10% either side. NRS097 allows the PV-inverter to remain connected even if you go outside the 10% margin (up to 40 seconds on the upper side), but outside 20% margins it disconnects much faster.
1 Like
Different countries that slightly different versions of essentially the same thing. The 10% ish theme is more common, but the disconnect time tends to vary.
There are also frequency tolerances and associated times.
I did trawl through various countriesâ codes at one time and some codes are more tolerant than others.
I canât remember the specifics of why, but I did form the impression that ZAâs codes were fairly stringent, but not too far off the general levels.
If I am not mistaken, the Victron inverter software is as easy a place as any to see a summary of the various codes in one place.
1 Like
Would this necessarily be the case? My understanding is that the cables would be able to accommodate only a certain percentage of the maximum load all the houses can draw at once. So instead of sizing the supply to be able to give 63x(number of houses) of amps (if their main breakers are 63A) it would be sized something like 10x(number of houses) because it is more practical, cost effective and should achieve the same thing.
The problem is that, while the draw of all the houses might have been taken into account when determining say the â10â, what wasnât taken into account was the amount of current that can flow upstream, because that is a function of the invertersâ capacities and not the neighbourhoodâs instantaneous demand.
Well, Iâd be very surprised if more than âa certain percentageâ of houses starts to push full power into the grid from their PV arrays. So technically, yes, the assumption for down vs up isnât necessarily symmetrical, but I still donât really think there is a big problem with the wire sizing.
Probably why it is so important to regulate maximum inverter sizes. PV is quite concentrated in an area, so would be neighbourhoods and wealth. I can imagine an instance where most houses in an affluent neighbourhood has oversized PV arrays with sufficiently large inverters to service a large chunk of their load, but fairly little load during the day (when affluent people should probably be out working). Anyways, donât need to be the most important reason to disallow grid feedback, but probably one to limit it to the % of maximum current that was allowed for in the downstream calculation.
The customers are quite entitled to draw full load simultaneously. They pay for it.
That is not an acceptable argument.
I donât know what the supply practices are at municipality street level, but a municipality wouldnât let their own supplier get away with that practice.
In reality though, local generation would actually alleviate the burden on the cabling for many years, long before that particular concern became an issue.
Two days to go for @GwedeMantashe1 to publish the new license exemption of 100MW. Is he going to move the goalposts by preventing wheeling and selling to non related entities?
1 Like
Well looks like the only thing that I need to be considered a dollar millionaire is a million dollars 
Yes I believe in a domestic situation they estimate the maximum current that would be consumed.
My son lived in a complex of 4 story student flats. There were a number of these buildings and they were all fitted with 3 phase 24kW instantaneous hot water systems. Now what is the possibility that all residents happen to be showering at once??
My understanding is that an industrial installation will be sized for maximum possible power consumption.
Apparently a problem comes in when what used to be large erfs are split up into many smaller houses. The grid wasnât designed for that additional load on the cables. So image there were 10 such erfs in a neighbourhood, each with one house. Now there might be 100 houses. If the fees charged by the municipality when such splits happen do not get used to improve the distribution network, you will start getting problems.
And from someone who used to work in this space, I understand that this often times were the caseâŚ
Iâve seen way worse than that. Iâve seen an old ârailway houseâ that was on a large plot demolished to be replaced by a block of 2-bedroom flats, about 80 of them⌠
Haha! Or something like that.
Somewhat on topic: Iâve got this weird thing that happens, not really sure it is an issue, but when I start my microwave and I turn on the kettle (or toaster) on the same circuit, the microwave goes from making more of a higher frequency humm to more of a lower frequency, deeper humm.
Iâve had a look at my grid voltage change when that happens. Basically only the microwave on dropped the grid voltage by 2V and then the kettle also dropped the grid voltage by another 2V. Nothing much else were on in my house, and the grid voltage was effectively at 226V when both were on.
Now, is this normal? I havenât measured the voltage on the microwaveâs circuit itself, potentially thereâs going to a bit more of a drop due to thinner cables.
Is this something to worry about?
Mine does exactly the same, and the change in tone is clearly because of the drop in voltage. It doesnât appear to affect the machine in any other way, I think the fan just slows down a little.
My voltage can be anything betweem 216V and 235V. The Sonoff POW2 in one of the back flatlets reported 212V the other evening (and there wasnât even any load on that side), which I think is partly poor calibration and partly voltage sag in the neighbourhood. What I should actually do is ask my neighbour across the street what his VRM site is so I can compare hisâŚ
1 Like
Oh! I can also have a look at my neighbourâs. My neighbour living next to the kiosk in our street also has a Victron system courtesy of Jaco. It could also be interesting to see if his power usage affects my voltage. I donât think we would branch off the same phase thoughâŚ
Would might be interesting is to put in some water and measure the change in temperature between a run with the volt drop and one without to see if it affects the energy output.
At least it is good to know that I donât need to worry too much!
The part Iâm interested to watch is, over time, that you guys start reporting even lower AC voltages ⌠i.e. specifically thinking of when Eskom comes under even more strain to keep the overall voltages inline.
Well, I moved house. At my previous house it was pretty much rock-solid around 237V, give or take a few volt. But that neighbourhood was born in 2000⌠this one was born in the 70s
Voltage isnât the biggest problem. You can compensate for that with tap changes on transformers. Phil already explained that at certain points in the network, this already happens automatically. So I donât really see voltage as a reliable measure, at least not a linear indicator. What I find more interesting is the frequency. A healthy grid connection seems to run just a tad over 50Hz, but what Iâve discovered is that during load-shedding I often see it run higher than normal, or sometimes slightly below 50Hz. That is to say, you see slight frequency deviations from normal when there is strain. That makes complete sense to me too, not only because you are running closer to the limit, but also because you are managing neighbourhoods coming online and others going offline, large blocks of energy changing hands around the 2-hour mark.
You are right ⌠I âthoughtâ Hz but wrote all about the volts. My bad.
The older neighborhoods would be the first to âshowâ national degradation methinks as the equipment is under more âstrainâ and/or being âolderâ so IF there is drama coming, methinks the Hz will start to fluctuation there first, the âolderâ transformers the first go.
Just a theory I have.
I remember when I was still a student I had problems with my 14" CRT monitor at some stage.
When we ended up measuring the power voltage in the house it was 172V.
That CRT lasted many more years, but was never the same. It started going green at random times like my very first XT with itâs old monochrome monitor (I see a few people googleing what an XT is⌠
)
If that happend you could smack the screen with a few hard claps for the colour to return.
Now there is nothing more annoying when you arranged your pals for a LAN (that was before internet gaming) and you are just about to pull the trigger on someone and the screen goes all shades of green and you canât see if that target is in the blue or red team. Perhaps I hit it too hard a few times then?
After a few years no amount of clapping help anymore and it was time to be replaced. Luckily they just started selling a new display technology: 14" LCD screens. Wohoo 
I think the refresh rate those days was 20ms 
Now that you googled what an XT is, it might be worth mentioning that the CPU ran at 4.77 MHz.
Mine had a little Turbo button that would take the CPU to 10MHz. You had to switch it off with some programs like games, as they were timed on the 4.77MHz and so your player character would run 2x faster at 10MHz and was uncontrollable. Hehe
It came with a hard drive built in. A full 20Mb! Peanuts today.
Yes indeed!
Some years back when we were having serious electricity supply issues I was privy to a senior Eskom engineerâs SMSs.
These automated messages were reporting on the generation frequency of all the power stations.
If the frequency dropped to a critically low figure a SMS would be sent to all the MMWCs.
The screen was full of these alarmsâŚ
All generators should run at the same frequency and if they arenât then this is the real indicator of a grid that is unstable. 