# How do micro-inverters stay in sync (phase-lock) in a grid-tied residential solar panel system?

I'm looking for a little more detail on how the all the micro-inverters sync up to the phase on the 60 Hz grid.

In communication circuits phase/frequency locking is done with a PLL (phase lock loop). I assume something similar is used in the micro-inverter. That is, each micro-inverter has a PLL to achieve phase and frequency lock. But this can't be an analog PLL, as the loop filter components would be huge (big capacitors, ...) So I guess a company like Enphase uses a digital PLL to do the phase/frequency locking. Does this sound right?

Or does Enphase use something simpler, like a zero-crossing detector, with additional information about the slope (at the zero-crossing) to maintain phase/frequency lock with the grid?

• This kit from TI will help you a lot. (I have one.) It includes info on MPPT and Software PLLs.
– jonk
Commented Aug 25, 2020 at 6:40
• A 60 Hz PLL needn't have huge capacitors. Commented Aug 25, 2020 at 9:29
• When you turn a hand crank, you don't need any special system to keep your hand in phase with the crank. It just happens when you hold the handle and try to apply a force. I like to think of all the generators in a country putting their shoulders to the grid and pushing it along. Commented Aug 25, 2020 at 10:53

It happens automatically when the inverter tries to synthesise a 'negative resistor'.

Consider how you take power out of a grid. You connect a resistor to it. Current flows proportional to the voltage. If you want to draw a certain power, you connect a certain value resistor.

The same thing happens in reverse when a grid-tied inverter wants to feed a certain amount of power into the grid. That amount of power is chosen by, let's say, the present value of its MPPT. From the power, and the nominal line voltage, we can derive a negative resistance which will deliver that.

It measures the instantaneous terminal voltage, then forces a current of V/-Rneg into the grid. As the voltage changes throughout the cycle, so the current forced into it changes. As it looks like a resistor, the current is always in phase with the voltage, and the power factor is 100%. This current for voltage computation can be done in analogue components, or using a fast enough DSP.

• This is one possible solution to the problem... Commented Aug 25, 2020 at 7:47
• Would it be an accurate paraphrase to conclude that the non-technical explanation is "each microinverter synchronizes itself?"? Commented Jul 18, 2022 at 0:04
• @Thagomizer no, it has to synchronise to something, and that something is the local grid voltage. I like tomnexus' comment above to the question. Think of the connected grid voltage as a hand-crank. Every microinverter pushes the hand-crank, aka adds current to the grid, by pushing in the correct direction on the crank. It doesn't change the position of the crank, that trundles round almost unaffacted, but it does add its own push. Rather like each load, as I talk about in my answer, draws current from the grid, trying to oppose the motion of the hand-crank. Commented Jul 18, 2022 at 4:28
• @Neil_UK Doesn't that mean that each microinverter is synchronizing itself, independently, to the grid? Commented Jul 19, 2022 at 18:18
• @Thagomizer yes, each has to synchronise to something, and that something is the local grid voltage. Commented Jul 19, 2022 at 19:37