开云体育

Hello everyone!

I'm building a PWM control circuit for my single-phase inverter.
I started by creating the control stage using a triangular wave source (Vp) and a DC source (Vc).
I included an inverting op-amp to invert the triangular wave, and the buffer at the comparator’s output is meant to isolate the signal.

After completing the setup, I tested it and noticed that the output from my comparator was a triangular waveform with a peak-to-peak voltage in the megavolt range, which is obviously incorrect.
Then I imported my inverter circuit and connected it to the comparator’s output to see if anything would change, but it still doesn't work properly.

Could someone help me figure out where I might be going wrong?
Could it be that the generic op-amp I'm using isn't suitable for this circuit and I should use a different model? If so, which one?

I will upload my .asc file named invmono_controle to the library. Feel free to test it.

Thank you!

On Fri, May 16, 2025 at 03:01 PM, <guilhermesouzam01@...> wrote:

After completing the setup, I tested it and noticed that the output from my comparator was a triangular waveform with a peak-to-peak voltage in the megavolt range, which is obviously incorrect.

Why is that obviously incorrect?

?

The ideal op-amps you used have no power supplies, and they are capable of driving infinite voltages, both positive and negative.? If that is not what you wanted, then you should have used a different one.? Consider using "UniversalOpAmp2" instead of "opamp".

?

That is especially important for U2 and U3,, which are both open-loop, without any negative feedback.? Configured that way, they act like comparators, with outputs that are unbounded.? The only thing keeping their output voltages from increasing indefinitely, is that they have a finite slew-rate, and the outputs are still slewing (ramping) when the input changes polarity and they start slewing in the opposite direction.

?

FYI, you should not have used ".lib"C:\Users\guilh\AppData\Local\Programs\ADI\LTspice\lib\lib\sub\opamp.sub"".? The instructions you saw when you added the "opamp" component was to add ".lib opamp.sub".? The full path is unnecessary and should not have been used.

?

Could someone help me figure out where I might be going wrong?
Could it be that the generic op-amp I'm using isn't suitable for this circuit ...

... and I should use a different model? If so, which one?

Use "UniversalOpAmp2".? It is located in the same section of the components picker, just a little further down alphabetically.

I will upload my .asc file named invmono_controle to the library. Feel free to test it.

When uploading your files, always upload them only into the "Temp" directory.? (I fixed yours.)

?

Andy

?

I wrote:

That is especially important for U2 and U3,, which are both open-loop, without any negative feedback.? Configured that way, they act like comparators, with outputs that are unbounded.? The only thing keeping their output voltages from increasing indefinitely, is that they have a finite slew-rate, and the outputs are still slewing (ramping) when the input changes polarity and they start slewing in the opposite direction.

I'm somewhat incorrect, in my explanation.? It is not the finite slew-rate that limits their output voltages.? It is the finite open-loop voltage gain, which defaults to Aol=100K.? They amplify the triangle waves from 24 V peak-to-peak, up to 2.4 MegV peak-to-peak.? The finite gain-bandwidth product (GBW=10Meg) makes their outputs a little less than that - only about 1.75 MV instead of 2.4 MV - but you get the idea.? Those amps never clip, so their outputs can be enormous.

?

Use a UnicersalOpAmp2 instead.? You will have to give them supply voltages too, of course.

?

Andy

?

?

Here's another suggestion, if I may add one.

?

When using numbers such as ".008", don't write them that way.? Writing a number with a leading decimal point can lead to mistakes when someone else looks at your schematic.

?

Also, I encourage you to make use of unit multipliers.? Nobody likes counting zeros.

?

So, instead of ".008", write either "0.008" or preferably "8m".

?

Andy

?

?