Electronics > Projects, Designs, and Technical Stuff

Oscillation in PSU simulation

<< < (7/7)

Zero999:

--- Quote from: xavier60 on April 19, 2020, 10:14:13 am ---
--- Quote from: Zero999 on April 19, 2020, 09:51:18 am ---
--- Quote from: xavier60 on April 18, 2020, 11:22:57 pm ---Would the voltage overshoot be acceptable with 100 milliohms for the CS resistor and 1A maximum load?

--- End quote ---
Are you asking the original poster or me?

I don't see the need for such a low value current sense resistor in a linear power supply such as this. 1V or so of additional voltage drop, at full load should be acceptable. Going for very low values will increase the error, from offset voltages, especially at lower currents.

--- End quote ---
I was asking you since you already had it in a simulator. Maybe the voltage overshoot will be acceptable with an output capacitor added and faster compensation.

--- End quote ---
The overshoot is worse, at lower voltage settings. Adding some anti-slew diodes, along with the associated feedback network helps.

 PSU Instr Amp.asc (4.85 kB - downloaded 66 times.)

Here's what it looks like, with the output set to 5V, without R7, R8, D3 & D4.


And with R7, R8, D3 & D4.


Note, I set the current limit to 1A, as that's what the original poster's schematic does.

xavier60:
I wonder if  Kleinstein's idea will work in this topology, C2 gets connected to the ORing node rather than to the CV op-amp's output.
The mod made a big improvement to my design but it's the current sourcing type.
With C2 connected to the ORing node the CV op-amp will immediately get miller feedback as soon as the overload is removed allowing the output voltage to rise.
The mod will also cause a large negative going transition at the CV op-amp's Inverting input as the CC op-amp pulls down.
There could be other problems that I'm not seeing yet.

Zero999:

--- Quote from: xavier60 on April 19, 2020, 01:23:46 pm ---
I wonder if  Kleinstein idea will work in this topology, C2 gets connected to the ORing node rather than to the CV op-amp's output.
The mod made a big improvement to my design but it's the current sourcing type.
With C2 connected to the ORing node the CV op-amp will immediately get miller feedback as soon as the overload is removed allowing the output voltage to rise.
The mod will also cause a large negative going transition at the CV op-amp's Inverting input as the CC op-amp pulls down.
There could be other problems that I'm not seeing yet.

--- End quote ---
Oh I missed that in the other thread. That's the trouble when there are so many posts. It's difficult to read all of them in detail.

 PSU Instr Amp.asc (5.23 kB - downloaded 67 times.)

Yes that works quite well. I was a bit worried about oscillation, but I also plotted the output of the op-amp and it's better than I thought it would be.

Navigation

[0] Message Index

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod