Grid-Tie Inverter Theory and Operation Confusion

[cinoaz]

I understand the principles of a Grid-Tie Inverter (GTI), but there's a few key items that aren't clear I was hoping to tap into the knowledge base on the forums. This is a GTI with no batteries, no charge controller. Just PV Arrays and GTI

A GTI is (first question) a dumb device. It's doesn't know anything about "load". Its job is to take whatever is connected to its input and drive as much power out as possible. If it can provide more power than the house needs, that raises the impedance of the home panel above that of the grid, and power flows out to the grid. When the Inverter produces less power, it lowers the impedance of the house panel and power flows from the grid to the home. I think this is correct.

Now for the more specifics.

A GTI has 1 or many PV inputs. People say your DC to AC ratio should be 1 to 1.5. This means to oversize your panels based on your inverter specs. This is the confusing part, since most inverters have two limits. The first limit is the total output power (i.e. 5Kw, 7Kw, or 10Kw inverters for example). That means the inverter, regardless of how much DC power gets shoved into it, will max out at its absolute limit. This is called "clipping".

Now back to the inputs. Each GTI has limits as to what can be connected to it in terms of input. There is a Voltage and Amperage range for inputs. Let's say the inverter has 2 inputs rated at 360v @15amps each. Now, if I overdrive a singular input with more panels, that means I'm either exceeding the inputs Voltage or Amperage rating based on the PV array configuration. From what I understand, the GTI doesn't "Clip" the input lines, only the output lines. So if you provide 360V @100amps on the PV input, the inverter will try to convert every ounce of that and most likely will burn something out in the process. The Inverters don't "clip" inputs, thus, why its so critical to size PV arrays correctly.??

a 10Kw GTI has two inputs rated at 360v @15amps for a total of ~5Kw per input for a max output of 10Kw. However, if someone tried to only connect one PV input with 360v@30amps, although the ouput is identical, the inverter will blow up because the the single input exceeded the amp limit of that input?

[Ampster]

I am not sure the GTI inverter can affect the impedance of a house panel. I prefer to think in terms of the grid presenting an infinite load to the GTI. It may be that the relative impedance of those two loads changes.
Also the clipping is typically done by the internal MPPT algorithm which happens in the MPPT controller section {DC} to match the output load capacity of the inverter {AC}. We see clipping as a flat top of a typical GTI output curve. As cinoaz mentioned, The MPPT controller is more sensitive to voltage limits on its input than current so that is the important variable in matching strings to inverters.

[cinoaz]

The MPPT controller is more sensitive to voltage limits on its input than current so that is the important variable in matching strings to inverters.

Overloading is based on Voltage or Amperage or both? If I have a 360v @ 100amps to a PV input expecting 360V @15amps, do we see smoke from the inverter? Does the Inverter try to gobble down ALL 100amps or does it "clip" it's amperage limit on the input somehow?

I think that's the key question here. When overloading a PV input and people talk about 1.4, or 1.5 overloading, are they talking Voltage, Amperate, combination thereof, or??

[Ampster]

Overloading is based on Voltage or Amperage or both? If I have a 360v @ 100amps to a PV input expecting 360V @15amps, do we see smoke from the inverter? Does the Inverter try to gobble down ALL 100amps or does it "clip" it's amperage limit on the input somehow?

I think that's the key question here. When overloading a PV input and people talk about 1.4, or 1.5 overloading, are they talking Voltage, Amperate, combination thereof, or??

i will try to answer the question initially from a physics standpoint and then my practical understanding. The way I understand how PV panels work is they put out voltage when there is any available light. Looking at graphs of my panels I see the voltage ram up and stay steady most of the day, but current may vary if clouds pass by and I see current ramp up and down, respectively in the morning and evening. The amount Current they produce is a function of the intensity of light which I believe is called insolation. As you mentioned it is important to think of both Voltage and Amperage when considering overloading. Of course the most common metric used is a Wattage ratio or DC panel capacity to AC inverter capacity. That hopefully answers your last question as to the common metric of DC to AC ratios. Overpanelling may have a more ambiguous meaning.

However when configuring strings or even micro inverters, voltage is important because the components in a MPPT controller or microinverter are designed for certain voltage ranges. As a practical matter component prices vary in relationship to the Voltage they can sustain, so engineers and not going to specify more expensive components than the device is designed for. Voltage is a variable for the most part outside the control of the device so that specification should not be exceeded when matching panels to a device. Current can be limited by the device based on loads or its internal capacity to handle current. The analogy I learned in basic electronics is that voltage as a push and current as a pull if that makes sense? PV panels themselves are describe as current sources which means to me that they can put out optimum current depending on the load that is pulling that current. I wont get into the nuances of they typical MPPT controller which can vary voltage and current to optimize Wattage output. Hopefully that answers your questions. There are others with more technical knowledge.

[cinoaz]

PV panels themselves are describe as current sources which means to me that they can put out optimum current depending on the load that is pulling that current.

Thxs for the feedback and yes, we agree in an overview how things align, but its the more gritty stuff that is unclear. Since a GTI is always connected to the Grid, LOAD is "infinite" in terms of the Inverter since to the inverter the GRID is infinite. So the "Pull" will always be whatever the Inverter can provide with no regard with how much the home needs. Since Current(amps) flow like water, if the Inverter is producing more amps in the panel than the home needs, the panel has a higher pressure "more amps", that means water will flow to the grid. If the home panel has less amps, water will flow from the grid. How much water flowing in either direction is based on the current need of the house and how much the inverter is providing.

But the basic question still applies. If the Inverter is a "dumb" device, that means, whatever you put on the PV input line, the inverter will use it all. This is why I think the mantra "size your panels" is so prevalent. For my system, I have 15 panels per string, in series, giving me 15amps @360v, exactly what the inverter says should be on the input. Knowing Voltage is a bigger concern, let's say I double my PV array by adding another 15 panels iand reconfigure Parrallel/Series to maintain a voltage remains at 360V but my amperage increases to 30amps. Now the Inverter input is seeing 360v @30amps, which is twice the amount of watts.

Does the inverter try to gobble up ALL 30 amps on the input, which is WELL ABOVE its rated capacity on that input or does it self limit to only 15amps no matter how many amps is on the line.

Yes, I understand if you oversize based on AMPs there's the possibility all these panels are wasted since the inverter will max out at 15amp (rated input) but that is the question, is it really limited or is that what you are supposed to meet in the PV array configuration? Is the inverter a dumb device and will gobble up whatever is sees, is that why "sizing" you array is deemed "critical" or is the only critical component the Voltage?

In order to reach 10Kw output on my Inverter, it needs at least 2 PV arrays (inverter actually has 3 PV input lines) but you need at least 2 to reach the max output numbers. Since each input is 360v@15amps, if I combine my arrays into ONE big one 360v@30amps and feed it into ONE pv input, I expect the inverter to go poof, since, it exceed the maximum allowable amps on that input and the inverter can't self limit its input, it just uses whatever is available regardless of any stated limits in the manual.

The ONLY clipping that ever occurs is on the output of the inverter.

[Ampster]

........... If the Inverter is a "dumb" device, that means, whatever you put on the PV input line, the inverter will use it all. This is why I think the mantra "size your panels" is so prevalent. For my system, I have 15 panels per string, in series, giving me 15amps @360v, exactly what the inverter says should be on the input. Knowing Voltage is a bigger concern, let's say I double my PV array by adding another 15 panels and reconfigure Parrallel/Series to maintain a voltage remains at 360V but my amperage increases to 30amps. Now the Inverter input is seeing 360v @30amps, which is twice the amount of watts.
Does the inverter try to gobble up ALL 30 amps on the input, which is WELL ABOVE its rated capacity on that input or does it self limit to only 15amps no matter how many amps is on the line.

The MPPT section of the inverter is not a dumb device because it has an algorithm to maximize Watts. The Inverters only intelligence is to limit how much it can pull from the MPPT controller{s}.
The rating of the panels is short circuit current and open circuit Voltage. The inverter does not "see" those Amps it only pulls the current it needs based on the capacity of the GTI to serve the loads. It does see the voltage. In your example of a GT Inverter it would only pull the Amps necessary to provide the Wattage of its output capacity

Yes, I understand if you oversize based on AMPs there's the possibility all these panels are wasted since the inverter will max out at 15amp (rated input) but that is the question, is it really limited or is that what you are supposed to meet in the PV array configuration? Is the inverter a dumb device and will gobble up whatever is sees, is that why "sizing" you array is deemed "critical" or is the only critical component the Voltage?

All the panels would not be wasted. All the panels output would be inverted up to the maximum Wattage. The result could be an earlier ramp up to capacity in the morning and a slower ramp down in the afternoon but there would be clipping mid day. Voltage is critical in terms of not letting the smoke out. Opinions vary, but my opinion is current is more of a use case issue which may be site specific. I am including electric rates as a component of use case. For example if you have an open field you could have some panels facing East, some facing South and some facing West. That might be optimal depending the cost of panels and your electric rates. You might even see some clipping but you might also see full inverter production for 6-8 hours instead of maybe only 5 hours. It is not a one size fits all in terms of DC to AC ratios.

In order to reach 10Kw output on my Inverter, it needs at least 2 PV arrays (inverter actually has 3 PV input lines) but you need at least 2 to reach the max output numbers. Since each input is 360v@15amps, if I combine my arrays into ONE big one 360v@30amps and feed it into ONE pv input, I expect the inverter to go poof, since, it exceed the maximum allowable amps on that input and the inverter can't self limit its input, it just uses whatever is available regardless of any stated limits in the manual.

The ONLY clipping that ever occurs is on the output of the inverter.

I do not know what the limits are of each MPPT input. Typically you would have to stay within the voltage specs but in most cases you could exceed the current specs. In my example of three arrays the ideal scenario would be one array on each input.
Technically clipping occurs at the interface of the DC output section of the MPPT controller and the DC input section of inverter.. The typical GTI inverter has a DC section and an AC section. We observe clipping as a flat portion of the AC output but the limitation may occur somewhere inside one or both sections.. In the case of three inputs there may also clipping in each MPPT circuit.

[bcroe]

From what I understand, the GTI doesn't "Clip" the input lines, only the output lines. So if you provide 360V @100amps on the PV input, the inverter will try to convert every ounce of that and most likely will burn something out in the process. The Inverters don't "clip" inputs, thus, why its so critical to size PV arrays correctly.??

The GTI has a number of limitations to be observed. In general these
are broad enough not to interfere with each other. My 7500 W GTI has
6 + and 6 - DC terminals. Rated 15A @ to avoid burning the physical
wire, they may be connected directly to 6 strings of panels. But here
the 6 strings are 250' away, combined there in a combiner box into
single 6 gauge (60A capacity) wire. A copper comb at the GTI inputs
ties several (or all) of the 6 together, so that my nominal 400V 20A
input from the 6 ga wire will be distributed among the individual
terminals. This gives an overall current capacity limit of 90A input.

100A @ 230V would be 23,000W.

For my GTI to operate at 100A DC input, the input would only be at
80VDC. This is completely out of the MPPT operating range of
230-500VDC. Panels delivering 230VDC could only supply 35ADC
at the limited power output. Not very critical. Bruce Roe

[davidcheok]

I People say your DC to AC ratio should be 1 to 1.5.

Shouldn't listen to what 'people' say. You should read the inverter manufacturer's recommendations for maximum DC to AC ratio.

To calculate the maximum number of panels per string, there are several things to take note of. The start up voltages and the maximum voltages and current of the inverter, the panel's max voltage and current and obviously the output voltages (to make sure its suited to your grid).

So say you want to work out which panels you can use and at a specific DC to AC ratio e.g. 1.4. So if your inverter is 10kW, multiply that by 1.4 to get 14kW. Then divide the 14kW with the panel power you want to use, e.g. 400w. This gives you 35 panels. Divide that by the number of MPPT, e.g. 2, and you get 17 panels per MPPT (lost one panel here). If your panels allow 2 strings per mppt, then these two strings must be identical in panel numbers thus divisible by 2. 16/2 = 8 panels per string (note you lost another panel here). Now multiply the 8 panels by the max voltage to get total max voltage per string e.g. 45v x 8 = 360v and check that this number is higher than the start up voltage but lower than the max input voltage. If yes, you have your first possible configuration which is a system consisting of 16x 400w panels per string per mppt and 2 mppt = 32 panels = 12800w matched to a 10kW inverter giving you the final DC/AC ratio of 1.28. Once you know how to work this out, start playing around with panel sizes ideally in a spreadsheet until you get the desired 1.4 ratio and your ideal panel size for that inverter. Keep in mind the limitations of the inverter as well as the size of the panel against your available roof or land space. Also note you can always use less panels per string but should avoid creeping close to the maximum input voltages. If you are going to split the strings in parallel then be wary of the max current specifications.

The voltage and current max are there specified by manufacturer to prevent the inverter from going poof so avoid like the plague.

Note. Once the strings are within specifications, inverters 'draw' from the panels. When the max output of the inverter is reached, it will not take more than it can use. However, do bare in mind that at maximum production also means highest heat production from the inverters so make sure there is adequate cooling.

As for the grid, it doesnt really care (but the power station does). Whatever that isnt used up by the loads in the house automatically goes through the net meter and out to the grid. The provider will usually set a limit on what you can export because large fluctuations in export causes issues with the turbines slower response rates (unless grid is buffered with batteries along the way).

And as others have pointed out, be wary of the cable losses as well as resistance to avoid melting them out.

Below is the result from a 1.4 DC/AC ratio system. I plotted out the orange line from the maximum outputs across various days and times to see what the potential could be. As you can see, clouds are a big factor to take into account of in my area hence the larger ratio. Its 18 kW panels 600w each with 13 kW inverter peak output using 880w micros.


Screenshot 2023-03-09 at 6.43.47 PM.png

Below is an example of the workings to get your ideal max panel configuration.

Screenshot 2023-03-09 at 7.52.16 PM.png

[cinoaz]

davidcheok

Yes, all good stuff, but, you missed the core question. In your graph, you are increasing VOLTAGE to increase your DC to AC ratio and by that definition your AMPs are staying uniform for your WATTS calculation across all voltages.

The question is the direct opposite of this. What if VOLTAGE remains constant and only AMPS is adjusted? If your inverter input is rated for 360v @15 amps and you size your array for 360v @1,000 amps (insane number but that's the point). Will the inverter, on that single input, try to max its output power, exceed the rating on that input and go poof.

In equation form:

GTI Input #1 = PV array 360v@15amps = ~5Kw (Contacts on input are rated for 360v@15amps)
GTI Input #2 = PV array 360v@15amps = ~5Kw (Contacts on input are rated for 360v@15amps)
GTI Input #3 = unused
Total GTI rated ouput = 10Kw

Now, change the PV array:

GTI input #1 = PV array 360v@30amps = 10Kw (Contacts on input are rated for 360v@15amps)
GTI input #2 = unused
GTI input #3 = unused
Total GTI rated oputput = 10Kw or is the total just 5Kw because the inverter is self limiting how much it pulls to 15amps although the PV array has 30amps available?

In both scenarios, the total output of the PV arrays are equal to the Inverter max output. As you can see, however, the second configuration OVERLOADED the amps on a single string. The question continues

Will the inverter try to convert all 30amps on the single line to max its output of 10Kw, or, will the inverter "pull" up to 15amps on that line and the remaining 15amps available is wasted, since the inverter will only "pull" up to 15amps on that line the output is limited to 5Kw even tho there is 10Kw available from the PV array?

Can the inverter self-limit how many amps it pulls, or, does it just do VoltsxAmps=Wattage until Wattage equals max output and it doesn't care, per se, what the voltage or amps are on the line, it will just try to convert it. If either the Volts or Amps exceed the input limit, the inverter goes POOF because its just doing VOLTSxAMPS until that combination equals 10Kw? Its up to the person to "limit" the PV array accordingly for BOTH VOLTs and AMPs not to exceed the input limits of the GTI for each input?

The inverter can reach it's max output in either scenrio, but, will the inverter reach its max output in the second configuration? Will the inverter go POOF on the second configuration?

[cinoaz]

The inverter does not "see" those Amps it only pulls the current it needs based on the capacity of the GTI to serve the loads. It does see the voltage. In your example of a GT Inverter it would only pull the Amps necessary to provide the Wattage of its output capacity

Yes, and that is precisely the question. If the inverter is trying to hit 10Kw output, and 30amps is available on the PV line, will the Inverter just keep sucking up the AMPs until something gives?

VOLTS x AMPS = WATTS and regardless of Volts or Amps, the inverter just does the math. So if WATTS = 10Kw, and VOLTS are set to 360v, will the inverter ramp up the AMPs it consumes until it reaches 10Kw, even tho the input is only rated to 15amps, or, will the Inverter self limit to 5Kw output and not PULL more than 15amps on that input?

[bcroe]

The system DC operating voltage is set by the panel array. It
must be set up so the open circuit voltage, and the MPPT voltage,
over the temperature range, are withing the inverter specs. Then
the inverter will draw as much MPPT current as the panels can
deliver. But if more panel energy is available than the inverter
rating, it will throttle back the current drawn to limit output power
to its rating. This of course will cause panel voltage to rise
closer to open circuit voltage.

The issue with a very large DC to AC situation, is the ability
of the inverter ability to throttle back. Instability could result
and possible damage, if this is taken to an extreme.

Something elso not discussed, is the EFFECTIVE DC:AC ratio.
While adding up all the panels here gives a ratio higher than 2,
it is impossible for all the varied orientation panels to all be
facing the sun at the same time. So the primary panels keep
changing over the day, and the effective DC:AC is around 1.2:1.
This arrangement can double the effective hours per day under
good sun, and double the output with light dispersed by varied
levels of clouds. Inverters here have been happy with this
arrangement for a decade. Bruce Roe

[Ampster]

Yes, and that is precisely the question. If the inverter is trying to hit 10Kw output, and 30amps is available on the PV line, will the Inverter just keep sucking up the AMPs until something gives?

VOLTS x AMPS = WATTS and regardless of Volts or Amps, the inverter just does the math. So if WATTS = 10Kw, and VOLTS are set to 360v, will the inverter ramp up the AMPs it consumes until it reaches 10Kw, even tho the input is only rated to 15amps, or, will the Inverter self limit to 5Kw output and not PULL more than 15amps on that input?

As I mentioned earlier I have no idea how the three MPPT sections share their output. You will have to do that research to answer the question. There are two distinct sections of that device, the MPPT section and an inverter section. The MPPT section converts the solar energy to a voltage and Amperage compatible to the inverter section.

[davidcheok]

davidcheok
gti input #1 = pv array 360v@30amps = 10kw (contacts on input are rated for 360v@15amps)
g

poof

[cinoaz]

poof

And that was my thought. I know we like to think these inverters are more than they are. I tend to believe they are way simpler than we are led to believe. Just Volts x Amps until it reaches max output. It's a matched set, the inverter and panels. Both have to be matched with each other and limitations adhered to.

[Ampster]

And that was my thought. I know we like to think these inverters are more than they are. I tend to believe they are way simpler than we are led to believe. Just Volts x Amps until it reaches max output. It's a matched set, the inverter and panels. Both have to be matched with each other and limitations adhered to.

More than likely the only risk of poof is by over voltage on the input side. It is also important when matching current specs with charge controllers to understand whether the limit is on the input side or the output side of the charge controller section. In your case with multiple MPPT controllers going to an inverter whose input voltage may not be known the input current could be significantly different than the output current. That can lead to confusion about how much over paneling could be tolerated or which configuration would be optimal.