Jump to content

Is your system legal? Capetonians have till 28 Feb 2019 to register their systems


Guest

Recommended Posts

@plonkster  I do not know if he read the whole document but would assume so since he is a consultant with a reputation to consider. It would seem to me that the bone of contention is really whether the output of the inverter is connected to the grid. As Marius explained, it is not. Whether there is a possibility that it could feed back into the grid for whatever reason and under whatever specific circumstances, would require intimate knowledge of the Axpert or any other inverter's circuitry. The only people who clearly have that knowledge, as far as I have noted from this forum, are @Coulomb and @weber who may have an opinion on the issue regarding the Axperts. But they are fortunate to live in a first world country where, to my mind and with limited knowledge of the related issues, they are not subjected to all manner of quite silly regulations dreamed up and approved by quite incompetent politicians. They  have both expressed surprise that the Axpert is deemed to be unsuitable for Eskom and some RSA municipalities' grids, notably the City of Cape Town Metro. I accept that the specifications may have been defined by competent people but then it is their side of the story. If you disagree, your only recourse are the courts and the supporting evidence of the likes of Marius. This would work just fine if your bank balance looks like Johan Rupert's or that of Nicky Oppenheimer. So mister SA taxpayer usually just shut up and pay up. Somewhat like the R3000 cordless phones we were subjected to during the 1980's. That bit of lunacy died a sudden death but was nonetheless formulated by experts, probably of Telkom and the Buro of Standards and then enshrined in legislation. I also recall the efforts by academia to have legislation passed that only graduates with an applicable university degrees should be allowed to work on desktop computers. Fortunately that bit of lunacy did not make it through the parliament of the day. Not so sure of the present one.

Link to comment
Share on other sites

1 hour ago, ebrsa said:

whether the output of the inverter is connected to the grid. As Marius explained, it is not

With the Axpert it actually is connected to the grid. When it charges, it literally connects the output side to the grid, and then it runs the whole pipeline in reverse. The Axpert actually has a very modern topology... it is very almost a grid-tied inverter. With a bit more hardware and software, you could run the pipeline in boost mode and switch it into AC while connected to the grid. But the designers decided not to do that, so the H-bridge remains off (turning it into a passive rectifier), so one might say it is very almost sufficiently capable of being grid-tied. Makes sense too... use a tried and tested design and just dumb it down a bit. If it was a dumb UPS... then yes, there would be no problem here. The trouble is it isn't that dumb...

Link to comment
Share on other sites

1) Last year I realised qualifications and having a reputation to uphold, does not mean you know what you are talking about.
2) That every single qualified person out there, suddenly have become specialists on the internal workings of solar inverters / UPS'es - overnight.
3) That the only person who counts is the one who can get a UPS with a solar capabilities (or one that is not on the list) connected to a DB and signed off by CoCT's department dealing with it all, as grid tied.

Off grid, generators, the regulations are in place since yonks now as we have said here over and over.

So my thoughts are:
Get a electrician to give a CoC for the inverter/s in question.
If it is off-grid, submit the papers to CoCT.
If it is grid tied, then continue and get a engineer to sign it off as grid tied, then submit to CoCT.

Let whomever who says it is all ok, let them put their signature on that doc and submit it to CoCT. Or let them write to CoCT stating their opinion.

Like I am busy doing, as is a host of other solar users using knowledgeable, qualified, trained, reputable solar installers and electricians with engineers to back it all up.

Someone somewhere needs to put something on a block.

Link to comment
Share on other sites

All I know is that Marius Fourie was adamant that the Axpert is not a grid tied inverter but I do note your views @plonkster. Still have to determine the views of Swartland Director of Electrical Services now that the standstill of the festive season is history. @The Terrible Triplett  I am sure Capetonians will find your comments most useful. Fortunately CofC regulations are of no consequence to residents in Swartland. So Swartland's regulations and views is what I need to be guided by and comply with. As soon as I have ascertained that, I will take the liberty to convey them to the forum as a matter of interest at least.

Link to comment
Share on other sites

11 hours ago, The Terrible Triplett said:

1) Last year I realised qualifications and having a reputation to uphold, does not mean you know what you are talking about.
2) That every single qualified person out there, suddenly have become specialists on the internal workings of solar inverters / UPS'es - overnight.
 

.

 

 

That`s a pretty wild statement to make.  You might upset a few peoples feelings making statements like that in the public domain

Edited by Jaws
Link to comment
Share on other sites

8 hours ago, ebrsa said:

adamant that the Axpert is not a grid tied inverter

Oh he is absolutely right. It isn't a grid-tied inverter. The point is that it doesn't have to be grid-tied for NRS097-2-1 to apply to it. It only has to be an "embedded generator" and NRS097-2-1 would still apply to it. So it comes down to what constitutes an embedded generator.

Selection_071.png.64c5eff793728d4a90749a5b3c77784e.png

So it seems to me that being grid-tied is not a requirement. All that is needed to be an EG is that you have PV panels and a power converter (aka inverter). Hence NRS097-2-1 applies to Axperts. Which kinda makes sense, otherwise I don't understand why section 4.4 is in there. This talks about a "UPS with embedded generation", that is, a UPS with solar panels attached.

Selection_072.png.404e813bd7119dca4bffcd4110ed43bd.png

For the Axpert, 4.4.1.2 applies. It is mentioned again in 4.4.3, which points back to 4.4.1.2, which again sends you to SANS 10142-1 regarding the changeover switch requirements.

Selection_073.png.df1eddb6312fe863777eec61535afb5b.png

So for the most part NRS097-2-1 has very little to say about the Axpert, because the majority of the stuff in that document has to do with units that can feed power into the grid. That much is true.

Link to comment
Share on other sites

12 minutes ago, plonkster said:

It isn't a grid-tied inverter. The point is that it doesn't have to be grid-tied for NRS097-2-1 to apply to it. It only has to be an "embedded generator" and NRS097-2-1 would still apply to it. So it comes down to what constitutes an embedded generator.

I've haven't followed this discussion, but that phrase refering to embedded generation namely, "operate in synchronism" has a very specific and clear definition.

It means operate with the same frequency, voltage and "in phase" with the grid. 

I don't know this inverter at all, but if its only  230Vac and 50 Hz but out of phase its doesn't qualify as being in synch. 

That is very unambiguous.

Maybe its the loophole that's needed?

 

Link to comment
Share on other sites

35 minutes ago, Jaws said:

That`s a pretty wild statement to make. 

Very good point! Never will I purposefully intend to hurt peoples feelings. Just getting a wee bit frustrated. Allow me to refine. 

There ARE a lot of solar installers out there who sell and do installs, who will proudly sign off the product they sold.

Back to my frustration.
Why must I pay so much to get my system correctly installed, with a new problem discovered around just about every corner!? The latest one the bonding.
Over and over the regs are posted here, as per Plonkster's post above, right? Any electrician worth their salt will read up first, before they install anything.
As to connect anything to a DB board has regulations. Things like a UPS (with or without panels), generators, grid tied inverters. Ask any UPS / generator installer, they know exactly what must be done. Most jacked electricians do too.

It is not cheap to have a inverter tested for certification via the official testing houses right? Any supplier can go that route, have their products tested, right?
Yet it seems to me that money is wasted i.e. to pay for that testing. All you do is ask anyone who has skin in the game ito profit, and they just know that xyz inverter is acceptable. ;-)

In the end the customer who paid good money for XYZ product, for the install, the CoC, is left with the problem if it was not done properly or according to regulations. On this forum there are people who have paid good money to "reputable" solar installers, who have been burned.

The PV solar industry is still in it's infancy in SA. Quick course here, or there, and you are certified.

My prediction: In another 5-10 years 95% of the installers today will be gone, same as the solar geyser industry way back. The ones that are left would probably be electrician businesses first, like plumber businesses today being the "main" installer of solar water heating.

Sorry if anyone is offended, but you get my point.

Link to comment
Share on other sites

9 minutes ago, phil.g00 said:

operate in synchronism

I grabbed a screenshot from the definitions chapter. The document also refers to synchronous generators here and there, which is why it is in the list of definitions, but it is not limited to only synchronous generators. It clearly goes across the border into UPS land and tells you what happens if you add embedded generation (aka PV modules) to your UPS. In fact, one could argue that it is being pretty generous: It essentially gets out of the way and says:

For it seemed good to the regulators and to us to lay upon you no greater burden than these essentials: That thou shalt comply with SANS10142-1.

Which is why I keep saying: Show that the Axpert complies and you are done :-)

(No prize for the person who googles my appropriation of ancient literature).

Edited by plonkster
Link to comment
Share on other sites

@lonkster thank you for publishing the relevant section od NRS097-2-1. It would seem to me that section 4.4.2.1 applies as it refers to a UPS that does not operate in parallel with the grid and the in brackets defines it as one that cannot export energy to the grid. The Axpert is not designed to export to the grid and one would be justified in reasononing that NRS097 therefore does not apply to it. The SANS requirements seems to apply though. But all this is of academic interest to me but not the citizens of CofCT to whom the list of approved apparatus is binding. So I have to ascertain the requirements of our municipality for clarity of what I have to do. With some luck I may get an appointment to discuss the matter with our Director in the next 2 weeks or so. It affects a growing number of members of our Residents' Association and publishing our municipality's requirements is a task that should receive attention of the committee of which I am the chair. With some luck things will be less onerous than in CofCT.

Link to comment
Share on other sites

I think it is both accurate to say that NRS097-2-1 applies to this inverter, and that it doesn't. It applies in as much as it delegates to the correct standard. Or suppose you ask me for a recommendation for some or other product... then I am both involved (in that I told you about it) and not involved (you went and bought it from a supplier yourself, I didn't sell it to you). It depends on how you look at it.

Now to get back to what I was talking about earlier: I think the reason it cannot comply is because if you charge with the Axpert while the sun is also shining, then there exists a path via the electronics between the PV panels and the grid. The pipeline is bi-directional too. Now if you have a proper interlocking changeover so that PV and grid are never simultaneously connected, then the regulations leaves you alone. Therefore it might actually be an idea to attempt to go the other way: Actually test it according to the grid-tied requirements!

Those are:

1. Always disconnect if the voltage or frequency is outside the accepted values

2. Flicker levels below what is required by NRS 048-2.

3. DC injection below 1% of rating

4. Harmonics and distortion at acceptable levels

5. Power factor limited to 0.9 (for injected power)

6. Synhronisation requirements

Now many of these simply won't apply. Pretty much 2 - 6 are met because you don't do any of that. So it should be easy, right?

But I suspect a testing house will have to do it and make a certificate for it. And that is where the problem lies.

Edited by plonkster
Link to comment
Share on other sites

Thanks for the trouble you have taken @plonkster, much appreciated. What is interesting is that it seems that the Axpert charges from either the panel or the grid but not both if set to SBU or solar before utility. I suppose the problem arises when it is set to solar and utility charging when under, what seems somewhat obscure conditions, it may feed back into the grid. If all else fails a remotely controlled transfer switch, controlled by either the RPi or the BMV to switch the load between the Axpert and the grid and using only solar charging may solve all the legal complexities, at least in the most economical way. As I understand @The Terrible Triplett's past posts on the subject and the changeover unit that he had made by an engineer, rendered his installation perfectly legal. However I hope that Swartland municipality's regulations will make all that unnecessary.

 

Link to comment
Share on other sites

12 minutes ago, ebrsa said:

However I hope that Swartland municipality's regulations will make all that unnecessary.

Forget for a moment that their are panels involved.

Ask the question: What are the regulations to connect a UPS to a DB board?

12 minutes ago, ebrsa said:

... changeover unit that he had made by an engineer, rendered his installation perfectly legal.

100% correct - idea came from a old toppie from the UPS industry, built by a electrical engineer. Every single installer I saw last year said it is 100% secure and safe, all would have signed it off. At the time it was built, 1 second was FAST!

Today there are newer smarter changeovers in the market, installed direct on the DB, I'm told, by electricians who install UPS'es / solar / generators. Much faster than 1 second.

Edited by Guest
Link to comment
Share on other sites

17 hours ago, phil.g00 said:

I don't know this inverter at all, but if its only  230Vac and 50 Hz but out of phase its doesn't qualify as being in synch. 

That is very unambiguous.

Maybe its the loophole that's needed?

If we're talking about the Axperts, unfortunately, this is not a suitable loophole. Axperts do run in synchronism (in frequency but not voltage) with the AC input, as far as I know mainly to make switching from battery to/from utility as smooth as possible. The models capable of having their outputs paralleled also need to be synchronised, for obvious reasons, but they seem to synchronise to the AC input whenever it is available.

It's easy to test. In battery mode, with the AC input connected, use a multimeter to measure the difference in voltage between the AC in live and the AC out live (one multimeter probe on each live). It might read say 15 V, and jump around a little with loads on the utility network, but won't slowly increase and decrease and sometimes read over 400 V (which it would do if the input and output were out of phase).

I believe that the regulators' intention was to catch all the grid-tied inverters without being excessively wordy, and not to catch any off-grid inverters, but they have (presumably) inadvertently included the Axperts in this definition. In my opinion, of course, based merely on observation.

I do hope that some other loophole or exemption is found, but it seems to me, this isn't one.

Link to comment
Share on other sites

15 hours ago, ebrsa said:

what seems somewhat obscure conditions

Absolutely agreed. It would feed into the grid only in a fault condition, and obviously only if the grid is off (otherwise it will just blow up with a loud bang).

15 hours ago, ebrsa said:

remotely controlled transfer switch, controlled by either the RPi or the BMV to switch the load between the Axpert and the grid and using only solar charging

I believe that will make it off-grid and solve the issue completely. You lose the ability to charge with AC, of course, but most people don't want to do that (it costs money), and only rarely have to anyway. If you do live in an area with long periods of overcast weather, even a small 48V charger of a few amps should be enough to apply a "slow charge" to prevent the batteries from sustaining damage (Victron has a built-in slow-charge phase that kicks in after 24 hours of being left discharged, which charges at 5A, so you'd mimic that).

Making it off-grid is by far the cheapest too. You only need a CoC, the small amount of registration schlep... and then an official MAY show up to make sure you're telling the truth (but probably won't).

In fact, people often talk of the cost of a 48V charger. Compare it to the cost of the engineer that has to sign this off... I'm pretty sure you can obtain a good charger for 3k-5k 🙂

7 hours ago, Coulomb said:

I believe that the regulators' intention was to catch all the grid-tied inverters without being excessively wordy, and not to catch any off-grid inverters, but they have (presumably) inadvertently included the Axperts in this definition. In my opinion, of course, based merely on observation. 

I think you are absolutely right. They probably thought they were being generous by carving out a hole for the UPSes.

Link to comment
Share on other sites

  • Thanks for your comments @plonkster and @The Terrible Triplett. Below is a changeover switch at a reasonable price, which I think will cope with my 2 x Axpert cluster 8KWA total output capacity and available from Electro Mechanica in Montaque Gardens, Cape Town. @Mike has already installed an 8 KWA overload switch for me and the inverter output could be routed through that.  The DB board will be split between grid and off-grid. I will only put heavy loads like the oven (the stove is gas) and the geyser (with evacuated tube solar heater connected) on the permanent grid DB board. Most of this with the exception of the DB board split will only happen if that is the only solution to Swartland municipality's regulations, yet to be determined. I always find it more productive to have a discussion with the boss man instead of relying on views of the underlings. That is something we still manage where I live but I doubt that it will be that easy in CofCT. 
  • Socomec-9505-4008-4-pole-motorised-chang

9513 4004media.nl?id=736030&c=4501092&h=281527058

ATYS S 4P 40A 2X230VAC C/O SW

R6,487.15 (Excl. VAT)


List Price:
R6,487.15
Supplier:
Socomec
Stock Availability:
 In stock, can ship immediately

 I have a question about this product

 

ATyS dS motorised 4 pole changeover switches

Incorporates supply redundancy without the need for additional wiring, by integrating a double supply (2 independent supplies) directly within the product.

Features:

  • 2 x 230 VAC dual power supply
  • Fail-safe mechanical interlock
  • Padlockable in the OFF position
  • Wide-band AC control voltage ±30%
  • Manual override facility (handle supplied)
  • Can be configured for impulse or contactor logic
  • AC 22A IEC60947-3 (load break switch standard)
  • AC 32B IEC 60947-6-1, GB 14048-11 (transfer switch standards)
  • Fully independent position auxiliary contacts one for each position ( I / O / II )
  • Two switches mounted back to back, motorised (easily replaceable motors)
  • Energy efficient power consumption is minimal and only required during transfer operation
AC-31 B 415V (A) 40A
NUMBER OF POLES 4
SUPPLY VOLTAGE 2 x 230 VAC
DESCRIPTION ATyS S motorised changeover switch
DIMENSIONS (MM) (H) 144 x (W) 198 x (D) 181
Link to comment
Share on other sites

3 minutes ago, ebrsa said:

That is something we still manage where I live but I doubt that it will be that easy in CofCT. 

In CoCT, not CofT, we can email the powers that be direct, the senior oaks. ;)

FWIW, my changeover box was maybe 1/4 of that price. Here it is again, see how simple it is.

What made it back then +-R4800, was the then expensive BMV addition. The rest is off the shelf stuff, even the lights and box.

post-122-0-98399200-1441704401_thumb.jpg

Link to comment
Share on other sites

Thank you for the videos @plonkster, I have seen them before and more by the guy on the first link. What I have been unable to find, with Google searches, was interlocking contactors  with a capacity of about 40A, which is what I would need for my 8 KWA cluster output. I suppose one could use 2 sets of contactors in parallel but then the price will probably not differ much from the transfer switch I posted above. Whether they will switch in perfect synchronization to prevent frying one of the sets if the other set should switch faster, is a concern. The second link idea I like as I already have Arduinos to use and for me programmable is always preferred versus hard wired. If you know of a source for suitable contactors, I would appreciate it.

@The Terrible Triplett as a matter of interest, what amperage can your transfer switch handle.

I wonder if @pilotfish has found time to build his transfer switch using contactors yet. Perhaps he would post on progress if he spots this post.

Link to comment
Share on other sites

4 minutes ago, ebrsa said:

... , what amperage can your transfer switch handle.

I do not recall and the pictures are not informative either. 

But you can buy any size breakers / relays to fit your needs I would think.

Link to comment
Share on other sites

Very interesting topic change and then back again...

I respectfully disagree with Marius on the NRS application. Yes, IEC might be for Parallel Connected inverters, NRS is for Embedded Generation.

From NRS, 2017 version

image.thumb.png.6d76960dab5df6396f139d5bd7475511.png

Also, This little interesting bit of information. Axpert is not classified as UPS if PV panels are connected. It then becomes an Embedded Generator

image.thumb.png.a5e6389937e6ceeeea4560677583edf8.png

Then on why the Axpert would fail type testing (additional requirements might make it compliant):

image.thumb.png.59996895bd95e19ca7807101e18cce71.png

Why would a Pr E.ng not sign-off on a system that doesn't have the type tests? Liability... 

For Cape Town, they require NRS compliance even though it is not a national standard... If NRS097-2-2 comes out, then no more Pr.Eng sign off (probably).

 

 

 

Link to comment
Share on other sites

37 minutes ago, ebrsa said:

What I have been unable to find, with Google searches, was interlocking contactors  with a capacity of about 40A

Possibly a problem with your description. Search for "Reversing contactor", that's their more common application.

I prefer them for their electrical and mechanical interlock. I did a quickie and found some.

Link to comment
Share on other sites

1 hour ago, ebrsa said:

What I have been unable to find, with Google searches, was interlocking contactors  with a capacity of about 40A

Well... RS tends to be avoided because they are usually not the cheapest... but this is not a bad option. And if you search around a bit, there are suitable contactors of other brands that are even cheaper. And here is your interlock (both electrical and mechanical) and this pretty lady tells you how to use it. You don't need the special connection set (those are usually to neatly and easily build reversing and star-delta things for motors), but the wiring instructions for the electrical interlock is still useful.

I don't know what that other transfer switch goes for, but you're talking about 4k-5k to do it with contactors. I also half-suspect this will be faster, I cannot imagine that a motorised switch will switch faster than interlocked contactors.

How fast will it switch? No idea... but probably significantly slower than the 20ms max or so that most UPSes do.

Link to comment
Share on other sites

Thank you for the links @plonkster and advice from others. I posted details of the wrong transfer switch, the manual one. Below is the one recommended by Electro Mechanica and this is the motorised one with remote control. Changeover time is given as <90ms and of course the price is about the same as an Axpert. But if you are over a barrel this may be the cheapest option unless I can find suitable contactors.

I sure hope @pilotfish will put one together at a better price, that is if he did not emigrate in the meantime.

  • Socomec-9323-4006-4-pole-motorised-chang

9323 2006media.nl?id=736030&c=4501092&h=281527058

2X63A ATYS REMOTE CONTROLLED

R9,355.25 (Excl. VAT)


List Price:
R9,355.25
Supplier:
Socomec
Stock Availability:
 In stock, can ship immediately

 I have a question about this product

 

ATyS dM motorised changeover switches

ATyS dM modular motorised changeover switches with positive break indication enable on-load changeover switching of two supply sources in automatic or manual mode.

Features:

  • Modular format to easily mount into standard distribution boards
  • Provide rapid switching (<90ms) excellent dynamic withstand and long life
  • Fully compliant with IEC 60947-6-1 (standard governing transfer switches)
  • Supply only active during transfer, uneffected by network voltage fluctuations
  • Auto or manual operation (with incorporated Allen key) padlockable in O position
AC-31 B 415V (A) 63A
NUMBER OF POLES 2
SUPPLY VOLTAGE 230 VAC
DESCRIPTION ATyS d M motorised changeover switches
DIMENSIONS (MM) (H) 245 x (W) 235 x (D) 7
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...