Oversized PV

[Alexphd1]

Forget the cost pay back for a minute but how could i fit a large Pv system say 15kW in a domestic home and limit the excess power into the grid or stop altogether going into the grid. 

[SteamyTea]

Split it into 3 sections and fit 3 separate 4kW inverters.

Then switch of the inverters you don't need.

 

Actually you may want to fit 3 3kW inverters  to get better efficiency.

Edited May 28, 2020 by SteamyTea

[ProDave]

At that size you will need permission from DNO (unless it is 3 phase)

 

Fit battery storage. then you only have to think about shutting down an inverter if the batteries are fully charged.

[Alexphd1]

I have a problem to remember to switch of lights.... is there not something on market to fit on downstream of meter to stop exporting electric? 

 

[ProDave]

12 minutes ago, Alexphd1 said:

I have a problem to remember to switch of lights.... is there not something on market to fit on downstream of meter to stop exporting electric? 

 

You can buy (but hardly off the shelf, think bespoke) G99 or G100 relays.  But these only limit the export to the maximum the DNO will allow. https://www.g59projects.co.uk/g100-export-limitation/

 

You would probably have to build / have built for you, some custom controls to monitor export and if too much shut down one or more inverter.

 

I dump my excess power to the immersion heater and a small panel heater if the export is more than the immersion can soak up on it's own.  When it gets too hot to tollerate any more heat in the house, I might put that heater in the garage out of spite, rather than export it for nothing.

 

[dnb]

1 hour ago, ProDave said:

You can buy (but hardly off the shelf, think bespoke) G99 or G100 relays

Exactly this. But they seem to be getting more common these days according to this:

Single phase inverter, apparently with G100 export limit

I applied for an "in principle" grid connection with SSEN recently using one of the above as my export limit, and they seemed happy. We'll see if that is still the case now the real application is ticking through the system. Admittedly I am only looking at 6kW or so of panel, but my export limit is likely to be in the region of 3kW.

[SteamyTea]

18 minutes ago, dnb said:

Admittedly I am only looking at 6kW or so of panel, but my export limit is likely to be in the region of 3kW.

Well it will be 16A ~3.7 kW

But that is governed by total module size, not inverter trickery (usually, there are always exceptions).

 

So decide what you can do with an extra 2 kW as an off grid system i.e. dedicated water heating/sewage plant running/battery storage.

[joth]

2 hours ago, ProDave said:

At that size you will need permission from DNO (unless it is 3 phase)

 

Even with 3 phase it would need permission, being over 11kW ? 

 

I'm in a bizarrely similar situation of wanting to limit export, not just because self-consumption is more £ efficient but because it will push the local voltage over 253V and shutdown the inverter. At times of high PV generation I might literally be better off (financially and in CO2 footprint) by running a pointelss convection heater in the garage or shed than having the inverter shutdown.  I SMH at how we've ended up with such contradictory regulations & incentives.

 

[ProDave]

4 minutes ago, SteamyTea said:

Well it will be 16A ~3.7 kW

But that is governed by total module size, not inverter trickery (usually, there are always exceptions).

 

So decide what you can do with an extra 2 kW as an off grid system i.e. dedicated water heating/sewage plant running/battery storage.

It is the inverter that sets the maximum generation.  My inverter is rated at 4kW but with the output current limited to 16A.  This caused the DNO to initially reject my application, saying it was over the limit for automatic approval, that was resp;ved when I sent them the data sheet stating the current limit.

 

 

[SteamyTea]

5 minutes ago, joth said:

because it will push the local voltage over 253V and shutdown the inverter

Does it shut the inverter down totally, or limit the voltage, which reduces the headline efficiency figure, but not really a real issue.

[SteamyTea]

1 minute ago, ProDave said:

It is the inverter that sets the maximum generation

In reality yes, but I think the DNO sets the limit on module capacity, or people would fit a 16A limit inverter, then attach 5 or 5.5 kW of modules to it.

This would cause the DNO problems, regardless of what the technical limit.

 

One way to check is to ask the DNO if it is ok to fit a diesel generator to their lines.

I believe this is one of the reasons that CHP is limited on the electrical generation capacity, to do with the hour running.

[Dan F]

I've been told by PV suppliers that some DNO's don't really accept G100 limiting and instead insist on a maximum inverter size.  How much truth there is to this though I don't know, and it may depend on your specific DNO.

 

Regardless of your inverter size though (which may be limited for various reasons) you can still oversize you PV array.  Solaredge single-phase inverters support up to 155% over-sizing, and if you consider "effective oversizing" rather than "STC oversizing"  (for example if you have panels on east and west roofs) you can actually oversize even more, you just need to run the calculations based on location and pitch/orientation.

 

In our case, according to the Solaredge designer,  we could put as much as 17kW installed capacity on the roof with a 10kW single-phase inverter and only loose maximum 4% to clipping in June.  We've decicded to go down the 3-phase route though, even though the DNO will allow 10kW export/inverter on a single-phase as we already have a 3-phase supply.

Edited May 28, 2020 by Dan Feist

[dnb]

14 minutes ago, SteamyTea said:

Well it will be 16A ~3.7 kW

They've indicated that there's quite a lot of PV already on "my" bit of wire, so I might not get the full 16A worth. (I won't know until I get the results of my application). They only seemed concerned with the export, not the total size. In reality, they aren't going to ever see any of it in the first place - it will be directed to local use of some kind for all but the most extreme circumstances.

 

28 minutes ago, joth said:

running a pointelss convection heater in the garage or shed than having the inverter shutdown

I too might be in this position. The house is specified to have full A/C - it's always good for soaking up a bit of spare power. Can always open a window to make it work harder...

[Dan F]

40 minutes ago, joth said:

Even with 3 phase it would need permission, being over 11kW ? 

 

No, because you can easily run 15kW array with a 10kW 3-phase inverter and still stay within the standard 3.68kW.  DNO care about export and potentially also the inverter size, but AFAIK they don't care how many panels you actually put on the roof.

[DamonHD]

Various of you are saying that the inverter size matters rather than panels, but I don't believe that was the case at least when I did mine.  The DNO/G83 and MCS cert seemed to be all about the nominal panel capacity IIRC.

 

Now in my case, because I was an early bird, and I think also because my array is split east-west and simply cannot all run at 100% all the time, the DNO let me put up ~5.2kWp on ~4.5kWp of inverters on my single-phase connection.

 

Rgds

 

Damon

[joth]

4 hours ago, dnb said:

 

4 hours ago, joth said:

running a pointelss convection heater in the garage or shed than having the inverter shutdown

I too might be in this position. The house is specified to have full A/C - it's always good for soaking up a bit of spare power. Can always open a window to make it work harder...

Interestingly I just found this:

https://shop.gwl.eu/docs/web/2018/EN_50438_2013-BS.pdf

Section "A.12 GB – United Kingdom" (page 31) clearly shows the stage-1 overvoltage protection for the UK only starts at 262.2 V (with 3s delay) and stage 2 is way up at 273.7 V.

 

SolarEdge have said (via our supplier) their inverters cut out at 255 V (presumably with ~5s delay) and hard cut off at 262 V, i.e. designed for European market, and there's no way to adjust these trip-points.  Time to look for alternative manufacturers, it would seem!

 

[I'm probably derailing the conversation here - I know the grid over-voltage issue has been mentioned many times but I can't spot another sensible thread to put this on?]

 

 

 

 

[SteamyTea]

Amy be worth reading though this short document as it gives some general details, and where to find more indepth stuff:

 

http://www.inbalance-energy.co.uk/further_reading_books/pv_installers/Guide to the installation of PV systems_2nd Edition.pdf

 

It is a decade now since I was involved in PV, but as far as I can remember, and what I have heard from some installers, two things have changed.

Inverters are not allowed in lofts anymore and the system capacity is set by total module capacity.

But as usual, there is probably more to it this this, the so called 'gotchas'.

One thing that is interesting is that inverters need to have isolators on both the DC and AC sides.  Not sure how that is implemented with micro-inverters.  Be a lot of switches.

Edited May 29, 2020 by SteamyTea

[joth]

2 hours ago, joth said:

Interestingly I just found this:

https://shop.gwl.eu/docs/web/2018/EN_50438_2013-BS.pdf

Section "A.12 GB – United Kingdom" (page 31) clearly shows the stage-1 overvoltage protection for the UK only starts at 262.2 V (with 3s delay) and stage 2 is way up at 273.7 V.

 

SolarEdge have said (via our supplier) their inverters cut out at 255 V (presumably with ~5s delay) and hard cut off at 262 V, i.e. designed for European market, and there's no way to adjust these trip-points.  Time to look for alternative manufacturers, it would seem!

 

 

As initial  validation of this:  the Luxpower hybrid inverters (as referenced several times by @Jeremy Harris)  have the stage1 trip set to 262.2 V, and stage 2 way up at 273 V, according to their G59 test report here (table 13.8.3.2, page 15).

 

Given that this hybrid inverter costs less than the SolarEdge (even without counting all the micro optimizers), and gives option of battery storage for only the cost of the batteries (which can be incrementally added), it's starting to look a very attractive alternative.

The main drawback is while it can handle 8kW of PV, it only has a 5kW inverter, so once the battery storage is full we'll be losing 3kW of generation. (I'm assuming it can dump 3kW to batteries at the same time as exporting 5kW via the inverter). So to minimize that wastage we'd need to buy LOTS of batteries, or get 2x 4kW inverters, either way pushing the price way back up again.

 

 

Edited May 29, 2020 by joth

[Dan F]

@joth Is it too late for you to go 3-phase if you are concerned about the voltage issue? 

 

Looks like this may be another validation that 3-phase is the best approach for us (vs. 10kW single-phase inverter), but I do wonder if there are any examples of practical issues caused by this or if it is more theoretical..

[joth]

38 minutes ago, Dan Feist said:

@joth Is it too late for you to go 3-phase if you are concerned about the voltage issue? 

 

Looks like this may be another validation that 3-phase is the best approach for us (vs. 10kW single-phase inverter), but I do wonder if there are any examples of practical issues caused by this or if it is more theoretical..

 

Yes, I think it is really. It would mean we wasted about £1500 moving the existing supply on top of the £3000-odd for the new one. Moreover, we'd also need to create a new place for the meter (the existing was allowed to remain indoors as it was already there, but a new supply has to go externally) and so potentially re-dig all the trenches for that, and likely a 6+ month delay given UKPN would need to close the road to make the connection and that seems really hard to do because of the school. (Cadent charged a pretty price to do this for the gas permanent disconnection, but in the end I think they gave up trying to get the permission and have just left it with a "temporary" disconnect just inside our boundary).

But I agree, I would have gone with 3ph had I realized this back in January.

 

When you ask for practical examples, you mean of inverter cutting off from being overvoltage? In the UK, my neighbour and @Jeremy Harris are the two I know of.

 

[SteamyTea]

2 hours ago, joth said:

losing 3kW of generation

Realistically it is not that often that a PV system delivers at full power.

It is easy to get carried away with the weather we have been having the last couple of months, but think back to last October, November, December and January.

[joth]

9 minutes ago, SteamyTea said:

Realistically it is not that often that a PV system delivers at full power.

It is easy to get carried away with the weather we have been having the last couple of months, but think back to last October, November, December and January.

Yeah, my mind spins both ways on this: if these sunny times are rare occurrences, surely I want to maximise them as much as I can!!  but you remind me of a good point: maximizing peak output is not necessarily going to dint the overall seasonal performance so much.

One thing I do like about the Luxpower is I can easily see a path for getting a minimal system installed, and then incrementally improving; adding more batteries, or even splitting to 2 inverters if we really found it would be justified

 

Edited May 29, 2020 by joth

[SteamyTea]

Being able to easily add to a system is useful.

Stecca used to do an inverter system like this.  You bought a 2 kW inverter that had all the controls in it, then just kept adding extra 2 kW basic inverters to it.

And it kicked in once the module string was generating 80V, which meant it actually generated more kWh than a comparable SMA system.

[joth]

5 hours ago, joth said:

8 hours ago, joth said:

Interestingly I just found this:

https://shop.gwl.eu/docs/web/2018/EN_50438_2013-BS.pdf

Section "A.12 GB – United Kingdom" (page 31) clearly shows the stage-1 overvoltage protection for the UK only starts at 262.2 V (with 3s delay) and stage 2 is way up at 273.7 V.

 

SolarEdge have said (via our supplier) their inverters cut out at 255 V (presumably with ~5s delay) and hard cut off at 262 V, i.e. designed for European market, and there's no way to adjust these trip-points.  Time to look for alternative manufacturers, it would seem!

 

 

As initial  validation of this:  the Luxpower hybrid inverters (as referenced several times by @Jeremy Harris)  have the stage1 trip set to 262.2 V, and stage 2 way up at 273 V, according to their G59 test report here (table 13.8.3.2, page 15).

 

OK OK looks like this might have been a red-herring, Solar Edge sending out misleading info to our supplier or something.

At least, I found the G99 test report for the SE8000H and it states that stage 1 trip occurred at 262 V and stage 2 at 274 V, and it can operate at 258 V for 2s without trip, so from that data it looks a lot more viable. Just need to get their tech support to confirm this.

 

[Gav_P]

On 28/05/2020 at 14:26, joth said:

I might literally be better off (financially and in CO2 footprint) by running a pointelss convection heater in the garage or shed than having the inverter shutdown

Time to dig a pool and heat it for free ?