Single Phase to three signal generator.

Do you have a visual, like a block diagram? That might help people better see what you are doing.

I'm not entirely sure what you're looking for, especially the ultrasonic bit!

A diagram would definitely help.

For the 3-phase generation portion, in theory that's quite easy using any microcontroller with PWM, and a simple filter to construct the waveforms from that:

/technologies/power-management/b/blog/posts/building-a-three-phase-mains-power-system-simulator

However you may have different requirements so it needs more info. (the microcontroller approach just mentioned only works at low frequencies, not ultrasonic).

Here is the circuit and the output. The input in the signal gen is 110kHz and 4Vp.

Here is my initial circuit, the output signal will be use as a frequency output for a transducer

It would appear from your schematic, that because the output of the first stage, feeds the input of the second stage, and the output of the second stage feeds the input for the third, the amplitude changes are being caused by a gain less that one in each stage.  You could test this by increasing the resistance for R2, R4 and R6 to increase the gains.  The gains appear to be about 0.5 right now, so you might want to double the resistance.  Good luck

Thanks! I will try this approach. 

I think that the amplitude loss is caused by the capacitor. The OpAmps seem to have unity gain.

Don't try it as you've drawn it. The maximum supply for a MCP6291 is 6V, 10V will blow it up.

At 110KHz, the 8.353pF capacitor has an impedance of 173KΩ, which is in series with the 100KΩ, so yes the capacitance is effecting the gain.  One way to change the gain is to increase the feedback resistance.

First, let me quote you. 

"I am designing a low-voltage power supply that will run single-phase to three-phase circuit and will be using ultrasonic frequency as its signal input. The current circuit design uses three operational amplifiers that has a 60 degrees phase shift."

To convert single-phase to three-phase should be at 120 degrees apart, Not 60 degrees as 60+60+60 = 180 degrees not 360 degrees. Also, your scope trace clearly shows two signals not three. I work at 400Hz and I use three audio amps to provide isolation. And besides my target voltage is 48 vac which I step up to 11 5vac to get a clean signal you will need to clean up the source I assume ( your input is a PWM output from sort of a processor )  you can clean it up with a RC network.  this would be phase A, then you need to delay it 120 degrees for phase B and one more time to get the phase C.