Victron MPPT over panelling

[timselectric]

Nahh I’ve seen the light and jumped on Tim’s platform, over paneling is only when a SCC is able to exceed its published specs of VOC and ISC, argue with me all you want. ?

You are on your own, in that category. lol

 

[Kiwi2000]

Nahh I’ve seen the light and jumped on Tim’s platform, over paneling is only when a SCC is able to exceed its published specs of VOC and ISC, argue with me all you want. ?

If that was the case then no SCC manufacturer would recommend or endorse oversizing.

They talk about oversizing as when the moment you have more PV going to the controller than the controller can output at the system's voltage.

 

[740GLE]

If that was the case then no SCC manufacturer would recommend or endorse oversizing.

Agreed in my definition (and Tims) almost all SCC are very limited in over paneling.

Except for those midnight units that offer hyper voltage ratings.

 

[timselectric]

VOC , is a hard line limit. Exceeding it can be damaging.
Overpaneling has nothing to do with VOC.
It's only about amperage.

 

[Kiwi2000]

Agreed in my definition (and Tims) almost all SCC are very limited in over paneling.

Except for those midnight units that offer hyper voltage ratings.

That's because Tim said something incorrect at the beginning of this thread then tried to change the common understanding of overpaneling to cover his incorrect statement.

This is Victron's definition of overpaneling/oversizing...

Matching solar modules to MPPT charge controllers - Victron Energy

Life used to be so simple; in a 12V battery system you took a ‘12V’ solar module, watched carefully that the maximum PV current would not exceed the charge controller maximum current and the system would work. Unfortunately due to the fact, that with PWM controllers the PV module is not feeding...

www.victronenergy.com

"Oversizing a PV array

Oversizing a PV array means installing more peak power (Wp) than the maximum charge power of the chosen MPPT charge controller."

 

[timselectric]

Agreed in my definition (and Tims) almost all SCC are very limited in over paneling.

Except for those midnight units that offer hyper voltage ratings.

that is your category. Keep me out of it.
You are on your own. lol

 

[timselectric]

That's because Tim said something incorrect at the beginning of this thread

True, my information was not completely correct.

then tried to change the common understanding of overpaneling to cover his incorrect statement

Not true.
"Common" applies to the majority. (Not the outliers)
Overpaneling in all but a few (like Victron, and someone mentioned another) SCC's.
Means exceeding the rated input amperage.
As I said before. If you choose to call anything else overpaneling, that's fine.
It looks like Victron calls it oversizing. But the two terms could be considered similar.

 

[Kiwi2000]

True, my information was not completely correct.

Not true.
"Common" applies to the majority. (Not the outliers)
Overpaneling in all but a few (like Victron, and someone mentioned another) SCC's.
Means exceeding the rated input amperage.
As I said before. If you choose to call anything else overpaneling, that's fine.
It looks like Victron calls it oversizing. But the two terms could be considered similar.

It's also the common understanding when dealing with say grid tied inverters. 6kw of PV to a 5kw inverter is consider oversizing.

 

[sunshine_eggo]

Can't we all just get along?

 

[OffGridIdaho]

I can not see any of this on the vrm

You wont see it in history graph you have to be running live to see it.

 

[Whats_per_question]

I feel that some forum viewers might have missed some points on how the SCC works (this applies to Victron but maybe not other SCC);
All this is presumed to be at STC - Standard Test Conditions - and your conditions may be above this, but probably below it.

The panels produce a Voc when in full sun and disconnected, but this voltage always falls to a much lower level as soon as current is allowed to flow. The panels will produce their Isc (short circuit current) when shorted out, and this is the max current they could ever produce (at STC) but this is well below the max POWER they can produce - the max power is somewhere between zero amps, and Isc amps (and depends on the incoming radiation - ie it changes over time and temperature).

1) The first fact that i think some people are missing is that at maximum power (Pmax), the current is usually very close to the short circuit current. This is just one of the reasons that there is no point in putting fuses in the solar cables for the purpose of protecting the cable from short circuits like it is with fuses or breakers in AC systems or battery system. The panels just aren't capable of damaging the cables by short circuiting, and indeed the cables are regularly handling *almost* the short circuit current anyway in their normal operation. Whether an SCC periodically shorts the panel cables together (in silicon) i think is moot for the above reason.

When you commission your panels, you should be shorting them out - this allows you to test the Isc of the array and confirm that all branches of the array are indeed contributing. This is *critical* to do in parallel or series/parallel panel installations, otherwise you might commission the system, it seems to work fine, and for weeks/months/years, but the system is only using half or a third of the panels because you saw the controller working on day 1 and though "this is fine", when in fact you forgot to connect the cables to the 2nd, 3rd etc parallel string.

2) The second fact that some might be missing, is that the first limit in the Victron SCC name (ie the "30" in 100/30) is the current limit of the FETs on the battery side of the controller. This is a little bit of a simplification, but its a good working definition. The next limit that Victron tells you is the max Isc. This limit is usually a fraction higher than the "30" and tells you how much current the SCC can handle and still protect you from damaging the SCC by reverse polarity errors on the panel side. As some have mentioned in the thread, if you have your polarity right you don't have to worry about exceeding this limit.

3) The next limit is not discussed, but its the derating when the SCC gets very hot. I've seen suggestions that in hot climates or confined installations, that you manually derate your SCC, such as telling a 100/30 to limit itself to 25A. You also do this in situations where you have say 4 controllers that can do 30A each, but the battery will be sad if you poke it with 120A, so you derate the SCC manually to say 25A, so that on a great solar day you have a happy battery accepting 100A.

4) One poster made the excellent point that panels continue to drop in price (or at least faster than batteries and SCC) so this topic should stay hot for a long time, so lets thrash out the definition of "overpanelling", and then convert the masses to our agreed definition!

 

[BjornM]

I feel that some forum viewers might have missed some points on how the SCC works (this applies to Victron but maybe not other SCC);
All this is presumed to be at STC - Standard Test Conditions - and your conditions may be above this, but probably below it.

The panels produce a Voc when in full sun and disconnected, but this voltage always falls to a much lower level as soon as current is allowed to flow. The panels will produce their Isc (short circuit current) when shorted out, and this is the max current they could ever produce (at STC) but this is well below the max POWER they can produce - the max power is somewhere between zero amps, and Isc amps (and depends on the incoming radiation - ie it changes over time and temperature).

1) The first fact that i think some people are missing is that at maximum power (Pmax), the current is usually very close to the short circuit current. This is just one of the reasons that there is no point in putting fuses in the solar cables for the purpose of protecting the cable from short circuits like it is with fuses or breakers in AC systems or battery system. The panels just aren't capable of damaging the cables by short circuiting, and indeed the cables are regularly handling *almost* the short circuit current anyway in their normal operation. Whether an SCC periodically shorts the panel cables together (in silicon) i think is moot for the above reason.

When you commission your panels, you should be shorting them out - this allows you to test the Isc of the array and confirm that all branches of the array are indeed contributing. This is *critical* to do in parallel or series/parallel panel installations, otherwise you might commission the system, it seems to work fine, and for weeks/months/years, but the system is only using half or a third of the panels because you saw the controller working on day 1 and though "this is fine", when in fact you forgot to connect the cables to the 2nd, 3rd etc parallel string.

2) The second fact that some might be missing, is that the first limit in the Victron SCC name (ie the "30" in 100/30) is the current limit of the FETs on the battery side of the controller. This is a little bit of a simplification, but its a good working definition. The next limit that Victron tells you is the max Isc. This limit is usually a fraction higher than the "30" and tells you how much current the SCC can handle and still protect you from damaging the SCC by reverse polarity errors on the panel side. As some have mentioned in the thread, if you have your polarity right you don't have to worry about exceeding this limit.

3) The next limit is not discussed, but its the derating when the SCC gets very hot. I've seen suggestions that in hot climates or confined installations, that you manually derate your SCC, such as telling a 100/30 to limit itself to 25A. You also do this in situations where you have say 4 controllers that can do 30A each, but the battery will be sad if you poke it with 120A, so you derate the SCC manually to say 25A, so that on a great solar day you have a happy battery accepting 100A.

4) One poster made the excellent point that panels continue to drop in price (or at least faster than batteries and SCC) so this topic should stay hot for a long time, so lets thrash out the definition of "overpanelling", and then convert the masses to our agreed definition!

Great explanation.

I'd add that the controller will only get to the maximum power point when the output current is lower than the max rating (or if it is derated in some way). If the controller is already at the max output current limit, it will not be in the MPPT-point, but most likely at a higher voltage/lower current point. I doubt that the MPPT algorithm is even run at this point, because it would be pointless.

So in theory you can put a 1 MW array on a very small SCC as long as you put a fuse in series with the array so the max Isc is not reached. This array will not be in a MPPT-point 99.99% of the time. In practical terms it would be totally pointless to do it, but it is still possible to do and stay within the specs of the SCC.

 

[Kiwi2000]

3) The next limit is not discussed, but its the derating when the SCC gets very hot. I've seen suggestions that in hot climates or confined installations, that you manually derate your SCC, such as telling a 100/30 to limit itself to 25A. You also do this in situations where you have say 4 controllers that can do 30A each, but the battery will be sad if you poke it with 120A, so you derate the SCC manually to say 25A, so that on a great solar day you have a happy battery accepting 100A.

I've done that on the two arrays that I oversized by 200%. I have a 150/60 set to 50A and a 100/50 set to 45A. They run significantly cooler that way and the bus bar that they're attached to isn't getting as overloaded. (It has other SCCs going to it as well)

The benefit of oversizing PV even if needing to derate the SCC is more consistent production through the day and from day to day.

 

[Whats_per_question]

Great explanation.

I'd add that the controller will only get to the maximum power point when the output current is lower than the max rating (or if it is derated in some way). If the controller is already at the max output current limit, it will not be in the MPPT-point, but most likely at a higher voltage/lower current point. I doubt that the MPPT algorithm is even run at this point, because it would be pointless.

So in theory you can put a 1 MW array on a very small SCC as long as you put a fuse in series with the array so the max Isc is not reached. This array will not be in a MPPT-point 99.99% of the time. In practical terms it would be totally pointless to do it, but it is still possible to do and stay within the specs of the SCC.

Bjorn; That's a good point, but lets take that (impractical) idea further; what size fuse would you use? Lets say you have a 100/30, and your panels are a massive array of panels with pairs of my current favourite panel (
430w N-Type TOPCon, with 30 year linear power output warranty ) which has a Voc of 38v (so your Array Voc is 76v - no risk of going over the SCC voltage limit even in the coldest climates) and a panel Isc of 13.84A (Imp = 13.23A). Lets say your array has 500 pairs of these panels, so you have 1000 x 430w = 430,000Wp (or 0.43 MWp - a small solar farm by modern standards even here in New Zealand)
Your array (at STC) can create 500 x 13.84A = 6,920A at Isc, and at max power you could be flowing 6615A.
Your controller ( Victron 100/30 ) has these specs;

@BjornM - What is your fuse size, and what is the reasoning(s) for this?

I'll start with some issues with this array;
- If part of the array is shaded, you will have a large amount of voltage on the positive cable on one side of the array trying to find the shortest/easiest path to the negative cable, putting stress on the panel diodes on the shaded side? There is no control of the current at the 500-to-one combiner box.
- Even though you have a massive array, its not going to turn moonlight into 30A at the controller - there is a significant hurdle to getting current flowing, so even though in near darkness, with a panel facing *away* from the light source (at least for panels i've tested) you can still get a good voltage (i've seen 30v on a 40v panel), but the slightest attempt to draw power results in zero current.
- We already knew it was a fantasy array, but you've paid about NZD$430,000 (for a while its been an easy formula that panels in NZD are about $1/watt) for an array that produces about 8.6kwh per day, netting you about $2.95, giving you a rough payback period of 454 years. Maybe you need 2 controllers?

 

[Pi Curio]

I've done that on the two arrays that I oversized by 200%. I have a 150/60 set to 50A and a 100/50 set to 45A. They run significantly cooler that way and the bus bar that they're attached to isn't getting as overloaded. (It has other SCCs going to it as well)

The benefit of oversizing PV even if needing to derate the SCC is more consistent production through the day and from day to day.

+1

600W array with 100/20 @12V set to 18A(-10%) - 8 full hours of set peak production in summer with stationary array. The SCC runs only slightly warm to the touch.

 

[BjornM]

Bjorn; That's a good point, but lets take that (impractical) idea further; what size fuse would you use? Lets say you have a 100/30, and your panels are a massive array of panels with pairs of my current favourite panel (
430w N-Type TOPCon, with 30 year linear power output warranty ) which has a Voc of 38v (so your Array Voc is 76v - no risk of going over the SCC voltage limit even in the coldest climates) and a panel Isc of 13.84A (Imp = 13.23A). Lets say your array has 500 pairs of these panels, so you have 1000 x 430w = 430,000Wp (or 0.43 MWp - a small solar farm by modern standards even here in New Zealand)
Your array (at STC) can create 500 x 13.84A = 6,920A at Isc, and at max power you could be flowing 6615A.
Your controller ( Victron 100/30 ) has these specs;
View attachment 190892
@BjornM - What is your fuse size, and what is the reasoning(s) for this?

I'll start with some issues with this array;
- If part of the array is shaded, you will have a large amount of voltage on the positive cable on one side of the array trying to find the shortest/easiest path to the negative cable, putting stress on the panel diodes on the shaded side? There is no control of the current at the 500-to-one combiner box.
- Even though you have a massive array, its not going to turn moonlight into 30A at the controller - there is a significant hurdle to getting current flowing, so even though in near darkness, with a panel facing *away* from the light source (at least for panels i've tested) you can still get a good voltage (i've seen 30v on a 40v panel), but the slightest attempt to draw power results in zero current.
- We already knew it was a fantasy array, but you've paid about NZD$430,000 (for a while its been an easy formula that panels in NZD are about $1/watt) for an array that produces about 8.6kwh per day, netting you about $2.95, giving you a rough payback period of 454 years. Maybe you need 2 controllers?

For that fantasy setup I'd just go with the max Isc of the controller: a 35 A fuse in series with the array. The fuse also has to have a breaking capacity of 7 kA, so any ordinary fuse wouldn't do. Because the array is so massive, 99.99% of the time it would be close to Voc, even at max controller current. If we assume a 24 V battery, the output would be around 24 V * 30 A = 720 W, so the array current would be around 9.5 A and not blow the fuse.

Dusk and dawn the controller will be at maximum power point. I've also seen exactly what you describe in low light conditions. Starts with a very low voltage of just 1 V, climbs rapidly to 30 V (per panel) but without the ability provide any useful current until 30 V. But the array is so huge that we enter unknown territory here. If it so happens that the array can give enough current at say 14 V (per panel), and the MPPT finds the point there, then the current will be 26 A. Lower panel voltages than 14 V can not be used by the controller, because it is a buck converter (the input voltage must be higher than the output). So it will still not blow the fuse, and that's as close as we ever will get to it. Well, except for the rare occasion where the controller shortens the array to protect itself. Worst case scenario 7 kA will flow, and if we picked a cheap fuse it will not be able to break that massive current, a plasma arc will form across it, the controller will catch fire, and the house will burn down. Yikes.

As for the shading problem, I'd put a blocking diode in series with each string before they are joined together in the combiner box. No wasted energy flowing into shaded panels, and also a more important function: if a string breaks and becomes a short circuit, the current from the other strings will not flow into it and cause a fire. Usually string fuses are used for this, but blocking diodes are even better. Or both, which I have in my own installation.

you've paid about NZD$430,000 ... Maybe you need 2 controllers?

Yes, at least!!! ?

 

[Whats-n-Watts]

The panel's Isc is 11.89, so 23.78 amps short circuit for the string.

I have seen isc multiplied by 1.25, 1.56 as well as 2. Can you tell me (I'm certain you can) when to use 2x? I understood it to be 1.25 for intermittent loads and 1.56 for continuous loads but your the first that I have have seen so far to use 2x. I been working on learning. Finally have all my equipment so I want to make sure I'm in the right table. TIA

 

[Whats-n-Watts]

The panel's Isc is 11.89, so 23.78 amps short circuit for the string.

 

[Hedges]

1.25x is the normal margin above continuous current to avoid nuisance tripping of thermal protection. (magnetic-hydraulic breakers might allow 1.00% or 1.05%)

For PV panels, an additional 1.25x is for extra illumination due to cloud edge effects.

1.25 x 1.25 = 1.56
NEC says to set OCP to at least 1.56x Isc, and wire ampacity must be at least that high.

As for what to consider regarding overpaneling an inverter, I think most of us use 1.0 x PV panel data sheet Isc rating. (Unlike for Voc, where we must adjust for cold.)

Don't connect PV string backwards, and if you do, recognize the mistake and disconnect a short time later. If multiple PV strings in parallel, leave the rest disconnected when first connecting inverter.

 

[Kahdeksan_minuuttia]

I have been playing with the MPPT sizing calculator by Victron (Victron MPPT sizing calculator example) with some random 420 W panels. This is just early stage planning, but I have been thinking of 48v ESS with Multiplus II in three phase setup and 6-8 kWp solar panels with Victron MPPTs.

The specs of a random panel are:

• Voc, 38,14
• Isc, 13,77
• Vmpp, 31,83
• Impp, 13,04
• TC Isc, +,03 %/°C
• TC Voc, -0,28 %/°C

For a setup with 2 string and 3 panels in series in each string the voltages are PV max. voltage @ min. temperature 128,8 V and PV min. voltage @ max. temperature 80,1 V.

The calculator recommends SmartSolar 150/35. Both PV max. current @ MPP min. temp. and PV max. current @ MPP max. temp. are limited to 35 A (126 %, oversized).

The second recommendation is SmartSolar 150/45 with PV max. current @ MPP min. temp. limited to 45 A and PV max. current @ MPP max. temp. 44,4 A (97 %, undersized).

The third recommendation is is SmartSolar 150/60 with PV max. current @ MPP min. temp. 58,3 A and PV max. current @ MPP max. temp. 44,4 A (73 %, undersized).

I would be happy to get 15 - 20 panels, preferably 18 or less. There are some shadows during the morgning hours and most of the winter time the panels would probably be covered by snow.

Three 150/45 with 6 panels in each seems the way to go. Are the calculations correct and are there some other aspects to consider? The maximum PV short circuit current for 150/45 is 50 A, how do I review this one? 150/45 with PV max. current @ MPP min. temp. limited to 45 A is over panelled (or is it?), but that's ok?