Why PV-Produced AC-Power Goes to the Loads before Grid-Provided AC-Power?

[LATTC Student]

I am still baffled how this part works.

One simple explanation stated that PV goes to the loads first because electricity takes the least resistive path (vs higher path thru transformer and grid). Another one mentioned that the PV inverter monitors the incoming utility line voltage, syncs to that voltage to maintain grid stability, and increases it's own output voltage that that above the utility voltage thus forcing the power to the local loads first. A third idea mentions the PV inverter is connected to the grid through a transformer. This offers an impedance. Do microinverters have transformers in them?

There seem to be conflicting theories. or maybe just my lack of understanding the tie-in

I do get the concept why the surplus goes back to the grid as the inverter is just a current source, they don't provide voltage and frequency reference by themselves. Instead they measure the (utility) voltage and freq, synchronize with it, and delivery the power output in synch with utility

 

[wwhitney]

What would it mean for the PV power to go to the grid and for the local loads to draw their power from the grid?

There's only one service, one connection to the grid. So that service is going to have power flowing through it in one direction or the other, either (PV power - local consumption) to the grid, or (local consumption - PV power) from the grid.

In the hypothetical superposition state of current from the PV going to the grid simultaneously with current from the grid going to the local loads, those currents are in opposite direction and will partially cancel. The only observable is the net current and associated net power flow.

Cheers, Wayne

 

[Carultch]

I am still baffled how this part works.

One simple explanation stated that PV goes to the loads first because electricity takes the least resistive path (vs higher path thru transformer and grid). Another one mentioned that the PV inverter monitors the incoming utility line voltage, syncs to that voltage to maintain grid stability, and increases it's own output voltage that that above the utility voltage thus forcing the power to the local loads first. A third idea mentions the PV inverter is connected to the grid through a transformer. This offers an impedance. Do microinverters have transformers in them?

There seem to be conflicting theories. or maybe just my lack of understanding the tie-in

I do get the concept why the surplus goes back to the grid as the inverter is just a current source, they don't provide voltage and frequency reference by themselves. Instead they measure the (utility) voltage and freq, synchronize with it, and delivery the power output in synch with utility

It is more of a matter of bookkeeping than reality, to think that the inverters "first feed the loads, and then export to the grid". The reality is that both of these events are happening simultaneously. Given an inverter that is producing 6kW, with site loads that consume 4 kW, the net result is that 2 kW are being exported through the service point to the rest of the grid. Your immediate neighbors will be consuming it.

No matter what kind of meter you have, your meter would be unable to tell the difference between 2 kW being exported, vs the alternative of ALL 6 kW being exported simultaneously with all 4kW being imported. All your meter can measure, is the net power at any instant in time at its location. And the reason for this, is that current itself doesn't simultaneously flow in both directions at once. Current that "attempts" to do this, will cancel out, and all you have left is the net current.

"Net" is a special kind of totaling, meaning you consider direction and sign (positive vs negative), as you add up numbers. Your meter measures consumption minus production, and adds this up as it accumulates through time. As will any other method of measuring power through any point within a circuit.

 

[jaggedben]

Imagine two batteries of the same voltage. Imagine a wire connecting their positive terminals to each other, and another wire connecting their negative terminals to each other. What you have now is two wires with a voltage between them, acting like a bus along which you can connect loads at any point you wish. The loads as a group will draw power roughly equally from both batteries, as long as they continue to provide the same voltage. If one battery's voltage drops lower, the other will supply more of the power to the loads. If you need to better understand this part then google Ohms Law as a start.

Now imagine a meter at the midpoint of your bus wires. If more of the loads drawing power are to the left of the meter, then the meter will show power coming from the battery on the right going to the loads on the left of it. If more of the loads are to the right of the meter, the meter will show the power from the battery on the left going to the right. If the loads are divided equally on each side of the meter, the meter won't show any power flow because each battery supplies the loads on its own side.

The situation with grid and the solar inverter and utility meter is really the same concept. For our purposes in this discussion, the fact that there's alternating current with frequency and power factor is just details.

There's one final thing to understand, which is that in most cases the grid and solar inverter do not match voltages quite like our theoretical batteries above. The grid is supposed to keep its voltage as close to nominal as possible. Meanwhile the solar inverter actually outputs a voltage slightly higher than the utility, in order to ensure that all its available power is outputted, thus maximizing solar production. (Many inverters can also actually be configured to respond to a meter and to raise or lower their voltage, and thus their power output, in order to keep the power flow at the meter as close as possible to a chosen number. This is common for battery inverters and can also be used where exporting power is not allowed or is restricted to a certain amount of power.)

 

[wwhitney]

Meanwhile the solar inverter actually outputs a voltage slightly higher than the utility, in order to ensure that all its available power is outputted, thus maximizing solar production.

Is this greater voltage exactly equal to what is necessary to overcome the voltage drop along the path to "the grid"? Or is it + epsilon?

E.g. say the utility no-load voltage at the meter is exactly 240V, and the PV production happens to exactly match the house load at some instant. I would think the voltage at the meter would be exactly 240V at that instant, which would mean no + epsilon.

Cheers, Wayne

 

[LATTC Student]

Thanks all for the responses, especially being so timely. The third response, from JaggedBen, really reinforces the first concept mentioned above and was at the root of the question -

(Why PV-Produced AC-Power Goes to the Loads before Grid-Provided AC-Power Does?) I left that word out and maybe formed the wrong question.

Namely "PV inverter monitors the incoming utility line voltage, syncs to that voltage to maintain grid stability, and increases it's own output voltage that that above the utility voltage thus forcing the power to the local loads first" is in line with "Meanwhile the solar inverter actually outputs a voltage slightly higher than the utility, in order to ensure that all its available power is outputted, thus maximizing solar production". So I am going to subscribe to this concept as my final understanding.

Additionally, I now have a clear understanding of the how the No-Export grid profile works based on the comment " Many inverters can also actually be configured to respond to a meter and to raise or lower their voltage, and thus their power output, in order to keep the power flow at the meter as close as possible to a chosen number". . . .that number being the at-the-moment consumption meter read. This explains the graphical view of my "tight" consumption/production graph using the CA Rule 21 No-Export grid profile. My capable PV generation is ~35% higher than is actually produced (or consumed) based on this microinverter setting

There were some other golden nuggets in there as well so thank you all who replied to my post. I do understand the metered "net" current that Wayne and Carultech wrote about.

I have PV fundamentals Lecture and Lab next semester and I am trying to understand the basics of DG renewable power and the variables in a rapid shutdown system with those DG power "supplies" as my starting point. You all helped a lot, thanks again.

. . . .and yes JaggedBen, I have a vague theoretical understanding of Ohm's law knowledge :O 1st semester school commitment

 

[Carultch]

Additionally, I now have a clear understanding of the how the No-Export grid profile works based on the comment " Many inverters can also actually be configured to respond to a meter and to raise or lower their voltage, and thus their power output, in order to keep the power flow at the meter as close as possible to a chosen number". . . .that number being the at-the-moment consumption meter read. This explains the graphical view of my "tight" consumption/production graph using the CA Rule 21 No-Export grid profile. My capable PV generation is ~35% higher than is actually produced (or consumed) based on this microinverter setting

What is really happening, is that the meter is sending a feedback signal to the inverter, to limit its power output. The inverter already has a power limitation feature in its firmware by design, in order to limit its production to its nameplate rating. The power feedback feature can dynamically adjust this setpoint, to values lower than its default full power setting, as it adapts to loads.

The way the power limitation feature works in general, is it shifts the DC voltage away from the Vmp/Imp sweet spot on the I-V curve, toward the open circuit condition. At Vmp/Imp, the array performs as best as it possibly can, which is whey the inverters are set up to seek this spot. At open circuit voltage (Voc), the array absorbs all sunlight as heat, because it can't produce any current with that voltage, when the current has nowhere to go. Shifting from Vmp to Voc, means less power leaves as electricity, and more power remains in the panels as heat.

 

[LATTC Student]

What is really happening, is that the meter is sending a feedback signal to the inverter, to limit its power output. The inverter already has a power limitation feature in its firmware by design, in order to limit its production to its nameplate rating. The power feedback feature can dynamically adjust this setpoint, to values lower than its default full power setting, as it adapts to loads.

The way the power limitation feature works in general, is it shifts the DC voltage away from the Vmp/Imp sweet spot on the I-V curve, toward the open circuit condition. At Vmp/Imp, the array performs as best as it possibly can, which is whey the inverters are set up to seek this spot. At open circuit voltage (Voc), the array absorbs all sunlight as heat, because it can't produce any current with that voltage, when the current has nowhere to go. Shifting from Vmp to Voc, means less power leaves as electricity, and more power remains in the panels as heat.

Whoa sir. There is a lot in that statement for me to absorb at once. I will reread a few times and comment further, thank you. I think there is something in there. And this is in reference to the "grid no-export" option or just synchronizing with the grid (meter) in general)? And does the additional heat generated (by suppressing the current output) cause any issues for the modules shore or long term?

 

[winnie]

I suggest you save questions about how load affects PV panel life for a different thread and focus on the power flow issue.

A PV panel in given sun conditions has its maximum power output, but can always produce less than this maximum. The actual power output depends upon the current draw by the inverter.

If the inverter is set up properly it can choose to draw less power from the panel and provide less AC power. Why it might do this depends upon external control input such as export limits, playing nice with a generator, etc.

Jon

 

[Hv&Lv]

Now to throw a little fuel on the fire…

An invertor monitors the voltage and sends power back to the grid by being just a touch above grid voltage, say 1 volt.
the utility is at 124.5 volts. The invertor sends power out at 125.5.
you have a neighbor that also has a solar install. What does his invertor see?
it’s not long before the grid sees voltages that are out of compliance and your invertors shut down about 255V, stopping production..

 

[electrofelon]

it’s not long before the grid sees voltages that are out of compliance and your invertors shut down about 255V, stopping production..

But that's your (utility's) problem

 

[Hv&Lv]

But that's your (utility's) problem

Not really… we can refuse to allow grid tie if needed per our interconnect agreement.

Actually our system isn’t saturated enough to cause problems now, but it’s becoming a problem in other areas of the country.
Its doubtful we ever will be that saturated.

We will occasionally offload a station on to another station, and we depend on some solar installations for load support on certain substations. We have some solar sites that will carry an entire circuit (2000 homes) or two, depending on loads.

 

[Carultch]

Whoa sir. There is a lot in that statement for me to absorb at once. I will reread a few times and comment further, thank you. I think there is something in there. And this is in reference to the "grid no-export" option or just synchronizing with the grid (meter) in general)? And does the additional heat generated (by suppressing the current output) cause any issues for the modules shore or long term?

I was describing the "grid no export" setup specifically. All inverters have power limitation by design. By default, it remains at the inverter nameplate value. Using grid no export tells it to change that setpoint, in response to what is needed.

The modules are rated to withstand all incident sunlight being converted into heat, as happens during an open circuit condition, so "leaving sun on the array" due to inverter power limitation is not an issue for them.

 

[electrofelon]

Not really… we can refuse to allow grid tie if needed per our interconnect agreement.

Actually our system isn’t saturated enough to cause problems now, but it’s becoming a problem in other areas of the country.
Its doubtful we ever will be that saturated.

We will occasionally offload a station on to another station, and we depend on some solar installations for load support on certain substations. We have some solar sites that will carry an entire circuit (2000 homes) or two, depending on loads.

From another thread you know I have a rather large (for residential) PV system connected thru two of my (customer owned) transformers. The voltage gets driven up quite high at the array and I've had to tap down one transformer 5% to keep the inverters from shutting down. I'd be curious what the voltage change is at the service point, I have never measured that. I'm about 8 Miles from the sub on an older 4800 line. I'm guessing a measly 16kw would hardly phase even that line.

 

[Hv&Lv]

From another thread you know I have a rather large (for residential) PV system connected thru two of my (customer owned) transformers. The voltage gets driven up quite high at the array and I've had to tap down one transformer 5% to keep the inverters from shutting down. I'd be curious what the voltage change is at the service point, I have never measured that. I'm about 8 Miles from the sub on an older 4800 line. I'm guessing a measly 16kw would hardly phase even that line.

For 16kW the neighbors would probably absorb it before it ever makes it back. Isn’t there a regulator close somewhere on your line? Seems I remember a thread some time back about a nameplate you took a picture of (on a ladder) and I was telling you about the regulator. Probably should be set to “cogen” on the relay panel so it doesn’t try to regulate both ways as it would on ”bi-directional”. It may be that way now…

 

[electrofelon]

For 16kW the neighbors would probably absorb it before it ever makes it back. Isn’t there a regulator close somewhere on your line? Seems I remember a thread some time back about a nameplate you took a picture of (on a ladder) and I was telling you about the regulator. Probably should be set to “cogen” on the relay panel so it doesn’t try to regulate both ways as it would on ”bi-directional”. It may be that way now…

Yup good memory. I think I was asking what the two pointers on the dial were for. That regulator is about a mile and a half from me.

 

[Hv&Lv]

Yup good memory. I think I was asking what the two pointers on the dial were for. That regulator is about a mile and a half from me.

Yes.. min-max drag hands, with present position hand.
then there are limit switch pointers (2) around the edge

 

[ggunn]

I didn't read the whole thread, so I may be repeating what someone else has said, but you are going down a rabbit hole unless you understand the difference between a current source and a voltage source. Grid tied PV inverters are current sources and the grid is a voltage source. Simply put, a PV inverter is going to supply all the current it can from the array and it has to go somewhere, while the grid supplies current on demand. If your loads are consuming more than your PV is supplying, your loads will use it all and the grid will make up the difference. If your loads are consuming less than the PV output, the PV will supply all the loads and the excess will flow to the grid.

 

[LATTC Student]

I didn't read the whole thread, so I may be repeating what someone else has said, but you are going down a rabbit hole unless you understand the difference between a current source and a voltage source. Grid tied PV inverters are current sources and the grid is a voltage source. Simply put, a PV inverter is going to supply all the current it can from the array and it has to go somewhere, while the grid supplies current on demand. If your loads are consuming more than your PV is supplying, your loads will use it all and the grid will make up the difference. If your loads are consuming less than the PV output, the PV will supply all the loads and the excess will flow to the grid.

Yes! Thanks GGunn for adding your knowledge here. Every bit helps me. I understood how it happens, it was the why that I was looking for. Your statement "a PV inverter is going to supply all the current it can from the array and it has to go somewhere, while the grid supplies current on demand" actually fills in the last piece of the puzzle for me. I hadn't thought of current as on-demand before but that's how Utility power does work. I remember Eric Stromberg referring to it (from a Mike Holt Video on Basics of DC last semester) when he mentioned "when it comes to current, it doesn't really exist, it isn't 'pushed' and the current isn't doing anything. It doesn't arrive until some need for it exists, generally from a load then it flows". . . . . strong emphasis on paraphrase

Thanks again.

 

[Carultch]

I hadn't thought of current as on-demand before but that's how Utility power does work.

What the utility really uses for feedback on aggregate power demand, is grid frequency. The utility has a standard to maintain grid frequency and voltage within a tolerance of the nominal values.

When it comes to conventional generating sources with rotating machinery, the grid frequency is directly determined by the rate that the turbines and generators rotate. Unlike wind power, where there is an inverter to allow turbine speed to be independent from the grid frequency, conventional generating sources have their rotation rate linked to the frequency they generate. There will be a fixed ratio between the RPM of the generator and the frequency in Hz, based on the geometry of the generator. For simplicity, let's consider an example where this ratio is 1:1. This would mean 3600 rpm corresponds to 60 Hz.

The generation stations take time to adapt fuel consumption to the varying power demand. Some forms are more adaptable than others, such as gas turbine plants and hydroelectric. Steam turbine power plants have their own advantages, but in this regard are better suited to supporting the steady base load. Prior to changing the fuel consumption rate, the generators will speed up when load decreases, and slow down when load increases. This means if it is spinning at 3610 rpm, they need to reduce the fuel, and if it is spinning at 3590 rpm, they need to add fuel, all while aiming for the target of 3600 rpm.