integrator help

[gauri_22480]

hello

herewith I am attaching circuit diag ....plz go through it......output of JK flip flop is given to integrator......here I want integrator to obtain triangular waveform fron square waveform ........but I am not getting expected results.........this integrator is not giving me triangular waveform........whats wrong with this integrator........here out put of JK flip flop is square waveform of 5 v , 50 H z........

here actually I am designing control circuit for boost converter..........so finally this integrator output will get compared with ref value ...using comparator .......accordingly will give switching pulses to switch(transistor/mosfet/igbt) of boost converter


plz help........any details related to triggering circuit for chopper

or

control circuit for boost converter

or

active current waveshapping techniques

will be appreciated.

[audioguru]

Hi Gauri,
I am sorry, but I see problems with your circuit:
1) R3 loads-down the output of U1 and is not needed. The circuit will still work with it since U1 is powerful, but it just wastes power.
2) R5 severely loads-down the output of the flip-flop and is not needed. It will make the output of the flip-flop very low, but not a sine-wave as shown on your 'scope.
3) You don't show which flip-flop part that you are using, but its set and reset inputs must connect to either the positive supply, or to ground. A 4027 flip-flop for example needs set and reset grounded. Is the flip-flop powered?
4) Integrators are inverting and have their integrating cap in the negative-feedback loop. Your "integrator" is non-inverting and its cap provides positive feedback. Swap pins 2 and 3 of U4. Is U4 powered?
5) Your integrating cap's value is way too high. It will take forever to charge and discharge. I don't have the formula handy but try a value between 0.47uF and 4.7uF for integration of a 50Hz input with the 10K input resistor.

Are you aware that your integrator's output voltage will be negative?

[gauri_22480]

hey thanx for your quick reply.........here I am using JK flipflop as frequencydivider........JK flip flop is in toggle mode............

can I use R-C integrator instead of op amp integrator?

I made necessary correction suggested by u but still not getting triangular output from integrator...... :-[

[audioguru]

Hi Gauri,
An RC network makes a lousy integrator. Its ramps are very curved.
Your opamp integrator should be perfect.

I need more information:
1) What output are you getting? Sine-wave? Square-wave? How many volts peak-to-peak? Is the entire waveform negative to ground?
2) What are the power supply voltages for U4?
3) Is the output of the flip-flop a square wave? How many volts peak-to-peak? What is the part number of the flip-flop?

[gauri_22480]

output of integrator is also square wave of 5 v , 50 hz ...... ??? ???

[audioguru]

Hi Gauri,
You didn't answer my questions so I am just guessing:
1) A 4.7uF integrating cap takes about 30msec to charge or discharge from the 5V across the 10K input resistor. 1/2 cycle of a 50Hz signal is 10ms. So the cap should be integrating unless my calc is wromg. Try 47uF. Are you certain that the resistor across the cap is not lower than 2.2M? Certainly not 10K?
2) I just realised that your flip-flop's output voltage is a 5V square-wave. You didn't say what it is even though I asked, but it is probably a Cmos IC and is probably operating with a 5V supply. Right?
Your schematic shows the flip-flop is being clocked with a 555 that has a 12V supply. The flip-flop is being destroyed by an input voltage greater than its supply voltage!
Use the 12V to also power the Cmos flip-flop.

Please post your changed schematic.

[gauri_22480]

hey.................I got the solution............

..thank you very much for your timely help.........

[audioguru]

Hi Gauri,
I'm glad that you have an outout signal but it isn't a triangular integrator waveform. I can see that its output will quickly reach its negative saturation voltage, because the value of the integrator cap is much too small. But I cannot see why the output slowly ramps in a positive direction.
The output of your integrator is supposed to be symmetrical.

[gauri_22480]

sir

I have made some modifications again ........have a look at it..........

now what u think about waveforms ?

Is this okay? or still some thing is wrong?

[audioguru]

Hi Gauri,
Good, you have an output that resembles a symmetrical triangle wave, but its ramps are exponential (curved) instead of linear.

Your 500 ohm value for R4 is so low (it should be megohms) that it takes most of the output current from the opamp so it can't be a current source (which makes linear ramps) to the integrator cap.

Also, your 200 ohm value for R3 is so low that it overloads the output of the flip-flop. If you change R3 to 10K then the integrator cap would need to be about 0.44uF to 2.2uF (the output resistance of the flip-flop increases the effective value of R3).

You could add an offset voltage adjustment resistor from the negative supply to pin 2 of the opamp to raise its output voltage to be symmetrical around 0V or entirely positive. Watch out for the polarity of the integrator cap.

BTW, you have plus and minus 20V supplies for the opamp. The absolute maximum supply voltage rating of an ordinary 741 is plus and minus 18V, with 15V supplies recommended.