Just now, plonkster said:

Again, something that is not uncommon. A brown-out below 180V would cause them all to disconnect in one smooth sweep, potentially making things much worse 🙂

And there we have the reason behind the limitation, this is to protect the network. So, in conclusion: CoCT and NRS097-2-3 limitation shall remain in force regardless of export into the Grid. Only Inverter size (and limitations) will be looked at. 

 

I had a response from CoCT on their interpretation and this will be VERY interesting to all on this Topic. Will post same on other topics as well:

Good afternoon Karel

 

1.Grid-tied hybrid SSEG:

The City’s drawings are schematics and site specific conditions might exist that must be provided by the applicant. The basis will be kept i.e. the separate suitably interlocked change-over switch for grid tied hybrid SSEG [as specified in Appendix 4 of the City’s ‘Requirements for SSEG’ document] for any hybrid grid-tied inverter must isolate an interconnected circuit from the grid before the storage system is used. For IN8 Sht3/6, the switch can also be in the sub-feeder circuit for the essential loads as indicated below for storage islanding mode. The inverter integrated network and system protection and two in series relays are dedicated for grid-tied functionality and may not be used for switching between grid-tied and storage modes. The registered person must sign off compliance of the switch in the CoC and test report.

 

FYI, note the following in draft SANS 10142-1-2:20XX

 

2.SSEG generation capacity:

NRS 097-2-3 defines the generation capacity guidelines for shared and dedicated feeders. The City requirements ito the ‘Requirements for SSEG’ for shared residential consumers are as follows:

The maximum generation capacity is inter alia required as follows:

·         Ensure network stability:

o    during changing irradiation and solar eclipse

o    when grid-tied embedded generators systems malfunction and property owner immediate load is transferred to the grid

o    during loss of storage capacity when and property owner immediate load is transferred to the grid of hybrid grid-tied systems [except from NRS 097-2-1: 2017 below]

·         Fair and equity generation capacity allocation possible to all customers in feeder groups ito NRS 097-2-3

29 minutes ago, Rautenk said:

the separate suitably interlocked change-over switch for grid tied hybrid SSEG [as specified in Appendix 4 of the City’s ‘Requirements for SSEG’ document] for any hybrid grid-tied inverter must isolate an interconnected circuit from the grid before the storage system is used

This confuses the heck out of me. Does it mean you're not allowed to bring the batteries into the picture until you've disconnected from the grid? In a Victron system the batteries are permanently connected (because that's where the power comes from, from the DC bus where the MPPTs leave it). Same is true of Goodwe. And what do they mean separate?

I suppose I should just be glad that my inverter is on "the list".

Edit: Or do they again have another interpretation of what "hybrid" means?

Edited January 24, 20197 yr by plonkster

The summary is basically:

1. Before the Unit ISLANDS it should be disconnected from the Grid, not sure how you can practically do this with Victron or Goodwe, but if you look at Annexre 4 of the SSEG requirements it details what is required for the switch.

2. NRS097-2-1 compliance is only for the "Grid-Tie" portion, ignoring hybrid functionality of the inverter (basically does it comply with Export rules)

15 minutes ago, Rautenk said:

Before the Unit ISLANDS it should be disconnected from the Grid

In my world "islanding" is the act of disconnecting from the grid. They are one and the same.

Is "after" used in the context of its physical position?

Edited January 24, 20197 yr by phil.g00

8 minutes ago, Rautenk said:

Before the Unit ISLANDS it should be disconnected from the Grid

Oh hang on! I think my understanding of "Island" is flawed. In the context of a Multiplus, this would be feeding power out of the input side to power things on your little island. Before you can do that (if you even can... which you can't with this inverter, because it has ANTI-islanding), you must disconnect from the grid first.

So if my understanding is correct this would then not apply to the it that is on the OUTPUT side of the inverter. Those are backup circuits... not an island.

Right?

Just now, phil.g00 said:

In my world "islanding" is the act of disconnecting from the grid. They are one and the same.

Technically it is not (according to how CoCT explains it).

They want you to first disconnect from the CoCT grid, then (and only then) is the inverter allowed to go into Islanding mode (provide power via batteries to the "Essential Supplies"). 

 

1 minute ago, plonkster said:

Oh hang on! I think my understanding of "Island" is flawed. In the context of a Multiplus, this would be feeding power out of the input side to power things on your little island. Before you can do that (if you even can... which you can't with this inverter, because it has ANTI-islanding), you must disconnect from the grid first.

So if my understanding is correct this would then not apply to the it that is on the OUTPUT side of the inverter. Those are backup circuits... not an island.

Right?

Correct, the Disconnect only applies to the "Grid-IN" side and not the "UPS-OUT" side (so you definitely do not need a change over on the output of the inverter, unless you want to go crazy and complex with a Inverter Bypass option)

2 minutes ago, Rautenk said:

They want you to first disconnect from the CoCT grid, then (and only then) is the inverter allowed to go into Islanding mode (provide power via batteries to the "Essential Supplies"). 

Which would rule out "power assist mode" from a Victron Inverter.

@phil.g00, I do not think so. Power Assist is if there is not enough power from the Grid Side to cover the load and the Inverter adds more with Battery Power, Correct?

This would be whilst still Grid Connected and because this inverter passed the NRS097-2-1 tests it would be allowed.

The topic here is if the Grid were to fail, then you need to disconnect the Grid-In side of the inverter from the Grid.

It is a very redundant safety feature, but according to SANS required. 

 

2 minutes ago, Rautenk said:

Correct, the Disconnect only applies to the "Grid-IN" side and not the "UPS-OUT" side (so you definitely do not need a change over on the output of the inverter, unless you want to go crazy and complex with a Inverter Bypass option)

Well what is funny about this is that internally the output is latched to the input while the inverter operates grid-parallel. There is technically no difference between IN and OUT while running in Hybrid mode.

There is however a separate two-series-relay device in addition to the backfeed relay with its own monitoring and redundancy. As is required by the regs.

1 minute ago, phil.g00 said:

Which would rule out "power assist mode" from a Victron Inverter.

Why?

@Rautenk, It would be akin to a PV inverter, but instead of the source being PV direct, the source would be batteries, which in turn are fed from the PV via MPPT's.

It wouldn't be a situation that the grid had insufficient power, but more like the load needed more than the system could provide and the extra would come from the grid.

But, no two ways about it, there is a battery involved in supplying the loads, while the grid is connected.

3 minutes ago, phil.g00 said:

But, no two ways about it, there is a battery involved in supplying the loads, while the grid is connected.

And if that is illegal, then just about every hybrid inverter on their list would drop off.

Which is why I start to wonder if we have the same definition of "Hybrid" in mind. Because that is what Hybrid means... it means I mix PV, battery and grid as needed. Can't really do that if the battery has to be disconnected...

I'm not saying it is, just like yourself I am trying to interpret the words used.

On 2019/01/24 at 12:38 PM, Rautenk said:

NOTE Some or all of the power generated may be consumed by the customers’ local loads. Where there is no
local storage the generator size is the active power rating of the installed generation. In cases with local
storage, the storage can be used to reduce the effective size of the generator by compensating for variations in
generation output, hence the definition used above."

Wow, thank you very much @Rautenk for this, and your other posts! I read through the rest of the thread, and tried to understand as much as possible. (E&E by training, but got lost in software like so many of our colleagues, and so I'm definitely out of my depth here.

How exactly should I interpret "the storage can be used to reduce the effective size of the generator"?

If my hybrid inverter is the NRS 2017 approved 5kVA EM-series, with a change / discharge capacity of 2.3kW (3.5kW peak for 10 seconds) into connected local storage, does this mean the effective size of my generator is 5 - 2.3 for the intents of the regulations?

On 2019/01/26 at 6:38 PM, cpbotha said:

How exactly should I interpret "the storage can be used to reduce the effective size of the generator"?

To be honest @cpbotha, I am grappling a bit on how the best way to "prove" this is. The onus will be on the user to prove to the engineer that there is the required settings to prevent the Grid from ever seeing more than the required 3.5kVA (on 60A).

 

On 2019/01/26 at 6:38 PM, cpbotha said:

If my hybrid inverter is the NRS 2017 approved 5kVA EM-series, with a change / discharge capacity of 2.3kW (3.5kW peak for 10 seconds) into connected local storage, does this mean the effective size of my generator is 5 - 2.3 for the intents of the regulations?

Could you please provide me with more information on this, what do you mean charge / discharge capacity limits into connected local storage. Are you referring to Battery Charge Controller limits? 

I have raised this specific point (regarding reduced sizing) to the City as well and will see what Ryno responds. 

Regards

To me this is becoming clearer.

The grid isn't worried about battery charging, that is clearly a load.

As custodians of the grid, they are accepting of the fact that throughout the course of a day, domestic loads are being switched on and off at random times and Independently of each other, and it's their job to manage the demand. It was always thus.

Now, along come inverters, and let's say they become as popular as "mobile phones" (everyone has one). Thing is about an inverter, they're all set to respond (trip out) for certain grid events. And by all I mean together the entire mass of generation would be lost, it would be the equivalent of all the random independent loads above being synchronized and happening at once.

With the present 1/4 of capacity limit, that would mean an extra and immediate 25%  load, and if future generation was allowed up to this limit. this swing in supply to demand could potentially be 50%. (Unlikely, but conceivable).

Now, if I have the reasoning right, I think there is a space where customer generation capacity could be greater than the limit, but the CB at the supply point would trip at 20A ( 4.6kA).

In other words the if the present philosophy changed from a 25% relative size of supply to an absolute 20A one.

The domestic IPP would be still be limited to causing the maximum load swing envisaged.

Of course, the IPP will have now limited himself to being only drawing 20A/phase from the grid, but in return he could increase his own self-generation capacity to above that limit.

I think this might be a preferable and practical option for some.

There would be no more onus of proof than already exists when sizing any supply. The size of the grid-controlled MCB at the interface sets that limit.

 

Edited January 27, 20197 yr by phil.g00

3 hours ago, Rautenk said:

On 2019/01/26 at 6:38 PM, cpbotha said:

If my hybrid inverter is the NRS 2017 approved 5kVA EM-series, with a change / discharge capacity of 2.3kW (3.5kW peak for 10 seconds) into connected local storage, does this mean the effective size of my generator is 5 - 2.3 for the intents of the regulations?

Could you please provide me with more information on this, what do you mean charge / discharge capacity limits into connected local storage. Are you referring to Battery Charge Controller limits? 

The specs on this specific hybrid inverter, the GoodWe EM 5kW, claim that in battery backup mode it can sustain 2.3kW to the house from the attached batteries (the so-called local storage), and it can supply 10 second peaks of 3.5kW from these batteries. I mentioned this, because in other cases, the inverters are able to deliver their full rated output from battery power only.

Anyways, I am super curious to hear what your CoCT contact says about how exactly storage (batteries) reduce the effective output of the inverter!

(I have read the "NOTE 3" exception you posted over and over, and it is slowly making more sense. I am now wondering how to prove, for generation that exceeds the 25% limit but has storage "included", that when the generation trips, the load *delta* suddenly transferred to the grid can never be > 25%. The fact that they specifically mention this configuration, means that there are cases where it can be shown. Super curious!)

 

Edited January 27, 20197 yr by cpbotha

4 hours ago, cpbotha said:

I have read the "NOTE 3" exception you posted over and over, and it is slowly making more sense. I am now wondering how to prove, for generation that exceeds the 25% limit but has storage "included", that when the generation trips, the load *delta* suddenly transferred to the grid can never be > 25%. The fact that they specifically mention this configuration, means that there are cases where it can be shown. Super curious!)

This is what i am putting forward, is a simple grid-owned MCB solution that is sized for the absolute current value of 25% as being an option.

Sure it would limit what you can draw from the grid, but at the same time it should also be allow you a greater capacity to generate for your own consumption.

Because, if you ever tried to transfer more than a 25% delta the MCB would trip. 

The question is would you prefer to have say 75% of solar, but only be allowed to draw 25% from the grid, or as now, 25% solar but be allowed to supplement that with 75% from the grid.

( Could in fact have more solar than 75% solar with a 25% MCB limit).

Either your MCB prevents that Delta, or your inverter size does.

I am saying if you paid for an existing 80A supply, the you applied to the authorities with your intent to exceed 4.6kVA ( 20A) of embedded generation, then they would come along and replace your 80A MCB with a 20A one with your blessing.

Naturally, the Grid will still have to charge you a standing grid charge for 80A, although you'll be limited to 20A. (They still had to up-size originally for you to downsize later).

Then there would be no need for you to be limited to a 4.6kVA  generation capacity, because the 20A MCB would automatically do that regardless.

It's a simple and cheap solution, not an administrative nightmare of officials and engineers. There would also be less guessing as to solar penetration rates, and and a real growth path for on-grid solar generation to alleviate grid generation pressures without the down-sides.

Already the size of the MCB installed in the meter box is the accepted standard that governs the capacity of a supply. - No change of policy required.

This might be preferable to some, I know it would be for me.

On 2019/01/21 at 10:40 AM, The Terrible Triplett said:

Most are not electrically inclined, like my Dr, mechanic, family. It must just work.

enter : "the idiots guide to solar power"  as proposed elsewhere... obviously the 1st hurdle is to acknowledge your status 

I'm probably fall into the "Dr, mechanic, family" category and all of this is getting too technical for me to follow.  We have a 60A feed but the Goodwe does up to 4.6kW.  Is it not as simply as applying for a 80A feed with the CoCT?

 

2 minutes ago, Niel said:

Is it not as simply as applying for a 80A feed with the CoCT?

the problem is that the cable to your property is probably limiting the current to 60A. To upgrade to 80A might require a new cable. Not sure who would have to pay for this.

COCT does not like offering an 80A supply to residential homes, however mine was done by simply applying and the reason i gave was to fit a lift......No cable change, paid R385 for them to change the breaker in the street box and i had to supply my own for my DB.

6 minutes ago, Fuenkli said:

the problem is that the cable to your property is probably limiting the current to 60A. To upgrade to 80A might require a new cable. Not sure who would have to pay for this

I recommend going the upgrade route to be honest. Thus far I have not had a suitable suggestion or idea of my own how to prove this. There is always cases that will result in full generator load being seen by the Grid.

Looking at the costs from CoCT, to install a New 60A supply is around R2000, so I would expect it to be less than this to upgrade to 80A. That way you are 100% assured that you will not be limited to 3.5kVA generation. The cost of upgrade will definitely be much less than the loss in Generation.

Thank you Mike for the actuals on your install, for R385 there should be no argument, but lets see what the CoCT has to say about the upgrade. Strange that City Power again prefers the 80A option...

 

 

Edited January 28, 20197 yr by Rautenk
Mike provided real world experience, updated post accordingly

11 hours ago, phil.g00 said:

Thing is about an inverter, they're all set to respond (trip out) for certain grid events

I know that Germany with a much higher solar penetration than South Africa had this problem with the old "disconnect regulations". The new regulations requiring a gradual reduction (instead of an abrupt stop) of power has addressed this problem.