recently design P.S with this IC, after i power-on , the output voltage is varying , from my setting 380V slowly drop to 372V, then IC function to give pulse to boost output again back to 380, of course, it will overshoot to around 390V. i try to adjust voltage loop, but it seems not working. does anybody meet same problem before? thanks, anyone can give me some suggestion.
What is your load and output capacitance?
What is your input voltage and frequency (AC, RMS)?
thanks.
i havent put any load. Cout is 680uFx3, nearly 2mF
input voltage from 100~240 fsw=50Khz , and inductor i give 360uH for each phrase.
any critical parameter need ?
Luo,
Pleaee send the schematic and scope shots of the performance.
Regards,
John
John Bottrill
hi John
thanks for your reply.
i have fixed this problem by increase capacitor of Vsense and Vinac, use 3.3nF capacitor connected between each pin and GND.
now the problem is load.......
i have monitored Vsense, can see some 25hz noisy there. i am trying to reduce it. later i post waveform.
any suggestion for load debug ? thank you
I Would still need to see the schematic before I could offer advise on how to fix the supply.
1.Rimo=20k Rsyn=150k c18=c2=3.3nf c5=2,2nf c20=10nf Rrt=150k Rdmax=120k, all other paramete remains same. my CT is 200:1
the pin color is feedback signal, at no load condition.
thanks, John
Under no load and light load conditions if you don't have enough boost capacitance you could be hitting OVP. Also if you have to much boost ripple voltage you could be tripping the slew rate comparator for an under voltage. This is when Vsense is below 2.8V you can see this comparator in the block diagram on page 10. We recommend that the boost ripple voltage be less than 5%. http://www.ti.com/lit/gpn/ucc28070 You might want to check the amount of output capacitance in your design with the excel design tool. http://www.ti.com/litv/zip/sluc114
hi M OLoughlin, i had checked my Capacitance, its enough for 5% ripple.
my problem now is load, i have tried parameters from excel design tool, but it doesn't work. now i am trying to tune it.
here i put some waveforms , if you can help me to figure out the problem.
pink color is Vinac, is this one accepted or not? when load, it will become better.
green is feedback, has some oscillation.
pink and green is from pin CAO,
yellow is input current, oscillation there. cannot go high load.
much thanks.
Your VINAC pin looks heavily filtered but should be able to work.
When I look at your waveforms it looks like the voltage loop/and or the current loop are not compensated correctly. I agree with you that it is breaking into an oscillation of some sort. You need to double check your voltage loop and current loops. The voltage loop should cross over at roughly 10 to 20 Hz. If you used the excel design tool it would have calculated the proper voltage loop compensation. http://www.ti.com/litv/zip/sluc114 It also gives a good idea on how to compensate the current loop. The excel design tool does go along with a pretty good application note which you can find at the following link http://www.ti.com/litv/pdf/slua479b
It might be worth while to get a hold of 300W EVM to evaluate you should be able to order one from the TI web site.http://www.ti.com/tool/ucc28070evm This way if you needed to you can evaluate the voltage amplifier and current amplifier to see how it behaves.
I found your schematics and waveforms hard to read in the forum post. In the future can you send them in a PDF so they can be blown up and evaluated easier?