click to expand...

To square a sine wave, all you have to do is give it massive gain so as soon as it crosses your threshold, it goes to the rail.