psipce integrator wont work

[panfilero]

Can anyone tell me why this circuit wont simulate... I've built it, and it runs fine... the first opamp is like a comparator, it saturates to one rail and that output is sent to the second opamp which creates a ramping voltage at the output, which is sent back to the other input of the first opamp and once it passes it's threshold it flips and saturates to the other rail... basically generating a square wave, while the integrator does it's sawtooth output... but I can't get it to simulate right in pspice... can anyone point out what I'm doing wrong?

much thanks!

Oh, I know I put the schematic in front of the waveforms... trust me they are all just constant straight lines all the way across... nothing is happening behind the schematic

 

[MrAl]

Hello,

There is too much hysteresis in the circuit (around the first op amp section)
so the output of the last stage can never overcome this much and so the
circuit can not start up.

The solution is to increase the R3 to R6 ratio.
For example, changing R3 to 30k makes the circuit start to work.
You should be able to take it from there...


Im not sure what you want this circuit for actually, but if you want a nice
symmetrical triangle wave (rather than sawtoothish) then you would need to
make R2 a lower value than R1. This would be to center the change in
voltage which governs the symmetricality of the output waveform.

 

[panfilero]

Thanks for replying, I tried changing R3 to 30k... and 100k and other values... and still nothing... my output lingers around a constant approx 5V.... it's not even a saturated output... which is what I would expect.... I'm still pretty clueless about what I'm doing wrong here...

much thanks

 

[ericgibbs]

Thanks for replying, I tried changing R3 to 30k... and 100k and other values... and still nothing... my output lingers around a constant approx 5V.... it's not even a saturated output... which is what I would expect.... I'm still pretty clueless about what I'm doing wrong here...

much thanks

hi, I'll try your circuit in LTspice. OK.

 

[ericgibbs]

hi,
Look at this LTspice, it works OK, your original circuit,

 

[MrAl]

Hello again,


I got mine to work with 15k too now. i used a different op amp model.
The output is more symmetrical however if R1 and R2 are swapped.

You must have something set up wrong if it isnt working at all.

 

[panfilero]

Hmmm... that LT simulation looks good... same circuit... and MrAI got it working too... I have my simulation profile just set at Time Domain (Transient) and set to run for 50s... I've tried 1s, 5s....etc is there anything I need to set up here? (I'm pretty new to psice/orcad so I might be missing something pretty obvious to an experienced user...)

 

[panfilero]

ok... if I kick up the value of C1 to 10nF... then the thing comes on.... I don't get it... it works fine in real life... why wouldn't 47nF work for me but work for you guys... is this something about orcad/pspice?

 

[panfilero]

ok, nevermind, it's working now with the 47nF.... i don't know how or what was wrong but it's going now... thanks all

 

[mneary]

This circuit is highly dependent on initial conditions. Different simulators will start differently depending on their LM741 models.

The LTspice model appears to show the LM741 as pretty ideal; I would be suspicious. At least try loading it with 1k to Vss (GND) and see what it looks like.

 

[MrAl]

Hi again,


Yes, i think that may be because of that feedback resistor i was pointing to
earlier. If the value is doubled however, that problem should go away.
The criterion i think is just how high can the second 741 output get too before
it tops out. If it can not get high enough to overcome the hysteresis (partly
determined by that feedback resistor) then it will not start. I think the design
may be somewhat marginally thought out. A little change however should
fix it for any device, 741, LM358, etc.