TOP260EN multi negativ voltage design stability.
I need to design a power supply with 3-negative -96V,-48V,-24V, +12 V voltage and one positive.
The power supply must provide an output current of about 2A for any of the outputs.
(To be more precise, then 2A - is the sum of the currents on all outputs.
For example:-96V - 1.5A,-48V - 0.5A,-24V - 0A.
Another example:-96V - 0A,-48V - 1.5A,-24V - 0.5A. And so on ...)
To use the software, I decided to design a power supply with 3 positive voltages and one negative.
After that, I changed the polarity of the primary winding and the rectifier diodes.
The resultant scheme I have attached as PowerInt.pdf.
I made the first power supply and became its test.
Unfortunately, the power supply begins to become unstable if connect even a small load to the voltage -96V.
It begins to turn on and off with the sound of falling drops.
(If the load is connected to any other outlets - everything is stable)
The first thing I tried to do - increased the capacity of the capacitor C15 (connected in parallel 330mkf)
and increased the resistance of the resistor R11 twice. (See PoverInt.pdf)
I'm not sure that this is a good way.
Could you suggest any ways to ensure the stability of the power supply?
Comments
Thank you very much for the advice.
I'll research the waveform voltage and describe in detail the situation later.
I wanted to draw attention to three things:
1) 96В no load - about 130V. (I guess try damping the diode.)
2) The instability begins at a very low load through 96В. I connected the 10K resistor.
3) The instability does not depend on the load on other outlets.
Thank you very much for the advice.
I'll research the waveform voltage and describe in detail the situation later.
I wanted to draw attention to three things:
1) 96V no load - about 130V. (I guess try damping the diode.)
2) The instability begins at a very low load through 96V. I connected the 10K resistor.
3) The instability does not depend on the load on other outputs.
I apologize for the incorrect sent of my previous answer.
My waveform voltage drain and the control voltage.
Load: 96В - 1.8K; 48B - 68R.
Capacitor C15 increased by 330mkf, or not working.
I put waveform from analog and digital oscilloscopes as digital oscilloscope can not display the full scale signal, But tha digital is more accurate.
The oscillograms show that impulses "trembling" - their image is blurred.
It may be the cause unstable operation is noise on the control pin of the chip?
Your drain waveform looks normal. After the output diode stops to conduct, the ringing is there. What is waveform of the other channel? Can you please add feedback on -96V?
> What is waveform of the other channel?
I do not understand what it means the "other channel".
In my design of single chip TOP260EN generates all 4 output voltages.
Please look at the scheme PowerInt.pdf.
Perhaps you mean the voltage across the secondary winding of the transformer?
I now have only a waveform of the channel - 48V. (Others will be able to send on Monday.)
> Can you please add feedback on -96V?
You mean the feedback from the two channels?
Since the output voltage is negative, I think, for this we need to use two SFH615A and two chip TL431. Is that true?
It would be very good. This further helps to avoid exceeding the maximum level of output-96V.
I really like this idea!
One strange thing.
In the beginning I tried to synthesize the project, using the feedback on the channel-96V. But I could not finish the synthesis of the project.
Then, for the feedback I chose-48V. And all was well.
Now I again tried to select-96V, and the project was successfully synthesized.
The first thing I'll do on Monday - I'll try to switch the feedback on-96V and see how the power supply will work. After that, I'll think about the feedback from the two outputs.
Another thing. I am concerned about jitter pulses at ostsilogramme and audible noise from the transformer. Is this normal?
Thank you very much for the good advice!
I connected the feedback on 96V.
The power supply is now working well.
Thanks for the recommendation!
Begin checking with different loads ..
I finished this project and all works well.
I changed the TL431 on the Zener diode, and added soft start circuit.
But on a loaded output, I see a very low-frequency small amplitude oscillations.
The frequency is about 220Hz and amplitude is about 30 mV.
(48V output 1A loaded.)
I tried to change all elements of the feedback loop - it does not affect.
It looks as if the TOP260 generates these signals itself.
Why?
On the picture: 5ms/div, 50mV/div
Attatched picture^
Please keep R11 and C15 at recommended values as they are related to control loop. You situation is probably auto-restart and is not an unstable issue but I don't have enough information from your description. I noticed that you have AC stack for secondary windings. So when you load the -96V only, the -96V will be lower than it should be (I mean the absolute value), thus the -48V is also lower due to the cross regulation issue. When your -48V is always below 48V (I mean the absolute value), there will be no feedback signal and TOP260EN will be at auto-restart mode. Please check the Drain waveform to see if TOP260 is doing auto-restart. Please verify the waveform of -48V and the voltage across opto transistor.