Synchronous Buck Converter with Negative Feedback Loop

Thread Starter

sers000

Joined Aug 5, 2019
25
Hello,

I am in the midst of creating a synchronous buck converter. The requirements of the converter is the following:
  • Vin = 30V - 50V
  • Vout= 3V - 5.5V
  • Io >=12 A
Now the power stage works perfectly with ideal as well as realistic components. After investigating that the power stage works, I started working on the closed loop feedback circuit following voltage controlled mode (duty controlled ) approach. When the VR1 is 1.06V and Vref is 5.5V, the compensator amplifier is outputting 4.44 V so the first half of the feedback seems working. I was not sure what the ramp voltage should be and the internet didn't help much to be honest. The main issue is that the output voltage varies greatly when the input voltage changes within the specified range. I have been trying to figure out this issue for the past month but not much progress so far. I will appreciate the help and thanks in advance
 

Attachments

ericgibbs

Joined Jan 29, 2010
11,095
hi,
Look at his rerun of your original, with the feedback loop open and fed by an increasing 'Vout error' voltage source, the Vout increases.!

E

For ref, this second image is with the test Voltage source sweeping down, note the Vout also going down.
E
 

Attachments

Last edited:

Thread Starter

sers000

Joined Aug 5, 2019
25
hi,
Look at his rerun of your original, with the feedback loop open and fed by an increasing 'Vout error' voltage source, the Vout increases.!

E

For ref, this second image is with the test Voltage source sweeping down, note the Vout also going down.
E
Thanks alot mate :)
 

Thread Starter

sers000

Joined Aug 5, 2019
25
hi,
Look at his rerun of your original, with the feedback loop open and fed by an increasing 'Vout error' voltage source, the Vout increases.!

E

For ref, this second image is with the test Voltage source sweeping down, note the Vout also going down.
E
Hey mate,
May you tell me how did you obtain the U1 amplifier's resistor's value and the V2 Reference voltage value?

Iam just trying to mathematically justify the importance of the addition of the extra amplifier.

Cheers
 

ericgibbs

Joined Jan 29, 2010
11,095
hi S,
My Post #2.

The input resistance is R6 and R8 in parallel, so the OPA gain is R9/(R6||R8)
G= 5k8/909R = 6.38

I selected R9 so that the Vout would be approx 5V

I added the extra OPA in order to invert the Vout error signal, without the inversion the Verr was driving Vout the wrong way

E
 

Thread Starter

sers000

Joined Aug 5, 2019
25
hi,
If you wanted to drop the extra OPA, it maybe possible to modify the original circuit in order to get the correct error sense.
E
Hey E,
I was trying to stick to one OPA and a Schmitt Trigger to obtain the required results, but as you mentioned before my Vout was increasing. I have followed Ti's formulas to obtain the cricuit's components. Basically im trying to go with a standardized circuit design to help me understand it but it did not go that well.
What do you think the problem was before If I wanted to use a Feedback whilst discarding the extra OPA?

Cheers
 

ericgibbs

Joined Jan 29, 2010
11,095
hi,
Tried modifying the original circuit, without the extra OPA, getting unacceptable results.
Will look at it some more, post if any success.

E
 

Thread Starter

sers000

Joined Aug 5, 2019
25
This is the waveform from before,

upload_2019-8-12_15-51-3.png

This is the waveform right now wich the following details:
upload_2019-8-12_15-58-0.png
For
Av = -R16 / (R13//R15) = Vo/Vin
I made Vo/Vin = 5.5/50 , then to find R16 , Vo(R13//R15)/Vin = R16 = 100 Ohms

Vref I made it 5.5V

Now I am getting less switching in the circuit. However the change in Vout is less consistent as Vin changes.
Vin = 30, Vout = 3.0
Vin = 40, Vout = 3.9
Vin = 50, Vout = 4.87

Even though the simulation is faster now and there is less switching behavior, why is the Error that big in the circuit even though its the same schematic?

Also, I really appreciate the time you spending on this, so thank you :)
 

Thread Starter

sers000

Joined Aug 5, 2019
25
Just for me to clear my doubt,
you dropped the extra OPA by inverting the reference and inputting it into the -ve terminal of the 2nd OPA?
 
Top