Triangle wave (170Hz) generator's circuit requirement

[wkrug]

For initially for 2 cycles of triangular wave was not good and bit low amplitude but after 2 cycles it was stable and works good

The problem is that the rectangular wave swings between 0V and +5V.
That will bring You a DC component of 2.5V at the - Input of Your OP-Amp.

You can avoid this when using a positive Voltage ( around 2.5V ) at the +Input of the Op-Amp by an voltage divider.
The exact Voltage depends of the Supply of the digital part's and the voltage losses of this parts, additional the input offset of the Op-Amp.
So the idea of gophert to use a pot there is very good.

Another Idea is to use only one supply with 10......12V ( than You have to use C-Mos 40xx chips ) and use the second OP of the LF353 to generate a virtual mass point ( +5V ).
The triangle wave swings in this case around 5V above GND.

 

[gophert]

You can avoid this when using a positive Voltage ( around 2.5V ) at the +Input of the Op-Amp by an voltage divider

note the two resistors and potentiometer connected to + input to generate said 2.5v virtual ground. The problem is the smallest DC drift causes the waveform to eventually migrate to one rail or the other.

Our OP cannot find a potentiometer in LTSpice. Is that truly missing or can someone point out how to find the Pot icon for him? I haven't used LTSpice in a long time and it is not on my current laptop

 

[atferrari]

Eric Gibbs is the man.
ericgibbs
He supplied the pot model years ago. Pity I lost it in my defunct PC. So much for a non-existing back up.

 

[wkrug]

The problem is the smallest DC drift causes the waveform to eventually migrate to one rail or the other.

You're right.
Probably You can use the 2nd OP to steer the reference Voltage ( + Input ) of the integrator?!
When put a Low Pass filter in the output of the integrator and feed with this signal the 2nd OP that comare that with an reference Voltage.
So the drift should be compensed.