High grid voltage problem for PV system exporting

[mark_m]

@larry17

most inverters have an output that you can feed into a computer.

The catch here is how each manufacturer delivers this capability. The second consideration concerns what is delivered.

The nearest modern parallel might be the electronic management systems in modern motor vehicles. Owners might be able to see individual data such as fuel instant consumption, km left in the tank or outside air temp. The ability to access logged data or more specific engine management parameters is hidden behind a manufacturer firewall.

Inverter manufacturers are no different.

How well are the data collection and access tools for each inverter product reviewed?

Until you get to live with each manufacturers product on a daily basis it must be hard to know for sure?

The perfect solution includes a connection direct between your inverter and PC, no cloud service or third party required.

[margotk]

Thanks very much for providing this information, Gordon - it is a topic that is close to my heart. I will read it thoroughly when I have the time.

I purchased a Modbus meter for solar monitoring, with the original installation of my system a year ago. It is handy and the data is granular, so it’s useful.

However, for those who don’t want to invest in this type of monitoring, some distributors offer free access to their portals, which provide data from the electricity meter. In my case, my distributor is United Energy, and their portal provides graphical views to show electricity consumption and feed-in. There are views to show the meter data aggregated:

• hourly over the course of a day;
• daily over the course of a week;
• daily over the course of a month;
• monthly over the course of a season;
• monthly over the course of a year.

You can navigate to prior periods, and each view also shows the equivalent previous period, for comparison purposes.

Here is a screen shot for my metered data for yesterday and the previous day.

In the graph, the solar export (feed-in) is shown as bars below the solid horizontal line. Consumption is shown as bars above the line.

At the time of writing (8:12am), I was able to fetch data up to 8am today so the data is quite current. I have a smart meter, which record readings at 30-minute intervals.

My father’s distributor is Ausgrid and they provide access to a similar portal - for free.

Data from the distributor-supplied meter only shows you the grid-centric view - being data that your household consumes from the grid, and excess solar generation that is exported back to the grid. Still, it’s very useful because you can see whether you’re exporting anything to the grid which - for most solar systems (depending on size and self-consumption) is enough to determine if your solar system is working.

What the distributor-supplied meter won’t tell you is the amount of solar energy that is generated by your solar system that is self-consumed by the household. Self-consumption can be measured via a “behind the meter” metering solution in the form of a optional extra in which one may opt to invest.

[PhilT]

My enphase system has a monitoring facility that shows:

orange: imported from the grid
light orange: solar used
blue: solar exported
light blue: solar generated
as well as the generation for each panel over the day

The distribution between N and W facing panels is to use as much as possible, not generate as much as possible.

If I had transient over voltage clipping I would never know, especially with the cloud patterns over Melbourne but excursions that would cause shutdowns would likely show as anomalies regardless. The system is tracking its projected output within a reasonable margin, noting our weather is what it is.

My grid supply is Ausnet and they provide data much like @margotk’s with peak, off-peak, and solar export components as well as daily costs plotted against consumption/generation.

[margotk]

Looking good!

Given the individual panel monitoring, do you have micro-inverters or DC optimisers?

Re voltage clipping, all of the instances that I’ve seen of this have resulted in the inverter tripping, which shows up on the usage graph as generation dropping to zero (usually fleetingly).

I have not experienced any inverter trips with my system and, when I harvest the voltage data as measured by the Modbus meter, it doesn’t go outside the expected thresholds.

[margotk]

United Energy has been trialing measures to improve grid stability in my area - the Mornington Peninsula - through its “Dynamic Voltage Management System”. I’m not clear on exactly what it entails although there is an ARENA media release on it here: https://arena.gov.au/news/united-energy-to-test-voltage-control-improving-grid-stability/

I did email United Energy several times to elicit more information - because I’m personally keen to see a big take-up of renewable energy generation - hence it would be great to address the issue of curtailment. However, they haven’t given me any insight.

The ARENA release states: “United Energy will install frequency monitors and use data from its smart meter network to quickly act to stabilise frequency across its 47 zone substations, which act as controlling points across the electricity network”. I wonder what exactly does that mean…

[PhilT]

@margotk

do you have micro-inverters or DC optimisers

Enphase are (in)arguably the leader in microinverters.

[mark_m]

@margotk

it’s very useful because you can see whether you’re exporting anything to the grid which - for most solar systems (depending on size and self-consumption) is enough to determine if your solar system is working.

Working or performing to specification?

Another and more immediate way to monitor the daily performance of a typical non hybrid PV system, no optimisers and no microinverters:

This is what the power output of a typical rooftop PV system will look like, sunny day all day, no faulting, and perhaps some shading first thing in the morning and late afternoon. This is a screen shot from the app on my mobile phone, accessible 24x7 for today or any day. Every install and system has differences.

But not everyday is perfect (19kWh for the following day vs 26kWh for the prior example, that’s 7kWh lost)

There are a number of possible explanations for the rapid changes in power output in the second half of the day. In this instance it was a cloudy afternoon. It might alternately be due to high line voltage. A good PV monitoring system will provide access to more detailed information. It does need a little more knowledge from the owner, and that ready access to the inverter data.

The core discussion point raised by @gordon for this topic is that:
1/ consumers are not fully aware of losses from residential PV systems shutting down or throttling due to inadequate line voltage management by the supply authorities. Typically through high or over voltage on bright sunny days when PV output is at it’s max.

2/ the attitude of the supply authorities to the issue is at best unclear, with no direct path to a short term solution.

P.S.
AGL also offers the ability to access the daily metering data from our smart meter. It is an indication of how the payback is progressing against the Solar Salespersons whispered promises.

[margotk]

@mark_m

Working or performing to specification?

No matter what data you get from monitoring, it would be difficult to make a precise determination that a system is working to specification. Considerations include (but are not limited to) panel orientation, panel angle, panel and component operating temperature, cable losses, shadowing (roof finials, antennae, trees, other objects), accumulations of bird poop, cloud cover and expected panel deterioration over time.

I agree with your statement re the discussion point. There was also information provided about monitoring solutions and I would like other consumers to know that one does not need to invest in expensive monitoring to get visibility of the household’s grid consumption and solar exports.

I did invest in a very expensive solar solution including monitoring - and am very happy with its capabilities, including providing me access to APIs to allow me to harvest data and store it in a local database.

However, this is not for everyone. Most distributors and some retailers also provide monitoring portals and mobile apps at no extra cost. The only caveat is that these are based on metering data and don’t provide a view of self-consumption. Nonetheless, this should suit most consumers’ needs.

[Fred123]

Here are a couple of photos of the output of our system for the past 2 days and a photo of how the power was used yesterday.

They clearly show that once it gets up to around 4.5kw, something is limiting it from reaching its rated maximum output of 6.6kw.

[nickh64]

I too have the Enphase Micro Inverter setup on our 20 panel North facing array and find this information very useful.

It isn’t perfect but at least I can see if any panels are under performing due to faulty inverter or being dirty.

The other bonus with micro inverters is that you have built in redundancy in the system in that if one of them fails, it doesn’t take down the entire array but merely operates at a reduced capacity. They are also reportedly to be more efficient.

[gordon]

@nickh64

They are also reportedly to be more efficient.

I think that is only the case if there is one or more shaded/damaged etc panel. For strings of unshaded panels in good condition string inverters are just as good efficiency wise.

[mark_m]

@Fred123

They clearly show that once it gets up to around 4.5kw, something is limiting it from reaching its rated maximum output of 6.6kw.

Solar PV Systems are now typically supplied with a PV array that has a greater nominal DC power output than the Inverter AC rating.

Eg 6.6kW of panels connected to a 5kW maxim AC output inverter, or 3.6kW of panels with a 3kW max AC output inverter. This is done to help offset losses through poor panel orientation, output/efficiency loss as panels heat up and other influences.

The maximum output power of an inverter is usually fixed and will not go above the factory set or built in output AC power limit.

Can you clarify the inverter outputs are for a 5kW AC rated inverter? In which instance the inverter might be expected to clip the power at 5,000kVA, or less?

The clipping of the PV output at approx 4.5kW may be due any one of a number of causes, including high line voltage reducing output power.

What does your installer have to say?

There’s are a number of performance settings as well as design features of the inverter which may be contributing. Although 35kWh for a winter day from a nominal 5kW system seems a very good result.

[PhilT]

@Fred123

They clearly show that once it gets up to around 4.5kw, something is limiting it from reaching its rated maximum output of 6.6kw.

@mark_m is right on. The 6.6kW figure and the one usually referenced by the sales people is DC on the roof, not AC from the inverter. Any inverter connected to the mains is required to meet grid standards, usually max of 5 kW AC for a single phase supply. Our Aussie available inverters are designed to be marginally under 5 kW to allow for tolerances without breaking the 5kW limit.

@mark_m

Although 35kWh for a winter day from a nominal 5kW system seems a very good result.

Writing from Melbourne, Yes!

My nominal 4350 kW output has not done more than 29.6 kWh on any day over the winter and peaks at 35.4 in the summer, the latter being 98.9% of the max forecast for my installation at my location and considering I am increasingly shaded by ‘immovable things’ to the west from late afternoon and noting my installation was designed for maximum consumption, not maximum generation.

[gordon]

@mark_m

Although 35kWh for a winter day from a nominal 5kW system…

Keeping in mind that 8th October isn’t winter
Often PV systems produce very good output in spring with longer days but not too much heat.

Fred’s 4.5kW max output is probably the max export limit, as mentioned above, the 6.6kW of PV is unlikely to be

@Fred123

its rated maximum output of 6.6kw.

which will just be the rating of the panels. Depending on the inverter, it may divert any excess power beyond the export limit into the battery, if it isn’t at full charge.

[grahroll]

@mark_m

Although 35kWh for a winter day from a nominal 5kW system seems a very good result

Cairns in winter…clear skies a lot of the time and still lots of sun…35kWh probably is achievable most days if it is a cloud free day, at least 11 hours of daylight per day rising to over 13 in Summer.

Sunrise and sunset times in Cairns

Longer days and shorter nights further up North actually makes sense to place large Solar generators in those areas particularly if you can avoid areas with large cloud coverage during the Wet season. Something a bit more West of the coastal regions seems a great starting point.

[Fred123]

35kwh is the best day so far, but when the weather is good, it has been producing 30kwh+ daily.

I understand that the FIT is limited to 5kw, but as I mentioned in an earlier post, I have seen it producing over 6.8kw twice and over 6.6kw once but I don’t know whether the battery was charged or how much we were consuming.

I will take note of all details so as to investigate it further.

[gordon]

@Fred123

…but I don’t know whether the battery was charged or how much we were consuming.

Which emphasises how useful full (external to the inverter) monitoring is.

[gordon]

Nick and others with Enphase might find this Blog entry interesting:

Enphase Micro Inverters Review | The Pros and Cons | Oct 2019

[nickh64]

Hi Gordon,
We have full exposure of all panels with no shaded areas as we are located in an area surrounded by open parkland with no trees within 300-400 metres so we are extremely lucky to be able to get 100% use of all available sunlight on any given day so we are very happy about that.

Of course not everyone with solar will be in the same position and for that we are extremely greatful.

[nickh64]

Thanks for that Gordon.
Great article explaining the pros and cons of Enphase Micro Inverters.
Greatly appreciated.