Reliable product to replace Chinese all-in-one inverter

[Brucey]

I think most are made in China. Maybe Victron is not.

Victron is generally built in India.

 

[timselectric]

OK, I will look into their options then (wanted to check them anyway). Looks cheap, but mid-range quality wise, though purely Chinese company.

I'm currently running 3 of them. 1 was turned on over 2 years ago.
When my system is complete, there will be 12.

 

[sunshine_eggo]

@Oak77

Wanted to elaborate on Tim's answer with some terminology:

Grid-tied = grid is required. Unit is interactive with grid. It synchronizes with grid and can draw from/send power to the grid. Most operate without batteries.

Off-gird = no grid required. It is a complete system without a grid attached; HOWEVER, many can be connected to grid and ONLY draw from grid if backup power is needed. Most require batteries.

 

[timselectric]

@Oak77

Wanted to elaborate on Tim's answer with some terminology:

Grid-tied = grid is required. Unit is interactive with grid. It synchronizes with grid and can draw from/send power to the grid. Most operate without batteries.

Off-gird = no grid required. It is a complete system without a grid attached; HOWEVER, many can be connected to grid and ONLY draw from grid if backup power is needed. Most require batteries.

Gotta be Mansplaining everything I say. lol

 

[Brucey]

Some of those aren't on the European market (EG4 for example).



Not in Europe.

Deye should be much cheaper versus Solark in Europe

 

[Oak77]

You can split the panel wiring near the array

I would still go with Victron depending on what server rack batteries you have

I'd love to go with Victron, just struggling find/confirm right product from their portfolio.
Not sure about "splitting" at the array. Do you mean to create two parallel arrays and feed them back using same cables? I think that would make things worse.
DIsclaimer: All calculations are rough and shouldn't be followed; I myself have to yet tripple check everything.
The situation with 1 string is, that 9*48V=432V (that's nominal voltage, with Voc is 49.5V -> 445V for 9 panels), which happens to be the limit based on battery short circut voltage set by Victron = 54V*8=432V. The I=4095/432=9.5A. If I split it to let's say 4+4 panels, I'll get 192V, but 19A and anything above 13A voids the warranty. In contrary, someone on Victron forum wrote me, that it's OK to leave it in one string, with voltage under 450V and current under 13A and that it doesn't really matter if it's 3KW or 4KW. But the product sheet does state that 3KW is supported...
If I can utilize 8 panels and be safe now, it would be a good compromise, though. 9 seems to be too much on the edge even ignoring 3KW "designation".

 

[Oak77]

Gotta be Mansplaining everything I say. lol

Ah, my mistake, apologies, I misunderstood the terms.
Anyhow, so far a quick search revealed that almost all Growatt available in EU AIOs are 3 or 5KW. I can see they offer 6KW via Aliexpress, which I want to avoid. 5KW would be of course sufficient, but I don't trust them to really stick to it. Even Victron has reduced output to 4800W in 32˚C (if I remember it correctly, with full power under 25˚C.

 

[sunshine_eggo]

Gotta be Mansplaining everything I say. lol

and spelling it wrong...

damn off-gird inverters!

 

[Oak77]

Deye should be much cheaper versus Solark in Europe

Did you mean Dyness? I've never heard about them and I'm not sure if their quality and support is much different from SunStone, They're all exploding in confidence, having tons of certificates, but the reality can be quite harsh.

 

[robbob2112]

I'd love to go with Victron, just struggling find/confirm right product from their portfolio.
Not sure about "splitting" at the array. Do you mean to create two parallel arrays and feed them back using same cables? I think that would make things worse.
DIsclaimer: All calculations are rough and shouldn't be followed; I myself have to yet tripple check everything.
The situation with 1 string is, that 9*48V=432V (that's nominal voltage, with Voc is 49.5V -> 445V for 9 panels), which happens to be the limit based on battery short circut voltage set by Victron = 54V*8=432V. The I=4095/432=9.5A. If I split it to let's say 4+4 panels, I'll get 192V, but 19A and anything above 13A voids the warranty. In contrary, someone on Victron forum wrote me, that it's OK to leave it in one string, with voltage under 450V and current under 13A and that it doesn't really matter if it's 3KW or 4KW. But the product sheet does state that 3KW is supported...
If I can utilize 8 panels and be safe now, it would be a good compromise, though. 9 seems to be too much on the edge even ignoring 3KW "designation".

Splitting at the array means you have two arrays and the wires between the arrays are not combined into a single larger wire each wire from the distinct array runs back to where the MPPTs live. Each MPPT needs to connect to 1 array or you can string the arrays together and form 1 array, but you can't take a combined wire and connect it to two different MPPT.

You would have two arrays if for instance you had some panels faced east and some west. Each direction would need its own MPPT. For that each array has a pair of wires connected to each MPPT and you have two MPPT.

Edit - this goes for the single channel smaller MPPT verse the 450/100 described below.

 

[Brucey]

I'd love to go with Victron, just struggling find/confirm right product from their portfolio.
Not sure about "splitting" at the array. Do you mean to create two parallel arrays and feed them back using same cables? I think that would make things worse.
DIsclaimer: All calculations are rough and shouldn't be followed; I myself have to yet tripple check everything.
The situation with 1 string is, that 9*48V=432V (that's nominal voltage, with Voc is 49.5V -> 445V for 9 panels), which happens to be the limit based on battery short circut voltage set by Victron = 54V*8=432V. The I=4095/432=9.5A. If I split it to let's say 4+4 panels, I'll get 192V, but 19A and anything above 13A voids the warranty. In contrary, someone on Victron forum wrote me, that it's OK to leave it in one string, with voltage under 450V and current under 13A and that it doesn't really matter if it's 3KW or 4KW. But the product sheet does state that 3KW is supported...
If I can utilize 8 panels and be safe now, it would be a good compromise, though. 9 seems to be too much on the edge even ignoring 3KW "designation".

What SCC is this? A 450/100 can handle 20A per tracker and that limit is a function of how far the units reverse polarity protection for the PV wires can extend. if you are confident you won't be connecting things in reverse you can actually go beyond 20A, footer note 2)

I think people are suggesting you connect the other array in parallel with the second, keeping voltage the same and doubling current.

Overpanelling (connecting more nameplate Watts than the SCC can actually output) is common on Victron and other controllers, as it's rare to get full nameplate value from a panel, this allows for better utilization in winter and clouds etc

 

[Oak77]

Splitting at the array means you have two arrays and the wires between the arrays are not combined into a single larger wire each wire from the distinct array runs back to where the MPPTs live. Each MPPT needs to connect to 1 array or you can string the arrays together and form 1 array, but you can't take a combined wire and connect it to two different MPPT.

You would have two arrays if for instance you had some panels faced east and some west. Each direction would need its own MPPT. For that each array has a pair of wires connected to each MPPT and you have two MPPT.

Yes, that's what I din't understand (thought he means using 1 MPPT with reduced voltage). What you're saying makes a perfect sense, but
I wrote before I can't put new cables in (feasibly) and understood the hint commenting that.
My trouble is, that the protection tube goes into the ground and around a house, having like 6 tight turns. Maybe I could pull the two cables out pulling a steel cable through, but I'm nearly sure I won't be able to pull 4 wires back. It would be a major construction hassle to dig up the tube, run the new cables and put it back.

 

[Brucey]

My apologies I didn't realize the easy solar has lower limits than an rs450/100, and there's no footnote saying the limit is due to reverse polarity protection. So paralleling two strings on the existing cable may be out.

 

[robbob2112]

Yes, that's what I din't understand (thought he means using 1 MPPT with reduced voltage). What you're saying makes a perfect sense, but
I wrote before I can't put new cables in (feasibly) and understood the hint commenting that.
My trouble is, that the protection tube goes into the ground and around a house, having like 6 tight turns. Maybe I could pull the two cables out pulling a steel cable through, but I'm nearly sure I won't be able to pull 4 wires back. It would be a major construction hassle to dig up the tube, run the new cables and put it back.

Well, as long as the tight corners are sweeps and if you know which end to pull from so they don't get stuck, you can tie some pull string onto the old wires and pull it through. Pull string is just really strong cord that won't break when you really tug on it. And you lay out your new wires nice and straight and have the person feeding them grease them with the pull grease so they are like a pig at the county fair.

Whatever you do, don't JERK them. If they are stuck back them out 10ft or so and feed them in at a slightly faster rate to see if you can get past whatever corner it is... If you get to a point you can't get past you can either pull it all out and start at the other end or pull back 10ft or so and twist the wires a bit as they go into the conduit.

So you can pull the new set of wires through. Note - the old wires can be included in the new wires, just tape the first few feet together really tight and I always taped about every 10ft. There are youtube videos to explain how to pull new wires.

And all of this assume you can fit the new wires and old wires inside the size conduit you have. One time I had to dig up a 250ft run because I used the size called for and later figured out I needed more. So I always go up 1 size or double from then until now.

 

[Brucey]

Did you mean Dyness? I've never heard about them and I'm not sure if their quality and support is much different from SunStone, They're all exploding in confidence, having tons of certificates, but the reality can be quite harsh.

Ningbo Deye Inverter Technology Co., Ltd. (Deye).

 

[Oak77]

What SCC is this? A 450/100 can handle 20A per tracker and that limit is a function of how far the units reverse polarity protection for the PV wires can extend. if you are confident you won't be connecting things in reverse you can actually go beyond 20A, footer note 2)

I think people are suggesting you connect the other array in parallel with the second, keeping voltage the same and doubling current.

Overpanelling (connecting more nameplate Watts than the SCC can actually output) is common on Victron and other controllers, as it's rare to get full nameplate value from a panel, this allows for better utilization in winter and clouds etc

View attachment 214286

Thank you, those are some valuable insights .

I was referring to Multi RS Solar 48/6000, the manual states:

The maximum operational input current for each tracker is 13 A. MPPT PV inputs are protected against reverse polarity, to a maximum short circuit current of 16 A for each tracker.

So maybe 13 A is protected against reverse polarity and 16 is the rated value.

Footnote (1) is what I calculated, with nominal voltage I'm little bit bellow and with VOC I'm right at the edge (9*48V=432V nominal voltage, with Voc it's 49.5V -> 445V --> over the limit). I have to check though, if I remember correctly Voc of my batteries (54V * 8 = 432V)

The footnote (3) is interesting, I was wondering that the watts don't add up to current and voltage... Even 450V x 13A = 5850 W, a much larger solar field that I have.

I'd rather disconnect 1 panel and be safe, rather than being slightly over the specs. But I wonder if my assumptions are correct and even 8 panels are safe on one MPPT tracker. That would be great, no construction work to pull 2nd set of cables and I can connect ~8 panels to the 2nd tracker in about 18 months, when my house extension is finished.

I wonder, if other combination of Victron products would be better. From my point of view Multi RS looks good, even if i wanted add an external MPPT it's possible (AFAIK), i.e. having current south-facing string, 1 west and 1 facing array (but I plan only 2 arrays).

 

[Brucey]

Thank you, those are some valuable insights .

I was referring to Multi RS Solar 48/6000, the manual states:

So maybe 13 A is protected against reverse polarity and 16 is the rated value.

Footnote (1) is what I calculated, with nominal voltage I'm little bit bellow and with VOC I'm right at the edge (9*48V=432V nominal voltage, with Voc it's 49.5V -> 445V --> over the limit). I have to check though, if I remember correctly Voc of my batteries (54V * 8 = 432V)

The footnote (3) is interesting, I was wondering that the watts don't add up to current and voltage... Even 450V x 13A = 5850 W, a much larger solar field that I have.

I'd rather disconnect 1 panel and be safe, rather than being slightly over the specs. But I wonder if my assumptions are correct and even 8 panels are safe on one MPPT tracker. That would be great, no construction work to pull 2nd set of cables and I can connect ~8 panels to the 2nd tracker in about 18 months, when my house extension is finished.

I wonder, if other combination of Victron products would be better. From my point of view Multi RS looks good, even if i wanted add an external MPPT it's possible (AFAIK), i.e. having current south-facing string, 1 west and 1 facing array (but I plan only 2 arrays).

Yes they updated mid 2023 to have two seperate trackers in the multi rs so a second array in a different direction would be fine (just make sure when you order that it was built after that change).

 

[Brucey]

Footnote (1) is what I calculated, with nominal voltage I'm little bit bellow and with VOC I'm right at the edge (9*48V=432V nominal voltage, with Voc it's 49.5V -> 445V --> over the limit). I have to check though, if I remember correctly Voc of my batteries (54V * 8 = 432V)

You are way too close to the 450V never exceed limit with VOC at 445V. 450V is a hard limit. You need to take into account that as temperature drops, voltage will rise. Beyond that VOC limit which is specced at 25C. There should be a temp coefficient listed with the panel specs. Most leave at least a 10% margin or base it on coldest recorded temperature for your location. I'm guessing 8 panels will be safe but depends on your location. The 8*bulk charge voltage limit is a soft one.

 

[Oak77]

Ningbo Deye Inverter Technology Co., Ltd. (Deye).

OK, my bad, found it on 2nd search attemt. Still unknown brand to me. They're available here, but not so much 6KW variant (SUN-6K-SG04LP1-EU), that one only pops up in online markets. The price little over half of the Victron Multi RS + Cerbo GX (it's like 2.5 times the 5KW version), but I'd still choose Victron if I could. But thank you, good to know the choices.

 

[timselectric]

Ah, my mistake, apologies, I misunderstood the terms.
Anyhow, so far a quick search revealed that almost all Growatt available in EU AIOs are 3 or 5KW. I can see they offer 6KW via Aliexpress, which I want to avoid. 5KW would be of course sufficient, but I don't trust them to really stick to it. Even Victron has reduced output to 4800W in 32˚C (if I remember it correctly, with full power under 25˚C.

5k is accurate. More for temporary surges.