EEVblog > EEVblog Specific

EEVblog 1487 - Do Solar Micro Inverters Take Power at Night?

(1/1)

EEVblog:
How much power do Enphase and other solar micro inverters draw at night time when switched off? It's actually a very interesting question involving real and apparent/reactive power, the system topology, and whether your storage battery is on-grid or off-grid, and also its efficiency curve.
Let's measure it, do some calculations, and look at when it might be a problem. Buckle up Dorothy!

I'm upgrading my solar power system AGAIN: https://www.eevblog.com/forum/renewable-energy/solar-upgrade-again/

00:00 - How much standby power does the Enphase micro inverter take?
04:53 - Solar Analytics power measurement at night
07:49 - The actual real power consumption of the Enphase inverter is...
08:32 - So why is it drawing 1.1A at night time?
09:20 - Reactive and Apparent power
10:10 - The Power Factor
11:41 - You only pay for REAL power
15:27 - Power Factor Correction & Energy Saver SCAMS
17:29 - DaveCAD: How does the Enphase Microinverter work?
23:38 - Bidirectional energy flow and untility VAR control
27:55 - How many microinverters can you have?
29:02 - What if you have grid connected battery storage?
30:46 - Battery inverter switching losses
33:19 - I2R copper losses
35:29 - Why not disconnect with a relay?
39:20 - Conclusion
 

rs20:
At 14:00, when you point out the moment when the apparent power flips from negative to positive, I think you got a bit derailed by some weird buggy maths in the graphs you're looking at.

Apparent power is always positive. Whether you define it as Vrms * Irms (positive * positive), or Sqrt(RealPower^2 + ReactivePower^2), it's always positive. So for the graph to be displaying negative apparent powers, it's doing some gross meaningless thing whereby it flips the sign of apparent power based on the sign of the real power (or maybe reactive power?). Either way, you can tell it's totally sus information to work off of because it flips in sign, but the magnitude stays the same for another hour or so. Which means that the giant flip-flop of hundreds of VA of apparent power might be due to, e.g., the tiniest drift of the real power from -1W to 1W.

Anyway, with the benefit of hindsight, it would have been much more appropriate to look at the reactive or real power graphs, rather than looking at a meaningless sign flip in the apparent power graph. I know it sucks to get tricked by the tools you're working with; should really file a bug with the makers of the dashboard to get that fixed.

EEVblog:

--- Quote from: rs20 on August 11, 2022, 07:43:11 am ---At 14:00, when you point out the moment when the apparent power flips from negative to positive, I think you got a bit derailed by some weird buggy maths in the graphs you're looking at.

Apparent power is always positive. Whether you define it as Vrms * Irms (positive * positive), or Sqrt(RealPower^2 + ReactivePower^2), it's always positive. So for the graph to be displaying negative apparent powers, it's doing some gross meaningless thing whereby it flips the sign of apparent power based on the sign of the real power (or maybe reactive power?). Either way, you can tell it's totally sus information to work off of because it flips in sign, but the magnitude stays the same for another hour or so. Which means that the giant flip-flop of hundreds of VA of apparent power might be due to, e.g., the tiniest drift of the real power from -1W to 1W.

Anyway, with the benefit of hindsight, it would have been much more appropriate to look at the reactive or real power graphs, rather than looking at a meaningless sign flip in the apparent power graph. I know it sucks to get tricked by the tools you're working with; should really file a bug with the makers of the dashboard to get that fixed.

--- End quote ---

It physically flips because when it's positive the microinveretrs have been switched on by the DC solar input and are actively outputting power. When it flips this is the microinverter completely switching off and you are left with just the output capacitors where current flows in the physically opposite direction. This is what they are trying to show here.

rs20:

--- Quote from: EEVblog on August 11, 2022, 12:08:38 pm ---It physically flips because when it's positive the microinveretrs have been switched on by the DC solar input and are actively outputting power. When it flips this is the microinverter completely switching off and you are left with just the output capacitors where current flows in the physically opposite direction. This is what they are trying to show here.

--- End quote ---

I think you completely missed the point of my post. I'm saying that you're pointing at the apparent power graph saying "woah look at those 100's of VARs disappear/flip the moment the microinverters switch on", giving the impression that, e.g., the reactive power has plummeted, or that the power factor has substantially improved in that instant. When really, all that's happened is that the real power has just gotten past the threshold of  getting from -8W real power to 0.001W of real power, and some dumb graphing maphs rendering that as a huge swing of 100's of VARs of "apparent power" (which again, a sign flip of apparent power is a meaningless concept.) 

In other words, the graph is displaying a huge sign flip, just because the phase between the current and voltage changed from 91 degrees to 89 degrees. This is not a major event worth celebrating, the inverter is still almost perfectly capacitative.

You're saying "look at the effect of the capacitors disappearing" (or something like that), but it's still there in the graph, just positive rather than negative (while it's flatlining before a real considerable amount of sunlight arrives).

Could you please post a screenshot of the Real Power, Reactive Power, and Apparent Power, with the x axis zoomed in down to the "flip of apparent power" (one in the morning looks cleaner) plus/minus 2 hours or so? Or maybe some exported CSV data or something? It'll be easier to demonstrate the point I'm making with that data.

Navigation

[0] Message Index