diy solar

diy solar

Midnite Solar Announced their new 10kw AIO at Intersolar Today

StorEdge, Sunny Boy Storage, Sunny Boy Smart Energy are HV battery inverters for consumers.
SolArk has a 3-phase HV inverter (think that one is 277/480Y).

What we don't have is protocols or BMS for HV DIY.
I'm curious about what could be done with an EV. Let it be your AC coupled battery for peak shaving and backup usage (only when left at home, of course.)

Given how much salt is in the ocean, and low toxicity/good safety (once reacted) the opportunity to use sodium in a battery is very attractive.

AGM fits for my backup purposes, but at $0.50/kWh isn't competitive with grid. Possibly FLA can be, but I think it is still well above cost of LiFePO4. What is sodium ion cost per kWh of cycle life?
Are those not all AC coupled...where you have to have a grid connection? I could care less about selling back and I want my system to work if the grid is down. With the way the world is right now I don't want to rely on a grid for things to function.
 
I think they are all AC coupled to the grid, but they also provide backup when grid is down.
Except for SolArk, all require separate auto-transformer/transfer switch.
(SMA has a 3-phase hybrid for Europe with with everything needed built in.)

One configuration without full functionality is SBS + LG RESU "HV" battery. Because the battery is actually LV + boost converter, and PV is AC coupled, it can't charge from PV off-grid. When bidirectional boost converter changes direction, GT PV gets dropped. It does work with BYD, a true HV battery.

One of the guys here has SBS + BYD.
I'm playing with SBS + LG RESU

What works better is hybrid, like StorEdge or SBSE, because both PV and battery are coupled to HV rail of hybrid. The PV is DC coupled, not AC.

But all have wimpy surge. What you want for offgrid is an inverter sized for 2x+ surge current.

Don't know if any HV battery inverters use battery voltage for HV rail, or if they all use bidirectional DC/DC to control charge/discharge.
 
Does anyone know the efficiency on a HV inverter such as Sol-Ark's 30k and 60k as compared to a 48v system? Seems to me that if there is not much difference in efficiency, then the "sweet spot" on battery voltage for a residential size inverter would be in the 75-125V range. High enough to run small (ish) cables (~#4-#6?) yet low enough to minimize shock hazard. Seems to me that a 200v+ battery should have 2 pole breakers, GFI protection, etc. to keep from someone getting electrocuted.

Where a higher voltage battery would be nice is for commercial size systems. (Such as Sol-Ark 30k/60k.)

On residential installs (at least at this point) most people don't need more than 15-20kW of inverter, and a 20kW 120/240v inverter would be pretyy big! More likely to stack 2x 10-15kW units. 15kW at 125Vdc is 120A. #2 Copper can handle that current, and with a cable length of 6 feet my calculator says the voltage drop would be 0.218%. Even 1/0 cable is not hard to work with!

@Hedges I could see though that using the batteries directly as the HV DC bus could increase efficiency and possibly make it worth going that direction.

And also, I agree that it seems most HV battery inverter options so far have terribly wimpy surge! (And even a lot of the AIO 48v ones as well.) Everyone seems to think that on a grid-tied hybrid inverter you can get away with not much surge, but the instant the grid goes down, you need that surge capability! And a widespread extended grid down scenario would probably pretty quickly get a lot of peoples' attention when their existing inverters would fall on their faces!
 
Price similar to BYD, $18k for 32kWh.

About 2x what cheap server racks go for.
Is there a (cell?) quality difference?
 
The biggest issue/risk that I see with high voltage batteries (other than........ well...... high voltage...... 😁 ) is cell imbalance issues. There don't seem to be many battery manufacturers that have active cell balancing.
Two comments there.

One, every high voltage battery I have ever seen has internal contactors that disconnect the output when there's no control signal (CAN or RS485.) Thus they are in many ways safer than, for example, a 48V lead acid bank which is always live. Dropping a wrench while you are hooking up a 400V battery is a non issue; dropping a wrench while connecting a 48V lead acid bank can be very, very exciting.

They are even a bit safer than the 48V server batteries that I often use. On those batteries (EG4) they are on as soon as you turn on the breaker. They are certainly better than always-on lead acid or LFP batteries, but if you have a problem and the communication bus gets disconnected, the batteries remain live.

Two, all EV batteries (which are HV batteries) have BMSes. Some have resistive top balancers (like the Nissan Leaf) some have quite competent charge transfer balancers (like the Tesla Model 3.)
 
Does anyone know the efficiency on a HV inverter such as Sol-Ark's 30k and 60k as compared to a 48v system? Seems to me that if there is not much difference in efficiency, then the "sweet spot" on battery voltage for a residential size inverter would be in the 75-125V range. High enough to run small (ish) cables (~#4-#6?) yet low enough to minimize shock hazard. Seems to me that a 200v+ battery should have 2 pole breakers, GFI protection, etc. to keep from someone getting electrocuted.

Where a higher voltage battery would be nice is for commercial size systems. (Such as Sol-Ark 30k/60k.)

On residential installs (at least at this point) most people don't need more than 15-20kW of inverter, and a 20kW 120/240v inverter would be pretyy big! More likely to stack 2x 10-15kW units. 15kW at 125Vdc is 120A. #2 Copper can handle that current, and with a cable length of 6 feet my calculator says the voltage drop would be 0.218%. Even 1/0 cable is not hard to work with!

@Hedges I could see though that using the batteries directly as the HV DC bus could increase efficiency and possibly make it worth going that direction.

And also, I agree that it seems most HV battery inverter options so far have terribly wimpy surge! (And even a lot of the AIO 48v ones as well.) Everyone seems to think that on a grid-tied hybrid inverter you can get away with not much surge, but the instant the grid goes down, you need that surge capability! And a widespread extended grid down scenario would probably pretty quickly get a lot of peoples' attention when their existing inverters would fall on their faces!
Some advantages to HV that I can think of.

1. Equipment runs cooler making it more efficient and overall performance should improve along with lifespan. Less fan noise since it's running cooler which should mean less idle consumption. Seems like surge performance would also improve since you don't have to ramp up voltage rather just step it down.

2. Less heat means less cooling the room it's in.
3. Less energy wasted inverting so in theory the battery storage should last longer.
4. Cheaper install since you can use much smaller wire.
5. longer distance between the batteries and inverter. Could be nice for us that want to store the batteries outside in a shed or something.

I think there's lots of people in residential settings that pull more than 20kW at once. Personally I'm like the 40kW sandi one, It's low frequency and has awesome surge capability but it's also like 400lbs. I remember last time we talked about HV breakers came up, it looks like breakers are coming on the market now I guess since manufactures like solark are moving this way. Again, it makes no since to worry about HV on the battery side but then turn around and run 600V PV...IMHO.
 
Two comments there.

One, every high voltage battery I have ever seen has internal contactors that disconnect the output when there's no control signal (CAN or RS485.) Thus they are in many ways safer than, for example, a 48V lead acid bank which is always live. Dropping a wrench while you are hooking up a 400V battery is a non issue; dropping a wrench while connecting a 48V lead acid bank can be very, very exciting.
I was mostly comparing LFP HV vs 48v. And from a manufacturer standpoint the 400V battery reference that you make is a whole lot more intricate and also there is a whole lot more liability interms of safety! A faulty breaker, contactor or BMS could mean death for someone, and a potential lawsuit for a company!
They are even a bit safer than the 48V server batteries that I often use.
Not a chance! 48v (or actually 51.2v on LFP server rack batteries) vs 400V, the 48V always wins in terms of safety! 48V you can touch positive and negative with your 2 hands and your good. 400V not so much.... you're toast at 400V from a battery! (Yes, I know those safety devices should fix this. But any component sometimes fails! )
On those batteries (EG4) they are on as soon as you turn on the breaker. They are certainly better than always-on lead acid or LFP batteries, but if you have a problem and the communication bus gets disconnected, the batteries remain live.
Proper BMS and/or breaker should take care of disconnecting if there is an issue such as overcurrent, overvoltage, overtemp, etc..... Losing communication bus is not a safety hazard in itself!
Two, all EV batteries (which are HV batteries) have BMSes. Some have resistive top balancers (like the Nissan Leaf) some have quite competent charge transfer balancers (like the Tesla Model 3.)
👍 I'm not familiar with EV batteries, but I'm interested more in learning more about Tesla's balancers!

Also, as someone else noted (maybe @Hedges ??) EV battery use is quite different than solar/storage! You will put many more cycles on a solar storage battery application than you will on an EV! (For instance, if an EV battery were to give 200 miles per charge, one the car had 200,000 miles on it, the battery has been cycled 1,000 times. On the flip side, for solar storage, with LFP you are looking at 5,000 cycles or higher! And also, the charge profile is drastically different.

Personally, I don't think you can compare an EV battery/application with a PV battery/application.
 
Some advantages to HV that I can think of.

1. Equipment runs cooler making it more efficient and overall performance should improve along with lifespan. Less fan noise since it's running cooler which should mean less idle consumption. Seems like surge performance would also improve since you don't have to ramp up voltage rather just step it down.

2. Less heat means less cooling the room it's in.
3. Less energy wasted inverting so in theory the battery storage should last longer.
I haven't seen specs or real world application data showing any huge improvement in efficiency. But I do realize that this could be the case.

As far as surge goes, that pretty much rides on the way the FET board is built and programmed. Look at the Midnite Rosie for instance. Massive surge capability at 48V nominal!
4. Cheaper install since you can use much smaller wire.
Cheaper is not allways better. (I like going the cheap route too though.... I have #2 cable going to my Victron 48/3000 inverters...)
5. longer distance between the batteries and inverter. Could be nice for us that want to store the batteries outside in a shed or something.
I wouldn't recommend that even with high voltage batteries for a few reasons. High voltage batteries still have high amperage capability, just like 48V (or even 12V for that matter). Shoot, I have a 7AH LFP starter battery on my 4-wheeler with a 150A?? BMS! So now thinking of going distance with say 400V and running.... #8...?? Or even #6 wire.... from a battery that has the capability to push hundreds of amps if the BMS and/or breaker were to fail... not a good idea! And then there is the whole aspect of controlling charge voltage correctly. (I know, closed loop commsis a must for high voltage, but still...)
I think there's lots of people in residential settings that pull more than 20kW at once.
No, I think your wrong there. Most PoCos will way undersize their transformers for residential services, because they know they can. If someone had an all-electric house then they will sometimes hit 20kW, but not often. Now an electric car charger is a whole separate ballgame. I realize that adds a lot, but most people who want storage for backup don't want enough to charge their car as well, so why would they need an inverter that's big enough for it?
Personally I'm like the 40kW sandi one, It's low frequency and has awesome surge capability but it's also like 400lbs.
400lbs quickly makes it not be installer friendly, which is why I say most people would opt to stack smaller units.
I remember last time we talked about HV breakers came up, it looks like breakers are coming on the market now I guess since manufactures like solark are moving this way. Again, it makes no since to worry about HV on the battery side but then turn around and run 600V PV...IMHO.
600V PV at 10-20 amps is a whole different animal than 400V battery with a short circuit capability of 300-500 amps! If you short out PV, you will get a terrific arc, but your wires won't melt! If you short out a 400V battery and the breaker and/or BMS fail, your wires will melt and your house will burn to the ground. There would be no turning back once that happens!

Hmm.... sounds like a fun experiment! 😁
 
Here are screenshots of Sol-Ark's spec sheets. Looks like exactly same rated efficiency for 15K at 48V, 30K at 300V and 60K at 600V.
I don't think those are battery to AC efficiency numbers, those are PV input to AC. It is not surprising to me that the PV to AC efficiency is identical in those units.

For example, EG4 18KPV:

PV to AC: 97.5%

Battery to AC: 94.0%

At 12 KW AC output, 94% means 766 watts of heat being wasted.

Efficiency of energy conversion is better when input voltage and output voltage are close to each other. This is why PV input of 360 volts is best for making 240 VAC (340 volts peak). If the battery was HV, then we would see battery to AC numbers similar to PV to AC numbers.

A HV battery system where the battery is closer to PV input and AC output will be more efficient and lower cost. That's just the basic nature of electrical power. EVs do this even though there wire runs are very short.

The problem with HV batteries is that anything above about 60 volts is a safety issue if you touch it. But given PV is already well above that, it doesn't seem like a huge step to treat the battery wiring with the same care. You really should be doing that at 48 volts, too.

The other negative for HV batteries is lack of available options for wiring them up. Putting 48 volt rack batteries in series is fraught with issues and you can't really buy economical HV batteries in 5 KWH chunks like 48 volt rack mounts. It is a chicken and egg thing, HV batteries are expensive because few people buy them because they are expensive.

Another aspect is that HV battery inverters will be lighter, the circuits are just not as heavy to deal with voltage over current.

Mike C.
 
Consider the alternatives.
(to premature announcement but not shipping yet)
 
Were you thinking dual The Ones?
Depending on the price...maybe. I'm not sure I would need two. Dual 10kva Quattro's is a bit overkill most of the time and dual 5kva MP2 would be fine most of the time but there are things I would like to be able to run that they couldn't support.
 
I'm only aware of Sandi and Aims for us DIY folks. The others like what outback and schneider have on the market are only AC coupled so they do not work without the grid. They're are a few guys running the sandi inverters, they are low frequency and built like a tank and are industrial quality from what I've read with mitsubishi guts. Sandi has HV battery now where it's plug and play. I was hoping to see American made inverters chalanging these soon, I'd rather support a American company.

https://www.cnsandi.cn/
Last time I got a quote, the price for a decent sized battery was pretty pricy. The inverter certainly looks like a tank!
 
Back
Top