I have developed a grid-connected single-phase PV inverter consisting of DC boost converter and a full bridge inverter separated with DC link. I noticed that 0.98 power factor gives better real power output than at unity power factor.
I've read that in UPS power supplies unity power factor can also cause inverter derating and generally there seems to be a practice in the industry to avoid unity power factor due to overcompensation/resonance and is best to stick with 0.96-0.98.
I can only add that the inverter is transformerless, the LCL filter is not ideally designed for switching frequency than what is being used and I notice MPPT is more stable at 0.98. Overall inverter current output is higher at 0.98, but still can't put my finger on why?
Hej,
Andres Julian Aristizabal: I beleive you mistook the efficieny (which caracterises the loss: output power divided by input power) and the power factor (which caracterises the relative shapes of the current and voltage: active power divided by Apparent power).
I would like to ask how do you load the inverter? Do you load it with a pure resistive load for a power factor one and with an inductive load with a power factor of 0.98?
That is how do you measure the power factor. Power factor for pure sine wave sources is cos phi. In case of the harmonics in the output waveform, how you will measure the power factor? I think that reactive loads affect more shaping in the output waveform.
I expect that the explanation of this observations lays on the the output voltage waveform.
Best wishes
Hej,
Power factor is defined by the ratio between active and aparent powers. It is not restricted to sinusoidal waveforms.
The cos(φ), where φ is the phase-shift between the sinusoidal voltage and the sinusoidal current is a specific case, valid for simple loads (R-L-C loads). In this very specific case: cos(φ) = power factor but the notion is much more general.
Regards.
Yes, sure Yoann, you are right the definition of the power factor is well known. I assumed that the source is normally a sinusoidal one and then the power factor will be cos phi.Then if the source is not a sinusoidal one, how he measured the power factor. As he said his filter is not ideally designed. i think the questions which i asked can help answer the question.
Thank you for your comments!
Indeed it is cos(φ) based on constant 50Hz LV source with no harmonics. I tried various load P/Q variations but PF is quite stiff and stays 0.98. I did not mention before that I don't use load to change the PF but inverter's own control reference.
Why current grows is understandable as Mohamed pointed out but that alone doesn't explain the power transfer discrepancy between theory and practice. At PF=1 I should be getting max real power. But that is not the case I get max real power at PF=0.98. In other words presence of lagging reactive power helps get 15% more active power. If I set reactive to zero, active power drops -15% too.
Perhaps it will help additionally to know that LCL filter was optimally designed for a mixed frequency inverter with one leg at 17.5 kHz and the other at 50Hz. However I use a single 10kHz frequency.
Dear Ognjen,
You have a grid connected inverter . So, your output voltage is the grid voltage and then you have to control the output current of the inverter to be fed to the network. Since your control circuit can not respond instantly on the set value of the current, but it takes a lag time to adjust the output current to the desired value, there will be some specific delay time between the network voltage and the injected current to the grid. This appears as an internal power factor. The reactive part of the current comes from the behavior of the current control loop of the inveter. From conceptual point of view if you want to track the current waveform with the grid waveform e.g. using a phase locked loop you can synchronize the two frequencies but there will be always some residual phase error . This may explain your observations.Now if you plot the grid voltage and the output current waveforms you will get the time delay and the phase differnce between the two and then this will be the effect.
You can test this hypothesis by changing the delay in the control loop. THEN IF THIS IS TRUE THE POWER FACTOR WHICH YOU GET WILL BE CHANGED.
BEST WISHES
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity