Electronics > Projects, Designs, and Technical Stuff
Problem with In-rush current
(1/1)
LoveLaika:
You guys really helped me out on my buck converter circuit, so I thank you guys for your help. The circuit works, but now I face a problem. It works when there's no load connected, outputting positive and negative voltages.
However, the issue is when I add a cable with a load at the end, a small load that draws around 400 mA. Turning on my bench power supply at 12 volts, it goes into current limiting mode. Something is drawing too much current.
I tried reducing my input capacitors from 220uF to 4.7 uF on both regulators, and I added an NTC thermister in between my bench power supply and my board at the voltage input. I also increased my slow-start capacitors to 10 nF each, but no matter what I do, my bench supply goes into current limiting mode. This only happens when I power up my board with a load attached.
What else can I do to reduce the in-rush current? Would increasing the resistance at the current adjust pin of each regulator help? If it helps, my bench supply is a single supply bench supply which can go up to 3 amps at the output.
LoveLaika:
Well, turns out....the issue wasn't with the input or output capacitors (I think) nor did using an NTC thermister solved the problem.
Turns out, it was an issue with the switch current. I guess my values for resistors to limit the switch current was still too high. So, I switched out the resistors with larger ones to make the current limit near the IC's threshold (1A and 3A respectively), and...that seemed to work for now. The thing is, when I power my bench power supply, it takes a few seconds for it to go to 12 volts, like 3 seconds while staying in constant current mode. It doesn't immediately go to 12 volts when I turn it on. I wonder if this is anything to be concerned about
duak:
If I understand your problem correctly, the bench supply is current limiting during startup when your power supply is loaded. This is to be expected with switching regulators because when they are operating properly, they act as constant power loads to the input power source (the bench supply). That is, to maintain a particular output power, the input current will increase as the input voltage is lowered. Neglecting inefficiency and input operating voltage range, if the input voltage is halved, the input current would have to double to maintain the same power.
When the input voltage is first applied, it has to start from zero and then increase as the various capacitors charge. At some point the switching regulators will start and try to bring their output voltages to the correct values. If the input supply voltage is lower than nominal at this point, the input current will be higher and is clearly high enough to cause the bench supply to current limit and not further increase its output voltage.
The LM2679 has a soft start feature and you may wish to increase the delay time. I see on p.13 of the data sheet that there might be an output voltage overshoot if the soft start capacitors are in a certain rainge.
LoveLaika:
Thanks for your reply. I increased the capacitances from 2.2 nF to 10 nF for my soft start, but that didn't work out. I mentioned that I managed to 'solve' the issue by limiting the switch current via resistors. My resistors were allowing 4.5 to 6 amps of current on startup. Reducing it down to 1 and 3 amps for each regulator worked, though it took a few seconds for the power supply to 'settle' at the fixed bench voltage. From there, with my load, the bench was drawing 300 mA.
Navigation
[0] Message Index
Go to full version