Enphase Micro Inverter Operation

[ggunn]

What do you mean? She said my "simple" explanation in post#2 was "wrong", with nothing presented to explain her comment. Let's experiment... you are wrong?

I'll take a swing at this one. PV inverters are current sources*, and as such do not need to have a higher voltage output than the grid to "push" energy onto it. Yes, there is voltage drop between the inverter and the service resulting in a voltage rise at the terminals of the inverter, but the magnitude of this rise does not determine the amount of power output. You could, for example, run a parallel set of conductors from the inverter to the service and thus cut the voltage rise in half and the power output from the inverter would remain unaffected. You could theoretically run a superconductor line from the inverter to the service and reduce the voltage rise to zero, and the inverter would still export power.

* Since the output voltage of the inverter is clamped by the grid voltage, that makes it in reality a current source at constant voltage - a power source.

 

[jaggedben]

I see... and additionally Line 3 for 3? system connection.

Actually, no. Enphase's 3 phase system uses only the wiring to distribute the phases. A given individual inverter sees only two of the phase conductors. The other leg bypasses every third inverter.

 

[jaggedben]

Why would they do that if they could combine the 120 V voltage vectors? FWIW, traditional utility residential meters meter Line-Line and thus have slight metering errors when the voltage is unbalanced.

Stupid me, I don't know why I took Bill's comment to suggest they measure only one line-to-neutral measurement instead of both.

 

[Smart $]

I'll take a swing at this one. PV inverters are current sources*, and as such do not need to have a higher voltage output than the grid to "push" energy onto it. Yes, there is voltage drop between the inverter and the service resulting in a voltage rise at the terminals of the inverter, but the magnitude of this rise does not determine the amount of power output. You could, for example, run a parallel set of conductors from the inverter to the service and thus cut the voltage rise in half and the power output from the inverter would remain unaffected. You could theoretically run a superconductor line from the inverter to the service and reduce the voltage rise to zero, and the inverter would still export power.

* Since the output voltage of the inverter is clamped by the grid voltage, that makes it in reality a current source at constant voltage - a power source.

I agree for the most part, but the utility voltage of consideration would be that on the grid, i.e. the line side voltage of the service transformer. Superconductors or not, the impedance of the service transformer must also be figured into the scenario. And this also assumes there are loads served beyond the tie in point. If connected on the end of a dead-end run, the inverter has to overcome voltage drop to up-line grid loads. Anyway you look at it, the inverter has to change its output voltage to push out the maximum available PV power.

 

[Smart $]

... A given individual inverter sees only two of the phase conductors. The other leg bypasses every third inverter.

After looking closer, I see you are correct. :ashamed1:

 

[mivey]

Regarding your reply, while you may not need access to all three phases (though the M215 does have access to all three... when wired correctly ), you still need access to three points (e.g. L1, L2, N) of the system to directly establish any difference in phase angle.

Correct. Without additional info, it is unclear.

 

[mivey]

What do you mean? She said my "simple" explanation in post#2 was "wrong", with nothing presented to explain her comment.

In other words: her comment did not really say anything useful.

 

[mivey]

I'll take a swing at this one. PV inverters are current sources*,

I thought that was what Smart was getting at in his simple explanation.

 

[ggunn]

I agree for the most part, but the utility voltage of consideration would be that on the grid, i.e. the line side voltage of the service transformer. Superconductors or not, the impedance of the service transformer must also be figured into the scenario. And this also assumes there are loads served beyond the tie in point. If connected on the end of a dead-end run, the inverter has to overcome voltage drop to up-line grid loads. Anyway you look at it, the inverter has to change its output voltage to push out the maximum available PV power.

It may be a semantic argument, but a sticking point for many folks is that a PV inverter behaves as a current source. It (virtually) pushes the same amount of current out irrespective of the load it sees. By contrast, most sources we deal with are voltage sources (like a battery or grid power), which keeps voltage (virtually) constant and the current they deliver depends on the load.

 

[Smart $]

It may be a semantic argument, but a sticking point for many folks is that a PV inverter behaves as a current source. It (virtually) pushes the same amount of current out irrespective of the load it sees. ...

No argument. I agree with you. But as a current source, it has to develop a voltage high enough to push that current into a charged system. Let's say the PV system has 2400W available. Local loads are all off. Service voltage is 240V and the inverter "sees" a system impedance of 5 ohms to the grid (includes service transformer). What voltage must be developed at the inverter's terminals to output the entire 2400W (ignoring I?R losses) onto the grid?

 

[ggunn]

No argument. I agree with you. But as a current source, it has to develop a voltage high enough to push that current into a charged system. Let's say the PV system has 2400W available. Local loads are all off. Service voltage is 240V and the inverter "sees" a system impedance of 5 ohms to the grid (includes service transformer). What voltage must be developed at the inverter's terminals to output the entire 2400W (ignoring I?R losses) onto the grid?

I am not saying that voltage rise does not exist, but in an ideal circuit a current source is perfectly happy to deliver current with a zero voltage differential. Yes, I know that the real world is not ideal and a PV powered inverter is not an ideal current source, but the distinction is, IMO, an important one. It is central to understanding the answer to what is probably the most often asked question about solar power - why a grid tied system must not keep running when the grid goes down. It's not just because it's illegal.

Current sources can be hard for some folks to wrap their heads around because virtually all the power supplies we come into contact with (batteries and grid power) act as voltage sources and deliver current on demand.

 

[Smart $]

I am not saying that voltage rise does not exist, but in an ideal circuit a current source is perfectly happy to deliver current with a zero voltage differential. Yes, I know that the real world is not ideal and a PV powered inverter is not an ideal current source, but the distinction is, IMO, an important one. It is central to understanding the answer to what is probably the most often asked question about solar power - why a grid tied system must not keep running when the grid goes down. It's not just because it's illegal.

Current sources can be hard for some folks to wrap their heads around because virtually all the power supplies we come into contact with (batteries and grid power) act as voltage sources and deliver current on demand.

Okay

 

[tallgirl]

Current sources can be hard for some folks to wrap their heads around because virtually all the power supplies we come into contact with (batteries and grid power) act as voltage sources and deliver current on demand.

Uh, no. Grid tied systems can, and have, run without the grid. The AC output stage matches volts and cycles to the load and merrily keeps on putting out power. This was not unheard of back before UL1741 came along.

For some gear, if you can produce a load that looks like the grid (load, not "source"), it will also merrily keep making power.

 

[Smart $]

Uh, no. Grid tied systems can, and have, run without the grid. The AC output stage matches volts and cycles to the load and merrily keeps on putting out power. This was not unheard of back before UL1741 came along.

For some gear, if you can produce a load that looks like the grid (load, not "source"), it will also merrily keep making power.

Uh, not exactly.


p)

 

[ggunn]

Uh, no. Grid tied systems can, and have, run without the grid. The AC output stage matches volts and cycles to the load and merrily keeps on putting out power. This was not unheard of back before UL1741 came along.

For some gear, if you can produce a load that looks like the grid (load, not "source"), it will also merrily keep making power.

With all due respect, I don't see how this could be, but I am always looking to expand my understanding. Can you point me to a reference?