I know my inverter is rated for 120-430V operating voltage with 450 open circuit max.
I currently have  9x405 panels in series with ~37,5V open circuit which is 337,5V max open.
System works like a champ on these settings.

I want to add 2 more panels, which will take me to 412,5 Open circuit, which is somewhat close to the 450 max.
This configuration would be ideal for my consumptions and 11kwh battery.
Is it safe enough? Will i have peaks on cold/sunny days above that based on others experience with your systems?

My axpert is close to running 2 years without problems, and i would like it to do another 2-3 years before i retire it/replace it, so running smoothly is of essence.

Any advice appreciated. 
Chart of past 7 days

Edited May 10, 20242 yr by Leog

I'd wait for others to confirm but I wouldn't worry about it. You're still well under the limit and the chances of getting anywhere close to open circuit voltages is very slim in our climate (unless you live in the middle of the Karoo). 

In my experience pushing a string as close to the MPPT's max operating voltage as possible is the most efficient way of doing things.

I would not do it. On real bold spells you might just exceed the 450V. Remember that 412V creeps higher for each degree drop in temp from 25°C. 

Compare it to a car you take up to the red line in each gear. It can but won't last as well as normal driving. When it pops one tends to bad mouth the car. 

Edited May 10, 20242 yr by Scorp007

Yes i agree with @Scorp007 don't do it at 0°C your Voc for 11 panels will be 438.8V much to close for comfort. My motto 50v bèlow mppt voc spec on high voltage mppt's 

 

Edited May 10, 20242 yr by TaliaB

58 minutes ago, LiFePOWER said:

You're still well under the limit and the chances of getting anywhere close to open circuit voltages is very slim in our climate (unless you live in the middle of the Karoo). 

 At hotter conditions the Voc decrease at colder temps your Voc increaces on the panel specs it is called Temperature Coefficient of VOC. 

You're never ever going to get panels below 25C with 1000W/m irradiance in South Africa. It's May and we're down to less than 800 already which is what NOCT (/NMOT, depends on the manufacturer) is measured at, and panels getting this aren't sitting at below 20C either.

NOCT open circuit is 35.12V for those panels so 386V for 11, well within the limits. 

Even NOCT is conservative and you can push the string voltage higher, not that it's recommended though and I'd use NOCT as the actual Voc limit. You also have to consider peak summer days and the max operating voltage of the MPPT, no point pushing Voc to max on a winter day when your MPPT starts clipping in the middle of Dec.

Why? The AM value, the only time a panel in South Africa is going to be cold enough to hit even the NOCT Voc value is first thing in the morning after a cold evening in winter, the AM value skyrockets at this time because of the amount of atmosphere the sunlight has to go through (red shifting it and therefore decreasing the irradiance) massively reducing the voltage the panel is producing.

NOCT test is at AM1.5, this is at a sun zenith of 45 degrees no minimal dust/pollution, Sutherland (which is where you're most likely to be cold and sunny) loses 45 degree zenith at midday before the beginning of May.

TLDR: use NOCT not STC as the limits when calculating string voltages.

Edited May 10, 20242 yr by LiFePOWER

9 minutes ago, TaliaB said:

 At hotter conditions the Voc decrease at colder temps your Voc increaces on the panel specs it is called Temperature Coefficient of VOC. 

 

Yes, but Karoo winter mornings = very cold and quite sunny. Which is where Voc likes to play, even then as above he'll be good with 11.

29 minutes ago, LiFePOWER said:

You're never ever going to get panels below 25C with 1000W/m irradiance in South Africa. It's May and we're down to less than 800 already which is what NOCT (/NMOT, depends on the manufacturer) is measured at, and panels getting this aren't sitting at below 20C either.

NOCT open circuit is 35.12V for those panels so 386V for 11, well within the limits. 

Even NOCT is conservative and you can push the string voltage higher, not that it's recommended though and I'd use NOCT as the actual Voc limit. You also have to consider peak summer days and the max operating voltage of the MPPT, no point pushing Voc to max on a winter day when your MPPT starts clipping in the middle of Dec.

Why? The AM value, the only time a panel in South Africa is going to be cold enough to hit even the NOCT Voc value is first thing in the morning after a cold evening in winter, the AM value skyrockets at this time because of the amount of atmosphere the sunlight has to go through (red shifting it and therefore decreasing the irradiance) massively reducing the voltage the panel is producing.

TLDR: use NOCT not STC as the limits when calculating string voltages.

I do think you are making a mistake by thinking the irradiance in the whole of SA is at 1000W/m² even in winter. 

Without looking at the coldest areas there are big parts in Mpumalanga that gets below 0°C and also no sun. Also some inland areas in the Easter Cape. 

Refer to the lows below 0° on this map. 

I am happy to be corrected but one should stay in a place for a few yrs to really know the temperatures. 

1 hour ago, Scorp007 said:

I do think you are making a mistake by thinking the irradiance in the whole of SA is at 1000W/m² even in winter.

I'm saying the irradiance is much less than 1000W/m when it's cold in SA which makes using the STC Voc as the limit for installations incorrect. 

To hit the STC limit you need:
Panel temp 25C or below
1000W/m
Sun zenith of 45 degrees or higher = AM1.5

Worst case scenario, let's say Sutherland, cold/sunny and be overly cautious:
1. Today 18C and say panels at 20C (remember it's cell temp that matters, not ambient, things in the sun get warm, solar cells especially so), 2C above ambient, NOCT (barely) and STC temp check!
2, 1000W/m, it's mid May and I'm only getting 700W/m in Pretoria, would need a CT forumite to confirm irradiance down that latitude. Again overly cautious and say 700W/m, NOCT irradiance fail, well below STC fail.
3. Zenith 40 degree, NOCT and STC fail. Sutherland lost 45 degree zenith at midday 2 weeks ago, never mind when the panels were cold.

Yes it'll get colder but AM/irradiance makes a MUCH bigger difference than temperature to a panels output. To increase irradiance you need to go north in winter. The north of SA loses 45 degree zenith beginning of June, and as it's Musina it's hardly troubling the temperature coefficients.

The only place I can think of in SA where it's going to have a zenith of 45 degree, panel temp below 20C and an irradiance of even 800W/m all at the same time would be a severe cold front with a bit of cloud edge in the escarpment above Louis Trichardt in July and that's just the NOCT limit, would have to be 45 degree, 20C and 1000W/m for STC.

TLDR: use the nominal operating data (NOCT/NMOT) not the standardised test data (STC) for installation calcs. STC measurements are not feasible in the real world and is basically a marketing gimmick to say a panel is X watts, you'll never approach this.
 

EDIT: and to be clear I'm not saying set your string to a NOCT Voc 2V under the inverter limit, be reasonable, but taking 10%+ of the STC margin "as a rule of thumb" is wasting efficiency, maximising PV voltage is the key, everything else cascades after that.

Edited May 10, 20242 yr by LiFePOWER

18 hours ago, LiFePOWER said:

I'm saying the irradiance is much less than 1000W/m when it's cold in SA which makes using the STC Voc as the limit for installations incorrect. 

To hit the STC limit you need:
Panel temp 25C or below
1000W/m
Sun zenith of 45 degrees or higher = AM1.5

Worst case scenario, let's say Sutherland, cold/sunny and be overly cautious:
1. Today 18C and say panels at 20C (remember it's cell temp that matters, not ambient, things in the sun get warm, solar cells especially so), 2C above ambient, NOCT (barely) and STC temp check!
2, 1000W/m, it's mid May and I'm only getting 700W/m in Pretoria, would need a CT forumite to confirm irradiance down that latitude. Again overly cautious and say 700W/m, NOCT irradiance fail, well below STC fail.
3. Zenith 40 degree, NOCT and STC fail. Sutherland lost 45 degree zenith at midday 2 weeks ago, never mind when the panels were cold.

Yes it'll get colder but AM/irradiance makes a MUCH bigger difference than temperature to a panels output. To increase irradiance you need to go north in winter. The north of SA loses 45 degree zenith beginning of June, and as it's Musina it's hardly troubling the temperature coefficients.

The only place I can think of in SA where it's going to have a zenith of 45 degree, panel temp below 20C and an irradiance of even 800W/m all at the same time would be a severe cold front with a bit of cloud edge in the escarpment above Louis Trichardt in July and that's just the NOCT limit, would have to be 45 degree, 20C and 1000W/m for STC.

TLDR: use the nominal operating data (NOCT/NMOT) not the standardised test data (STC) for installation calcs. STC measurements are not feasible in the real world and is basically a marketing gimmick to say a panel is X watts, you'll never approach this.
 

EDIT: and to be clear I'm not saying set your string to a NOCT Voc 2V under the inverter limit, be reasonable, but taking 10%+ of the STC margin "as a rule of thumb" is wasting efficiency, maximising PV voltage is the key, everything else cascades after that.

You raise a number of valid points that are not always quantifiable. The point of ambient vs panel temps for me is before panels generate any heat in the 1st light is that panels will be at ambient temp. Even this morning at 08h in Pta with string generating 20W the panels were only 3°C higher. 

I have also raised the fact that when panels are connected to the MPPT one will never see Voc. The risk remains when during the worst part of the day when they get disconnected and cooling down the moment you reconnect that's when you are at risk. 

For me it's also a matter of tolerance of components used. Better to have an extra margin than to risk a MPPT. 

A real measurement @ 08h this morning was Voc=213V and the moment you connect to the MPPT it dropped to 168V producing 20W.

Thus I personally belief in rather not chasing maximum power but have a more reliable system that can work for yrs to come. Have we not seen how many guys chase saving just to find batteries discharged too low and system shutting down. 

It remains an interesting topic with just so many variables that are difficult to calculate to the last Volt or Watt. Play it safe!!!! 

Hey guys, i am in Greece, no South African here, but you got a great forum, and have helped me a lot in DIYing my system with all the knowledge.
It will be very difficult to get full irradiance (sunny) with 0c where i am.
Heck we almost never have 0 with SUN. We can go 0 or negative only in the nights with clear skies.

From your answers though i get its better to play it safe and not go 11.

It doesn't matter that the irradiance at 0°C is weak; it will eventually charge the capacitors in the solar charger and these will store enough energy to cause avalanche in the semiconductors, and slowly damage the electrolytics if the voltage is over their rating.

@TaliaBI like your attitude on this. Make a personal hard limit & have your reasons for it & don't venture past this limit...