Measuring AC energy flow direction

[RCinFLA]

Now, if you cut that resistance in half by increasing the conductor size or paralleling conductors, you cut that voltage drop in half, but the current from the inverter remains the same. Keep cutting the resistance in half, the current stays the same. Take that resistance all the way to zero (OK, not realistic but imaginable); the voltage differential becomes zero but the current still flows at the same rate.

Not precisely, for a given illumination level there is a certain amount of power to be pushed to grid. If GT inverter sees a lower grid voltage it will deliver more current to push the same amount of power into the grid. You obvious have to be within the grid voltage tolerance acceptance range.

Some GTI will oscillate on and off grid if there is too much wiring resistance loss to inverter. It sees normal grid voltage before connecting. It connects and start to push power to grid. With too much wire resistance the current push raises the voltage the inverter sees. If it raises too high, out of high end limit, it disconnects from grid. Grid voltage seen by inverter then drops to normal grid voltage and after several minutes it tries to reconnects. A good grid testing algorythm will detect too much voltage fluxuation and disconnect shortly after a connection attempt to grid made.

[inetdog]

Mmm now you've confused me- posts from inetdog (#24) and boB (#16) suggest otherwise- that the relative phases of V and I waveforms DO represent power flow direction. But you're a PV engineer... arrrrghh
I appreciate the other solution which would involve measuring in two places (at the inverter and at the loads) but it does complicate the wiring.. and would rather have one point of measurement for simplicity's sake.

Disclaimer: I am a Physicist, so I will (almost) always be right about the theory, whether I manage to get something to work or not. An Engineer, on the other hand, will give you something that works, but may have trouble explaining it.

"In theory there is no difference between theory and practice. In practice, however..."

What ever explanation you follow, you will have to have two measurements to be able to tell which way the power is flowing. The combination of a DC clip-on Ammeter and a clamp-on capacitive voltage sensor will do the job. Measuring the current (including direction) at the same time you measure the voltage will do the job. It does not matter whether you measure the current with an Ammeter which senses the instantaneous direction of the current and correlates that with the voltage or you measure the current by looking at the voltages at two points separated by a known resistance. The result of that measurement will be the same. But a meter that just tells you the magnitude of the AC current and another meter that just tells you the magnitude of the AC voltage will not be able to tell you what you need to know.
If the current is flowing towards the grid when the voltage is + and away from the grid when the voltage is minus you are selling. If the current is flowing away from the grid when the voltage is + and towards the grid when the voltage is -, you are buying.

Now, just for fun, I will take a crack at ggunn's and boB's explanations directly:

"I don't believe that the relative phases of the voltage and current waveforms will have any relationship to which way the current is flowing,..."

If you follow the idea that the voltage out of the inverter has to be higher than the grid voltage to push power out, you need to recognize that the voltage from the inverter also has to be in phase with the voltage from the grid. That is the simple constant side of the formula.
That also means that if the magnitude of the voltage from the inverter is slightly larger than the magnitude of the voltage from the grid and they are in phase, then the + peak from the inverters is slightly higher than the + peak from the grid. And the - peak of the inverter dips slightly lower than the - peak of the grid. (So far so good?)
What that means, by Ohm's law, is that during the + cycle current is flowing toward the grid and during the - cycle current is flowing away from the grid.
If you substitute a resistive load in place of the inverter, the voltage at the load will be slightly lower at the + peak than the grid voltage (otherwise no current would flow, goes the argument.)
So if the current flows away from the grid during the + cycle and toward the grid during the - cycle, power is flowing into the load.
The output of a simple AC ammeter is only going to tell you the magnitude of the current. A current transformer, as in a clamp-on ammeter or in some utility meters, knows exactly which direction the current is flowing at any moment of time, but the AC meter does not bother to tell you. If you look at both waveforms on an oscilloscope, you will see that in one case the current and the voltage are in phase and in the other case current and voltage are 180 degrees out of phase. You can try the experiment yourself if you do not believe me.

If you can measure the load power and the output power of the inverter (which can be done with two power meters which do not need to run backwards as both BB and ggunn stated) you will know whether you are buying or selling. If the utility will not pay you for excess you send them (or worse, charge you for it) you will need to make sure that you never overproduce, even if that means adding extra load. (You cannot throttle down a grid-tie inverter except by limiting the DC input to it.)

[stephendv]

Lazza,

Have a look at this discussion here: http://www.navitron.org.uk/forum/ind...c,13808.0.html
The simplest solution seems to be to buy a standard electricity meter that only counts in one direction, and then wire it backwards. So when you're a net importer it doesn't move, but when you start exporting it'll start counting.

[boB]

Well, consider this: Say you have a GT inverter connected to the grid. The current flowing through the resistance in the conductors between the inverter and the grid results in a voltage drop in the direction of the grid, in turn resulting in a higher voltage at the terminals of the inverter, true enough. Now, if you cut that resistance in half by increasing the conductor size or paralleling conductors, you cut that voltage drop in half, but the current from the inverter remains the same. Keep cutting the resistance in half, the current stays the same. Take that resistance all the way to zero (OK, not realistic but imaginable); the voltage differential becomes zero but the current still flows at the same rate.

Superconductor ! Perfectly legal. Ya got me there GG !!

OK, so imagine you are looking at the voltage outputs that WOULD be present and then connect the GT to the grid via superconducting wire.
The GT's voltage would be slightly higher than the grid in order for current to flow into the grid when you finally connect them together.

Otherwise, no current (or power/energy) would flow into or out of the grid from/to the GT inverter if they were exactly the same voltage.

(Yes, I know, techincally with a current source the voltage when disconnected would be off the map for a true current source but voltage source
would certainly behave this way. And, not all GT inverters are current sources anyway)

This example along with the pictures I posted were only to help illustrate the concepts of direction of current that was
asked and was not intended to get into the details of current sources and stuff like that. It usually works well as an illustrative aid.

boB

[Cariboocoot]

Negative Voltage. Reverse current flow. Peak Voltage. RMS Voltage.

Lions and tigers and bears! Oh my!

Looks like another case of theoretical physics colliding with practical application.

And that's what makes sparks* fly, right?

(*And I don't mean Dave. )

[boB]

Negative Voltage. Reverse current flow. Peak Voltage. RMS Voltage.

For AC, think negative thoughts !!

[Cariboocoot]

For AC, think negative thoughts !!

I always think negative thoughts, Bob; I'm a dreadful cynic!

But, when it's negative on one wire it's positive on the other wire so the net effect must be zero current flow! That means inverters don't actually back feed the grid at all and none of this stuff really works!

It's all in our heads!

And before Bill or Niel or Tony jumps on me I promise to stop razzing everyone about this now.

[nsaspook]

Mmm now you've confused me- posts from inetdog (#24) and boB (#16) suggest otherwise- that the relative phases of V and I waveforms DO represent power flow direction. But you're a PV engineer... arrrrghh

I appreciate the other solution which would involve measuring in two places (at the inverter and at the loads) but it does complicate the wiring.. and would rather have one point of measurement for simplicity's sake.

Knowing the relative phases of V and I waveforms can't show power direction unless there is a point of reference "ie.. the Poynting vector".

Here is what you need to establish the direction of power flow(at a single point):

A reference direction
The angle between the voltage and current



P is real power
Q is imaginary power
S is the complex power

The direction of P can be left or right depending on the point of reference. In this case the grid is the point of reference.

http://en.wikipedia.org/wiki/Electric_power

[inetdog]

For AC, think negative thoughts !!

I think we can all agree that if you can buy power for $.06 per KwH, that is an *absolute value*.

[Ken Marsh]

Good Grief Guys, lets give lassa something he can work with.
Here is a circuit for an indicator which seems like it should work.
That is if I can figure how to post a .jpg
I looks like we have a lot of electrical / electronic types on this board.