diy solar

diy solar

All grid tied inverters in UK/EU/US/AUS limited to 5.5kW?

Rotating mass is good to overcome the inrush.
But it is bad also because slow to respond. An inverter + battery can correct in ms.
More the mass, slower the response. So it is good for base power plants. But fast reaction power plants need fast response.

If you have a rotating generator, the RPM (for at least some types) corresponds to 60 Hz line frequency.
When you throw a big load on it, it pushes out current but starts to slow down. So governor gives the diesel more fuel, or valve lets more water into turbine.
For a pelton wheel (a type of high head turbine), nozzles spray stream of water through the air against buckets. Adjustable nozzle can control flow while still producing same velocity.
I'm less clear on how low head turbine vary their water consumption with load. Water flowing through blades seems like almost fixed displacement.

For pumped hydro, need variable volume of pumping (which could be multiple pumps turned on or off.) With lots of PV, would want pumps running and varying their volume to match excess production.

Of course the grid has multiple rotating generators, and they have to be in sync. At least some are excited by the grid and push power into it. Not sure how close GT PV inverters behave to that.

The grid generally didn't have a problem, if not overloaded, before PV. It seemed to be able to handle load variations. Shouldn't be that different with PV, except more rapid swings as a cloud moves over L.A. with PV on all roofs.
 
Are all grid tied inverters for use in UK/EU/US/AUS all limited to 5.5kW?
No.

Connection limits will be set by the local electrical distribution authority/company and these not only vary by country but also by region within a country.

I am in Australia and have a 10kW grid tied inverter. Some I know have 30kW.

Also, is there a limit on the amount of electrical energy you are allowed to export to the grid each day?
Again, any specific connection restrictions or requirements are unique to each electrical distribution zone/region/country.

In Australia it's most common for exports to be restricted to a maximum of 5kW per phase, but in some areas the limit is 10kW while in other areas it's zero or somewhere in between. Where I am it is 3kW/phase.

If so, then what happens when your battery is full, and your house load is low, and your panels are generating, and your already above the maximum amount of export allowable?

If you system has the capacity to generate more power than the total of the export limit + household load, then the inverter clips production to maintain output within the export limit. The system will have its own consumption meter to monitor energy flows in from / out to the grid so as to not exceed the export limit. This is how mine is set up as my export limit is less than the inverter's output capacity, although not by much.

So for instance if you have 13kW of PV, 10kW inverter, a 5kW export limit, it's a nice sunny day and the house (and battery if you have one) is consuming 1kW, then the inverter will limit total output to 6kW so as to not exceed any more than 5kW export to the grid.

If the house at that time then switches on another load, e.g. a water heater with a 3.6kW draw, making total household loads 4.6kW, then the inverter will increase production to 9.6kW (assuming sufficient solar PV is available).

In such situations it makes sense to move as much discretionary load to peak solar production hours. Use it or lose it.
 
A while back I saw an article that talked about requiring the utility to control personal solar production but have heard nothing since.... (but me not hearing means very little).

The biggest technical hurdle with allowing utility control would be coming up with the standard to do it. Getting all of the cats herded into the same standard will not be easy.....Then add to that the probability that each country will end up with a different standard makes it even more difficult.

If they ever do start doing something, it should be targeted at what is going onto the grid, not what is produced.
actually, i read that us companies already started adding this to their codes.
using frequency shifts to shutdown solar production
 
The features are part of UL-1741 SA. Some mandatory, some optional.

The U.S. grid is divided into three parts: West, East, and Texas.
Frequency can only be a single number in any given part.
Frequency-shift works great for my stand-alone island grid when disconnected from utility.
For grid-wide control, I don't think frequency shift would be great. L.A. might need all the power it can get, while a PV producer up a long branch wire might be boosting voltage too high.

I would have thought volts-watts (another optional feature) would do better.

The plan is for some kind of networked control of the inverters.
I think wireless (pager) would be practical, with broadcast commands being received by many at once.
 
The features are part of UL-1741 SA. Some mandatory, some optional.

The U.S. grid is divided into three parts: West, East, and Texas.
Frequency can only be a single number in any given part.
Frequency-shift works great for my stand-alone island grid when disconnected from utility.
For grid-wide control, I don't think frequency shift would be great. L.A. might need all the power it can get, while a PV producer up a long branch wire might be boosting voltage too high.

I would have thought volts-watts (another optional feature) would do better.

The plan is for some kind of networked control of the inverters.
I think wireless (pager) would be practical, with broadcast commands being received by many at once.
well, in europe similar regulations are comming..

from my point all reason more to build for grid-assisted at best..
but thats me
 
If you have a rotating generator, the RPM (for at least some types) corresponds to 60 Hz line frequency.
When you throw a big load on it, it pushes out current but starts to slow down. So governor gives the diesel more fuel, or valve lets more water into turbine.
For a pelton wheel (a type of high head turbine), nozzles spray stream of water through the air against buckets. Adjustable nozzle can control flow while still producing same velocity.
I'm less clear on how low head turbine vary their water consumption with load. Water flowing through blades seems like almost fixed displacement.

For pumped hydro, need variable volume of pumping (which could be multiple pumps turned on or off.) With lots of PV, would want pumps running and varying their volume to match excess production.

Of course the grid has multiple rotating generators, and they have to be in sync. At least some are excited by the grid and push power into it. Not sure how close GT PV inverters behave to that.

The grid generally didn't have a problem, if not overloaded, before PV. It seemed to be able to handle load variations. Shouldn't be that different with PV, except more rapid swings as a cloud moves over L.A. with PV on all roofs.
Actual loads on a large grid are very predictable. Before wind and solar the load and demand were the same things from the utility point of view. Utilities could plan pretty accurately when they needed to ramp up or down, even for systems that take a long time to ramp. (Some take several hours to ramp up)

With wind and solar in the mix, the load is 'hidden' by all the renewable energy that the utilities have no view of, let alone control of. Now an unexpected change in demand can happen with little or no warning and the utilities may not be able to react fast enough.

Currently, almost all grid batteries are used to react to sudden changes to give the utilities time to adjust their primary generation. Little or none of the grid batteries replace any of the base generation the utilities operate.
 
actually, i read that us companies already started adding this to their codes.
using frequency shifts to shutdown solar production

I think all allowed Grid-tie inverters have to have this function (part of VDE, AS, G98, ...).
But this can not be used in the public grid. It is much harder to maintain the constant frequency than the constant volt.
Big load starts -> freq goes down. Big load stops, or too much power -> freq goes up.
It is constant struggle to maintain the balance. (hard to control big rotating masses ... easy with inverters)

Also grid-tie inverters have the overVolt shutdown.
Utility "uses" that here (silently). Like there is a lot Grid-tie inverters in a line ... and they do not upgrade the transformer.
All Grid-tie inverters raise the V to be able to sell the power.
So the V is rising from 230V until 252-260V. And in this range the grid-tie inverters start to shut down (for pre defined 5 or 10 minutes).
V goes down, stabilizing.
In this way utility limits a street max grid export capability. This is not "graceful", but works.


The plan is for some kind of networked control of the inverters.
I think wireless (pager) would be practical, with broadcast commands being received by many at once.
I think I wrote about the challenges this have. Uncontrollable amount of inverters do unknown amount of power production.
To poke it and send a message to them to lower 50% or shut down ... that could result in uncontrollable situations ?
 
Actual loads on a large grid are very predictable. Before wind and solar the load and demand were the same things from the utility point of view. Utilities could plan pretty accurately when they needed to ramp up or down, even for systems that take a long time to ramp. (Some take several hours to ramp up)

With wind and solar in the mix, the load is 'hidden' by all the renewable energy that the utilities have no view of, let alone control of. Now an unexpected change in demand can happen with little or no warning and the utilities may not be able to react fast enough.

Currently, almost all grid batteries are used to react to sudden changes to give the utilities time to adjust their primary generation. Little or none of the grid batteries replace any of the base generation the utilities operate.

Also before there was not very often that a 100-200 kW load just starts or stops.
Now in the time of the superchargers it happens every minute.
Unpredictable high load changes ... only inverters with batteries can react fast enough.

Yes, big battery does not replace base power plants like nuclear. Can not. Big battery stores and releases energy. A power plant makes/generates energy.
 
Also before there was not very often that a 100-200 kW load just starts or stops.
Now in the time of the superchargers it happens every minute.
Unpredictable high load changes ... only inverters with batteries can react fast enough.
Good point.... but I wonder how the superchargers average out with the rest of the load. As large as they are, they are probably a drop in the bucket compared to the total load on the grid. As you say, they start and stop all the time, are there enough of them that this just looks like an increase in the overall load?

BTW: I heard some of them are using local storage to handle the large peak power. This allows them to be built without paying the utilities to put in larger lines and larger transformers.
 
Good point.... but I wonder how the superchargers average out with the rest of the load. As large as they are, they are probably a drop in the bucket compared to the total load on the grid. As you say, they start and stop all the time, are there enough of them that this just looks like an increase in the overall load?

BTW: I heard some of them are using local storage to handle the large peak power. This allows them to be built without paying the utilities to put in larger lines and larger transformers.

Here the supercharger I see has 8 place. Each up to 120 or 150kW max power.
If a Tesla fanclub comes and all starts to charge ... then it is almost a 1 MW power consumption jump. And it is only one supercharger place.
A nuclear reactor block produces 500-1000 MW ... just to have something to compare.

If they charge only 40-50% then that is 8x 30kWh = 240 kWh. Big numbers.
A Tesla Megapack has 3 MWh power, so it could fill up 8x4x3 = 96 cars (to 40-50%).
 
I think the question in the title of the thread has been answered. There are a variety of Grid Tie inverter sizes and in only a few places are the export to the grid limited to 5.5 kW or some other number.
 
Thanks, but i am virtually certain that in UK at least, export to grid from a single household , is limited to 5kwh per day.
 
using frequency shifts to shutdown solar production
Across a grid that is highly unlikely, as grid frequency needs to be kept within a very tight range across the entire grid. Large deviations from the standard frequency signal something very wrong has happened (e.g. a main inter-connecter has gone down, or a large power station has blown up) and requires immediate action (e.g. tactical load shedding) to prevent a complete system blackout.

Frequency control of a household solar PV system is something used when the system is isolated from the grid. e.g. a AC coupled battery inverter uses frequency shifting to moderate the power output from the solar PV inverter.

Grids/networks have a few options for curtailing distributed generate sources.

One is via grid voltage limits. As grid voltage rises, most modern grid-tied solar PV systems will first derate output, then eventually when an upper grid voltage trigger has been reach they will shut down and not restart until a lower grid trigger voltage has been reached.

Grid operators do also have the ability to adjust voltages to force these changes to occur more rapidly, or more broadly, or in specific zones. This is fairly crude and indiscriminate and does not fairly share the pain of shutdown since everyone experiences different voltages on the same network, but in aggregate it works. There is a limit however as the operator must not supply too high a voltage as that can be damaging.

There are new inverter standards being introduced in parts of Australia (South Australia) where inverters will require the network to have the ability to derate or shutdown that specific inverter in special circumstances. The quid pro quo is the grid is permitting much larger solar PV systems to be connected enabling much more solar supply at times when there isn't an impending supply-demand imbalance. You may still choose to opt out of the network control but your system's export limit will be dropped to only 1kW, instead of 10kW. Comms network control comes with its own challenges which is why it is on trial at present. Given they are the grid with highest penetration of rooftop solar PV in the world, it represents an excellent case study on how such power supply systems can be successfully managed.

As an example, this is the mix of solar PV supply in the South Australian grid over the past week:

Screen Shot 2021-10-11 at 7.48.08 am.png

There have been and will be periods in the day in South Australia where rooftop solar PV alone has supplied nearly 100% of the state's total demand. Now it's not a isolated grid, there is an interconnecter with a neighbouring state so there is some import/export and a new interconnecter with another large state is being built - and this is a good thing as it mean less curtailment of the distributed renewable supply and more system robustness given the geographical spread.

The old grid supply paradigm is changing, and the means, technology, systems and processes required to transition to a more distributed supply are happening. You either plan well for it, or get left behind.
 
Thanks, but i am virtually certain that in UK at least, export to grid from a single household , is limited to 5kwh per day.
Yes that would be the "other number" I was referring to. Apparently it may vary by Country, Province, State or other area where a power company sets a limit. My point was that not all inverters are limited to 5.5 kW in the UK/EU/US/AUS or anywhere else for that matter as the OP asked.
 
Across a grid that is highly unlikely, as grid frequency needs to be kept within a very tight range across the entire grid. Large deviations from the standard frequency signal something very wrong has happened (e.g. a main inter-connecter has gone down, or a large power station has blown up) and requires immediate action (e.g. tactical load shedding) to prevent a complete system blackout.
Agreed, however there is another method that the electricity industry uses called ripple control which is a form of frequency control (and might be what @houseofancients was talking about). But instead of changing the grid frequency, they inject a high frequency signal on to the network which is picked up by special relays that control non-essential loads (and potentially solar inverters). The signals are normally generated at each substation and only controls loads supplied by that substation. So effectively they can turn off all non essential loads in a suburb. And if they added relays to grid tied solar installations (and used a different frequency to control them to what they use to control non-essential loads). They could remotely disconnect and reconnect our GT PV systems.
Grid operators do also have the ability to adjust voltages to force these changes to occur more rapidly, or more broadly, or in specific zones. This is fairly crude and indiscriminate and does not fairly share the pain of shutdown since everyone experiences different voltages on the same network, but in aggregate it works. There is a limit however as the operator must not supply too high a voltage as that can be damaging.
Yes this is possible, and they may use this in Australia, however as you said, it is very crude and indiscriminate as they can only adjust voltages at the HV level (here in NZ they can only remotely alter the 11kV and higher networks) and not the LV networks which supplies your home. And when they do increase the HV voltage, they don't know how much of an impact this is having on the LV voltage. I.e your local LV transformer could normally supply 249V and the one down the road could have more load on it and only supply 225V. So a change to the HV voltage is going to knock your inverter off, but not the ones connected to the other transformer.

There are new inverter standards being introduced in parts of Australia (South Australia) where inverters will require the network to have the ability to derate or shutdown that specific inverter in special circumstances. The quid pro quo is the grid is permitting much larger solar PV systems to be connected enabling much more solar supply at times when there isn't an impending supply-demand imbalance. You may still choose to opt out of the network control but your system's export limit will be dropped to only 1kW, instead of 10kW. Comms network control comes with its own challenges which is why it is on trial at present. Given they are the grid with highest penetration of rooftop solar PV in the world, it represents an excellent case study on how such power supply systems can be successfully managed.
And this is likely what we are all going to move to in the near future I suspect. And hopefully they will not only be able to curtail generation, but also ask your battery inverter to supply more energy into the grid when needed (for the appropriate fee).
In New Zealand we are trialing something similar called DERMS (or distributed energy resource management system). From what I have read it uses a communications protocol called OpenADR, which hopefully aligns with what is being used in SA and in other countries.

Yes that would be the "other number" I was referring to. Apparently it may vary by Country, Province, State or other area where a power company sets a limit. My point was that not all inverters are limited to 5.5 kW in the UK/EU/US/AUS or anywhere else for that matter as the OP asked.
My distribution company here in Bay of Plenty New Zealand limits you to 5kW per phase, which no limit to the amount of energy per day as far as I can work out.
 
Agreed, however there is another method that the electricity industry uses called ripple control which is a form of frequency control (and might be what @houseofancients was talking about). But instead of changing the grid frequency, they inject a high frequency signal on to the network which is picked up by special relays that control non-essential loads (and potentially solar inverters). The signals are normally generated at each substation and only controls loads supplied by that substation. So effectively they can turn off all non essential loads in a suburb. And if they added relays to grid tied solar installations (and used a different frequency to control them to what they use to control non-essential loads). They could remotely disconnect and reconnect our GT PV systems.
Yeah I wasn't providing an exclusive list of options.

I have a ripple control relay in my main circuit board (we refer to them as a "controlled load") which is for my off-peak hot water circuit. It operates late at night mostly, although sometimes they switch it on during the day on weekends. I did a heat map recently of when my HW system draws energy by day of week and by hour of day (it usually only needs to heat for a couple hours max). This is over the last few years:

Screen Shot 2021-09-30 at 7.04.58 am.png

Can see it turns on sometime after 11pm usually, runs for a couple of hours (3.6kW element, average daily consumption is 5.1kWh). But occasionally it operates for longer if we used more HW that day and sometimes the grid operator leaves the circuit on during the day on the weekend when they consider it prudent to add some daytime load. This has the effect of making Saturday a bigger day for HW energy consumption while Monday is much lighter on.

That said, I don't see ripple control relays being used here for the purpose of disabling solar PV systems. Indeed in one state (VIC) when they mandated smart meters in all homes many of the ripple control circuits were removed as such devices were pushed to operate at off-peak when people were transitioned to time of use tariffs. The issue was the smart meter could either meter solar PV, or a controlled load but not both.

Where I am (NSW) they provided two smart meters, a 3-phase unit for regular metering including solar exports and a separate single phase unit for the controlled load.

In Sth Aus the off-peak time is now middle of the day from 10am to 3pm. It's more expensive to buy energy overnight.
 
So I think it is better if every substation has a 50-100 MWh battery with 20-40 MW inverter.
First it is not centralized. A centralized big battery is a SPoF (Single Point of Failure).
Second it can work as a local UPS. Here is a town part where a substation has a 2 MWh battery with an 500kW inverter (in a container). And it can separate the substation from grid, make a microgrid to the homes and give power even it does not get from the main line.
It is exactly what we do, only bigger. And frankly, it is the utility company's job to do it.

I think batteries have to get a lot smaller before that is achievable. I installed a 1MW, 2MWH tesla battery storage system at a substation here in NZ a few years ago and it was almost the same size as the rest of the substation. I.e it was the biggest we could fit on the land available, and most substations don't have a lot of space available for battery systems.

I think the answer is to have batteries at the point of demand, and to have this controllable by the grid operators.
 
most substations don't have a lot of space available for battery systems.
Haha, here in California many PG&E employees store their boats and RVs at substations. There is plenty of room for a MegaWatthour of storage and that is where it should be located in my opinion. They could charge with the excess solar they complain about screwing up their control systems with bidirectional flow.
 
Last edited:
5kw limit: That may be true in Australia, based on the link you posted. There is no limit based on physics but each Country may have some limit based on unique rules in that Country.
Definitely not true here in Australia. Managed to convince the parents to go solar and they got a 15kw inverter installed with 20kw of panels on their roof.
 
Back
Top