3
\$\begingroup\$

I have a PC consuming 400 watts hooked up to a 220 VAC automatic backup 1500VA 12V dc inverter since AC grid line is not 24/7 in my country.

The main problem is that sometimes when the AC power goes down the inverter successfully turns on and backup up power is provided to the PC with no problems. Other times the inverter makes a strange sound and the PC restarts.

Using a cheap DSO oscilloscope and an ACS712 module the problem seems to be high inrush current reaching 15 A during the switching time of the inverter which is around 12 milliseconds.

I have tried the following with no success:

1) A thermistor of 3.5 ohms for a certain time then shorting it with a relay. This did not limit the inrush current.

2) A 20-ohm high power resistor for a certain time then shorted it with a relay. This method limited the current but the PC restarted. Probably due to a voltage drop at this current level.

3) A solid state relay with zero crossing trigger to facilitate the slow rise of current. This minimized the inrush current but the problem remained.

4) An attempt for a soft switch using a triac and angle phase control Soft Start Traic

Is there any other way to make this work?

\$\endgroup\$
  • 1
    \$\begingroup\$ An NTC in series? A different power supply with less inrush? Increase value of 1). Tune the time constant and value for 2). \$\endgroup\$ – winny Jan 21 '18 at 16:46
  • \$\begingroup\$ Also, some inverters have built in soft start for this very reason. Expensive but easy solution. \$\endgroup\$ – winny Jan 21 '18 at 16:59
  • 4
    \$\begingroup\$ You should be using a UPS for this - as the power to the pc comes from the ups whether the grid is available or not so there is no switching issue. \$\endgroup\$ – Solar Mike Jan 21 '18 at 17:09
  • \$\begingroup\$ Your question makes no sense. The battery and inverter are always working to produce the output voltage and therefore there is no inrush as you describe. \$\endgroup\$ – Andy aka Jan 21 '18 at 17:35
  • \$\begingroup\$ @Andyaka The automatic backup probably has an integrated automatic transfer switch. So the inverter is not supplying power until incoming grid power fails. \$\endgroup\$ – mkeith Jan 21 '18 at 21:56
1
\$\begingroup\$

I don't know how to solve your AC inrush problem. But it may not be the only way to approach this.

The most common problem with inverters is that people underestimate how large the DC wiring needs to be. It is very possible that during startup, the DC current is high enough to cause excessive drop in the wires leading from 12V battery to inverter. If so, this drop could cause the inverter to fault out due to low DC input voltage. That is the first thing to check, since you have test equipment and knowledge of how to use it. Check the droop at the input terminals first. If it droops a lot during changeover, then work your way back to the battery. It is possible that the cabling is adequate, and the battery is weak or undersized. Measuring will help you sort it out.

If that doesn't fix it, another option is to use the inverter strictly as an inverter, and run the power supply from the inverter at all times. Buy a battery charger to keep the battery topped off from the grid. With this setup, there will never be any inrush. This is what I have in an outbuilding. I have an 80Ah battery serving a 35W AC load through a 600W inverter. An intelligent battery charger keeps the battery topped off at all times. It is worth getting a good inverter, because it will be running 24/7.

\$\endgroup\$
0
\$\begingroup\$

400 watts is little when the system is already running but way enough to jump cycles at the Thyristor at 1500VAC from the initial startup of the filtering prior to the DC transformers..
Try jumping the psu input filter straight to the rectification. The pfc choke at input may be adding some inrush pull which multiplied at the inverter low frequency may be having trouble with a reducedl transformer driving its own feedback loop producing some current burst at thyristor larger than the cycle itself. At a few cycles the current may peak beyond the transformer magnetic limit and find resistance bouncing back to the thyristor and maybe heating despite of the reboots. You may think compensating the thyristor in some way, instead of adding a inrush limiter, adding big coils before and after using some formula.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.