Inverter recommendations for 48V off-grid installation

[Shimmy]

Seriously though, potted board which I think Victron uses should be a plus. Maybe they also have larger spacing between high voltage terminals.

So, I was all set on an XW-Pro, and then I read this. We have geckos... lots of geckos. They get into everything, shit, lay eggs, etc. They are known to short out mini-split control boards that aren't potted, although I have never heard of them causing problems with solar equipment. Any wisdom?

The schneider response about altitude de-rating the dielectric seems odd; usually it is about cooling in lower density air. I would think the conductivity of 95% humidity air would be much more of a problem.

 

[Hedges]

You could add on finer screen, would have to clean it more often.

Most equipment has screens to keep out larger things. Some is sealed, with fins to transfer heat. Motors called "TEFC" (totally enclosed fan cooled) are this way. Sunny Island and most Sunny Boy have SMA's "OptiCool" design, which is a sealed volume for electronics and a fan cooled 3R compartment for transformer with fins to cool the electronics. They may have an internal fan as well to cool some components.
The Sunny Island -US models aren't quite sealed because they have a mechanical breaker and SD card. I sealed that with sheet of silicone. New European models 8.0H are fully sealed because they don't have those parts, only membrane keypad and display.

3/4 of the way down is a drawing:

Inverters: power electronics for a clean power supply

Outback has "GTVX" (vented) and "GTFX" (sealed) models, with different cooling capability.

"Condensing" humidity would be a shorting problem. Otherwise it is Paschen effect, each gas supports breakdown at some voltage depending on pressure x distance, basically the number of molecules between electrodes.

So long as the water remains as H2O vapor in the air, it has only slightly lower breakdown voltage than air (which is predominately N2 and has same breakdown as N2). Gasses like He, Ne, Ar have significantly lower breakdown voltage. (and blend of Ne + Ar shown considerably lower.)

Figure 1. The Paschen curve shows the dependence of the breakdown...

Download scientific diagram | The Paschen curve shows the dependence of the breakdown voltage, Vt [V] on the product of gas pressure, p, and the minimum breakdown distance, d, for four molecular gases composed of H2, H2O, CO2, and of air. The classical Paschen curve depicted here, has a minimum...

www.researchgate.net

Figure 2. Paschen Curve for various gases [4].

Download scientific diagram | Paschen Curve for various gases [4]. from publication: The ratio of enthalpy minus piezoelectric energy versus the sum of the 1st ionization atomic energy in N moles of an ideal gas in an isochoric process. | Andrew Somorjai May 12 th 2015. The ratio of enthalpy...

www.researchgate.net

Some of those curves are significantly different from what I remember seeing, which was voltages < 200V and < 100V (at some pressure and distance), like this:

Happy Birthday, Friedrich Paschen

German physicist Friedrich Paschen improved our understanding of how electric current flows through gas. Learn about his life and work here.

www.comsol.com

Lowest for air was at something like 50,000 or 100,000 feet. A problem for rockets. Your location is just part way down the curve on the right hand side.

 

[copec]

I never even realized the higher altitude was a problem for breakdown voltages - I thought the thin air was purely a cooling issue. That's really interesting!

I would probably look at all the components and see if I could submerge some budget inverter in mineral oil to save $0.02

There were several tons of SMA 6048's on ebay a few years ago (presumably after that solar trailer company for powering venues went under), I wish I would have bought some.

 

[Hedges]

They're still on eBay and Craigslist, though not as heavily discounted. Some new close to $3000, some used.
At those prices, use with a transformer may make sense. When they were half that, buy two (or four) even if you don't really need the watts. (I did.)

SMA Sunny Island Off-Grid Inverter (SI6048-US-10) for sale online | eBay

Find many great new & used options and get the best deals for SMA Sunny Island Off-Grid Inverter (SI6048-US-10) at the best online prices at eBay! Free shipping for many products!

www.ebay.com

DC Solar Trailer Generator 50kw Batteries, 2 SMA Inverters 6000w ea | eBay

GVWR tandem axle trailer with electric brakes, pintle hitch, trailer jack, If you want to change it to a ball bring a 2 5/16 coupler and two 15/16 wrenches.Trailer is 22'6" long x 8' 2 1/2" wide and 9'2" tall.

www.ebay.com

 

[copec]

They're still on eBay and Craigslist, though not as heavily discounted. Some new close to $3000, some used.
At those prices, use with a transformer may make sense. When they were half that, buy two (or four) even if you don't really need the watts. (I did.)

SMA Sunny Island Off-Grid Inverter (SI6048-US-10) for sale online | eBay

Find many great new & used options and get the best deals for SMA Sunny Island Off-Grid Inverter (SI6048-US-10) at the best online prices at eBay! Free shipping for many products!

www.ebay.com

DC Solar Trailer Generator 50kw Batteries, 2 SMA Inverters 6000w ea | eBay

GVWR tandem axle trailer with electric brakes, pintle hitch, trailer jack, If you want to change it to a ball bring a 2 5/16 coupler and two 15/16 wrenches.Trailer is 22'6" long x 8' 2 1/2" wide and 9'2" tall.

www.ebay.com

Yeah, my next project is doing a solar shed on skids that could hold some serious battery that could also AC couple to my Deye in my house once its non-firecracker chemistry battery is full. Starting with one of those trailers would work really well for the most part. I enjoy recovering cells but haven’t done anything “powerwall” with them because I’ve wanted to keep that away from my house.

Stumbled on this the other day:

 

[keepsake]

I got the newer Schneider Insight Home to perform the same with modbus and webpage as the 2-year old gateway version.
Now, grid codes is another crap shoot.
I can set them with the codes on both boxes but only system will only sell to grid with older gateway.
When I move my Xanbus to the Insight Home, my XW 6848 Pro drops selling to utility.
So if my system sees a gateway or no interface it will sell fine. When it sees Insight Home its like a broken dongle, selling stops.
Any clues would be great.

 

[ChrisG]

Been a while and a lot of water has passed under the bridge. After consulting with AltE and Schneider about the altitude issue, it was strongly recommended by both not to exceed the altitude limitation. The concern is arcing across circuit boards due to the air at altitude being less of an insulator than at lower elevations. So... all the Schneider stuff went back with accompanying expensive shipping.
After more research and shopping, I settled on Victron equipment, which is rated up to 5000 m. I am planning to haul it to the cabin and start installing as soon as the snow melts. The list so far:

• 2X Quattro 3000 KVA inverters (which coordinate between themselves to provide split phase 120 V and 240 V AC and are stackable)
• SmartSolar MPPT 250/85 SCC
• BMV-712 Battery Monitor
• Cerbo GX Comm Device
• GX 50 Touch Screen

I've settled on Longi 355 W panels that I can get locally at a good price in a 5S3P arrangement, providing nominal 5325 W. Also I will probably go with an Iron Ridge ground mount with locally sourced pipe. For batteries, I am considering Crown lead acid batteries for about 10-15 KWH. I bought an inexpensive combiner box off Amazon. Basically all I have left is wire and switches. Now if the snow will melt...

@jsocolof what was the outcome of the build?

 

[jsocolof]

@jsocolof what was the outcome of the build?

As of today, everything is installed and operating except the panels. I am awaiting some muscle to help get the panels on the roof. But the inside components are all working well - the two Quattros, the battery bank, and the AC panel with four circuits so far. The genset connection is working as well, which is how I am charging the batteries. The only issue I am having is that the battery monitor (Victron BMV-712) seems to have the state of charge (SOC) decaying very rapidly. I believe I'll start another thread with this issue.

 

[iSwannie]

As of today, everything is installed and operating except the panels. I am awaiting some muscle to help get the panels on the roof. But the inside components are all working well - the two Quattros, the battery bank, and the AC panel with four circuits so far. The genset connection is working as well, which is how I am charging the batteries. The only issue I am having is that the battery monitor (Victron BMV-712) seems to have the state of charge (SOC) decaying very rapidly. I believe I'll start another thread with this issue.

Hi jsocolof,

How is the Victron system working out for you now that you have it up and running? Any issues wiring the devices up or configuring the software? Is it performing as designed?

Iswannie

 

[cara2000]

Looking for recommendations for off-grid inverters for a system with these characteristics:

• 48V
• un-inspected: off-grid, no building or electrical inspections required
• ~6 KW solar panels expanding to 12-18 KW over time
• PV array would probably settle in at around 400 V, 19 A initially, expanding to multiple arrays
• battery storage (looking at the EG4 batteries at the moment)
• 9100 ft elevation
• many days below freezing, with many also below zero
• ~5-6 KW of inverter capacity initially, expanding if required
• 240 V required
• The cabin will be uninhabited 3-5 days/week
• generator backup, and I have a Honda EU700is generator, currently without autostart

If you have an inverter brand/model recommendation based on your experience, please indicate why you recommend, and pros/cons. I've seen a lot of discussion of issues with certain brands on this forum, but not anything collecting it into one place for my scenario.

Research so far:

• Sol-Ark 12K: Perhaps not the best for off-grid use. Seems to include features that benefit grid-tie systems that I would not want to pay for. Seems touchy, as a lot of folks have struggled with it shutting down with slight imbalances on the AC legs. Needs an additional autotransformer to overcome this. Pricey.
• Victron Quattro 5000 KVA: Seems straightforward. You buy two to get 240 V. Not UL listed, which does not matter for me. Less pricey than Sol-Ark for two. Reputable brand.
• Growatt 5000W 48V: Price seems too good to be true. Needs an autotransformer to produce 240 V. Relatively high idle watt usage. Reputation of the brand? Reliability? Longevitiy?

Would appreciate any advice you guys can provide as I embark on my design.

Have you decided on any inverter? I am looking for a good one too and I think warranty should be an important factor. I think MPP is only 1 or 2 year warranty - so not too good. Not sure about other inverters proposed here.

 

[keepsake]

Beware of Fronius. I had 5 inverter changes in their 10 year warranty. Now out of warranty. Displays have almost lost segments year after year

 

[Hedges]

Any functional failures, or just the display?
What kind of display?
The LCD on my Sunny Boys fade in the sun. I now have a flap of aluminized duct tape over them as a sun shield.

When I was looking up inverter failures, I found a large number of Solar Edge "not working" on eBay, but vanishingly small number of all other brands. I thought Fronius was a reliable, respected brand, so this being the first complaint I've read I am curious.

 

[keepsake]

On Fronius -- on the sixth entire box now. Had 2 IG 5100 installed in 2011. There error codes are nightmares. Out of warranty one display cycles codes 24/7 or really 10 hours a day til sun goes down. It does produce about 80% of the other one. The full output one has multiple segments out.
Fronius never had board replacement available. Only positive thing is once I found a local outfit to service, no charges on to or from shipping. Beware most solar outfits WILL CHARGE their own fee from you. Never found another outfit that didn't take manufacturer $$$ and demand you the owner pay more.

On the other hand, have SMA 5kw Sunny Boys from 2016, still 100%, never an issue.

Schneider/Electrics all fine even after 10 years.

 

[keepsake]

So total Fronius production is at about 85% of original output. LCD displays

 

[Hedges]

Do you think Fronius is now gathering 85% of available PV power? Or did panels degrade?
Can you swap strings between SMA/Fronius inverters, see if the 15% power loss stays with the inverter or moves with the panels?

Each inverter - is it transformerless or transformer type?
Same panels or different? Are they "PID Free"?

 

[keepsake]

PID Free ? Never heard that.
I attribute 5% to panel loss in 11 years.
I have done the string exchange. Each inverter has 2 strings. Only one mppt circuit for the entire 5 kw. Have 12 panels in each string.
SMA is 150 feet away, so no swap. But I use current transformers to determine production. Been watching in the same fashion for 10 years now.
That IG5100 has no data path.

 

[keepsake]

Just don't get micro inverters

 

[Hedges]

Look up PID (Potential Induced Degradation) if you're not familiar with it.
Some panels degraded badly. Use in positive ground vs. negative ground array caused/prevented the problem for susceptible panels. Transformerless, can't avoid half the panels being biased the wrong way.
Some panels are said to be "PID free", actually meaning PID low enough to be negligible.

Swapping strings between Fronius and SMA, did you find Fronius produced lower wattage for any string (or array) than SMA did?

There are differences in MPPT algorithms. If 12s2p and some panels are bad, the Power/Voltage curve could have two peaks. One algorithm might stop at a higher voltage but lower power peak, another can find the lower voltage higher power peak. Could be Fronius and SMA algorithms differ, so one does a better job of getting past a bad panel in a string.

To get around that, try disconnecting one string at a time so you're swapping single strings.
But individual panels can drag down a string in the same way. Check for differences among the various strings.
Use a clamp DC ammeter to compare strings. I did that, found one disconnected (loose wire nut.) Found another underperforming, tracked it down to several bad panels.

Measuring PV panel performance

I suppose I could buy a solar panel tester like Will reviewed, but that doesn't feel DIY enough to me. Besides, I'm a cheapskate, and I try to do things in the most difficult (and expensive if the equipment wasn't on hand) way. I'm presently working on an underperforming string of Sharp 165W...

diysolarforum.com

 

[jsocolof]

Hi jsocolof,

How is the Victron system working out for you now that you have it up and running? Any issues wiring the devices up or configuring the software? Is it performing as designed?

Iswannie

My system is working very well overall. I had one issue about a month ago when the breaker from the solar charge controller to the batteries kept tripping, but I reviewed the sizing against my design and found I'd installed a lower-rated breaker than desired. So I replaced that and the problem is solved. The only other issue is a bit of corrosion on two battery terminals, but I've ordered some NO-OX-ID A-Special- Electrical Contact Grease and will go and clean my terminals and coat them when I'm next at the cabin.

I still have a couple of issues I have to sort out: My battery temperature sensor, which is attached to the Victron battery monitor, reads a constant 52 F, which is not right. Also, after watching the performance for a while, I am doubting the state of charge the battery monitor is reporting. The SOC only gets down to the low 70% range with the battery voltage in the low 46 V range, and then the inverters trip due to low voltage. The way I have it calibrated, the floor (0% SOC on the monitor) should be at the battery's 50% charge state. With my loads, it's discharging about 80-150 Ah at night, which should be less than 1/3 of the usable battery discharge capacity. It's not a huge issue, but I'll have to work through that when I have the time.

As for performance, the panel array is rated at 5.7 KW and I've been seeing about 4.2 KW max. The array is roof-mounted, so is not at the optimal tilt angle, but I knew that going in, and the performance is fine with me. It's also winter, which reduces the sun's angle. I wish I had more battery, but I have an 830 Ah, 48 V bank (430 Ah usable, since it is a FLA bank), and it's all I can afford at the moment.

Other observations: Overall I really like the Victron equipment and the Midnight Solar e-panel and breakers. Installation was pretty straightforward and the system looks pretty clean I think. I'll post some pictures sometime when I'm at the site. Also, I really like Victron's remote management portal. Once I worked out some network kinks, it's been rock solid for me to monitor the system from home, and the information available on the portal is more than adequate. I have the inverters set up for 6000 W 120 V split phase or 3000 W 240 V, which is working well enough for my loads.

 

[jsocolof]

Have you decided on any inverter? I am looking for a good one too and I think warranty should be an important factor. I think MPP is only 1 or 2 year warranty - so not too good. Not sure about other inverters proposed here.

Yes, I have two Victron Quattro 48/3000 running in split phase 120 V. Works great.