A few years ago I was invited to an “energy roundtable” hosted by the South Australian government. The idea was we’d all sit around and hatch some big ideas to push the energy transition forward. This was peak excitement in SA – Elon had recently cut the ribbon on the Big Battery in Jamestown.
All I remember from it was the state energy minister in charge at the time telling us of one idea suggested to his department that we might consider: changing the default electricity supply from AC to DC. The logic? We’d save the conversion losses when running modern electronics off DC.
This was probably the stupidest thing I’ve ever heard in all my 25 years as an electrical engineer. Yes, converting AC to DC does cost you some efficiency. In the real world about 5-15%. But ripping up an entire state’s grid to claw back that sliver, while simultaneously losing efficiency in all the appliances that actually run on AC, is brain-meltingly dumb. I walked out at the first smoko and never went back.
A flashback of this experience hit me between the eyes while browsing the comments on our latest YouTube video about home batteries.
Critical Comments
We got a few viewers pointing out that the video didn’t mention how AC-coupling “has a LOT more loss” than DC-coupling, because of the extra conversion step.
It’s worth tackling this furphy of inefficient AC-coupled batteries, because this is a favourite trick of some battery salespeople who only sell DC-coupled solutions. They’ll bash AC-coupled systems to steer you toward the model they are set up to sell. The pitch sounds technical, it’s factually correct, and seems important. But just like the Minister’s big DC dream, scratch just below the surface and it’s readily apparent it doesn’t matter in the real world.
A Quick Explainer: AC & DC Coupling
- Solar panels generate DC electricity.
- Batteries use DC electricity.
DC-Coupling
If you DC-couple, the solar panels are connected directly to the battery1.
It’s DC → DC. No conversion to AC required in order to charge the battery.
AC-Coupling
If you AC-couple, you connect the solar panels to the battery via your AC switchboard.
It’s DC → AC then AC → DC.
The solar inverter converts the solar first to AC , so it can be connected into your switchboard. Then the battery inverter takes that AC from the switchboard and converts it back to DC for the battery to charge. Two conversions.
So yes, AC-coupling is less efficient.
Why It Doesn’t Matter
1. The energy you’re “wasting” is often worthless. Daytime surplus solar in Australia is practically free. It’s worth maybe 2–10c per kWh (whatever your FiT is), or zero if you’re hitting an export-limit. If your AC-coupled battery needs a little more solar energy to charge than a DC-coupled battery would, you haven’t lost anything meaningful.
2. The losses are small. With two conversions at ~95% each, you lose ~10% charging a battery with AC-coupling. On a 20 kWh battery, that’s an extra 2 kWh of solar required.
The cost:
- $0 if you’re hitting export limits. 20c if you’re still hanging on to a 10c FiT. Chump change.
3. Other factors are far more important. AC-coupling makes retrofits easy, avoids warranty dramas, provides redundancy (2 inverters) and often delivers stronger backup power. These things genuinely affect your system’s usefulness.
Make A Smart Decision
If you’re building a new system, DC-coupling is usually neatest and cheapest. If you’re retrofitting, AC-coupling is often the smartest.
Don’t be fooled by the “AC is inefficient!” sales pitch. It’s just the home-battery version of the “convert the whole state to DC” brainfart – spectacularly irrelevant.
Next time someone bangs on about AC-coupling being so inefficient, just nod politely. Then remind yourself they’re asking you to choose their battery over a competitor’s to save an extra 20c a day.
Phase Shift is a weekly opinion column by SolarQuotes founder Finn Peacock. Subscribe to SolarQuotes’ free newsletter to get it emailed to your inbox each week along with our other home electrification coverage.
Footnotes
- Technically via a DC-> DC converter in the Hybrid Inverter ↩
About Finn Peacock
RSS - Posts
The other benefit of DC Coupling is the ability to charge the batteries when the grid is down, a pretty rare need unless your power is prone to regular extended blackouts.
Battery backup for grid outages has been quite useful for me a few times in the past year – I work from home. The only extended black out we have had was the Brisbane 2011 floods, over a week out then. But the batteries and inverter would have been under water and toast anyway.
You can charge the batteries when the grid is down with AC coupling.
So both DC and AC coupling set up will charge the battery when the grid is down? Is that standard or is it something that should be set up like blackout protection?
It’s standard with any good AC coupled system. Obviously, just like DC coupled systems you need a gateway for backup functions.
Be careful Finn, some inverters will shut down completely without a power source – e.g. grid to sync to.
Finn,
I’m currently getting a quote for a ~29kWH Bluetti EP2000 battery system noting I have an existing 13kWdc/10kWac system (Fronius Symo inverter, three phase). Currently my standalone solar system does not operate during a blackout.
The technician tells me that, in order for both our solar and new battery system to operate during a blackout, I need DC coupling rather than their standard AC coupled installation.
Does that sound right?
Hi Jeremy,
You already have what is one of the best inverters on the market. I would get a Fronius Gen24 Symo and both systems will appear on the same app.
Fronius Are currently working on firmware that will allow both units to run during an outage, not just the Gen24.
This makes it easy to get connection compliance or instigate controls, like solar surplus EV charging, hot water or export limits without compatibility problems.
AC coupling is fine but DC coupling is better and in many cases is the only way to get true black start capability.
Not that I’m aware of, AC Coupled batteries are charging off the grid at a rate to offset the solar discharge.
One thing to mention.
If DC coupled, through a hybrid inverter, you’ll have the ability to have much more panels on the roof. Sungrow for instance allows for 200% panels, as it’s able to generate 100% AC and charge 100% DC simultaneously. That’s a win that can’t be had with AC coupled batteries where inverter size is limited by the distributor. Good for winter…
Could you please provide some more info about how AC coupled batteries need to be configured to enable the charging of a battery when the grid is down?
I have an Enphase microinverter solar system and an installer is telling me it’s not possible for an Enphase AC coupled system to charge a battery when the grid is down.
Hi JD,
Your installer could be right if he’s selling a Fronius Gen24, they run at 53Hz when the grid is down to kick any other AC sources off the system. This might change with a firmware update sometime but I can’t promise that.
For any other battery hybrid I know, you can use Enphase as an AC coupled source and use frequency shift to control it.
The hybrid must generally be bigger than the Enphase (or use a throttle circuit when off grid) but the only other governing limit might be the DNSP rules.
10kW/phase is a rule of thumb but you may find differences across 17 Australian networks.
This was also what I was led to believe with the Enphase iq7 vs iq8 however when ac coupling your existing retrofit to in my case a Sigenergy with gateway it is done thru the smart port so the battery still enables blackstart capability to recharge from the solar in the event of a blackout. I’ve also added an extra solar array to my shed and dc coupled this to the Sigenstor hybrid ems simultaneously along with the ac coupled existing so I can now have the maximum benefit efficiency of both.
All Enphase is support with Sigenstor (I have old M215 and S270 all working). For blackout support of this solar it needs to be on the Smart Port, or can in fact be on the Backup port if you want the Smart Port for something else, or different location without needing new cables. Black Start is also support in this way. However AC Coupled black start requires reserving some battery capacity for that. However, if you also have DC Coupled solar, then there is no need for the black start for AC Coupled solar, so I would not enable that and keep the battery capacity for regular us.
Like Finn said, you can charge your battery when the grid is down using AC Coupled. As long as the battery has enough juice it will start the solar inverter. The AC Coupled advantage in your case as mentioned in the article is that AC Coupled has DOUBLE the inverter and can therefore provide DOUBLE the backup capacity. In case of blackout, AC Coupling means you can draw BOTH power from PV Inverter and Battery inverter separately, to power your house.
Another big DC coupling advantage if you have a Sigenergy type system is the direct DC charging of your EV. Plus in Tassie, winter production can be horribly low so losing 10% can actually be a big deal.
exactly. its a non event.
The biggest advantage of DC batteries in my opinion is combining them with oversizing your panels.
Especially helps when your hitting DNSP inverter limits but can fit more panels.
When I next upgrade my system if there’s no rule changes I’ll try applying for a negotiated connection with Ergon for 15kw’s of inverters but there’s no guarantees there’s as Ergon only has default allowances for 1-10kw and 30-100kw with the in between left for needing a negotiated connection agreement.
Otherwise I’ll just end up with a 10kw inverter but the same number of panels just solely feeding into dc-coupled batteries.
Btw an article suggestion with the current battery rebate is a comparison of the low to mid range price wise batteries taking into account the rebate along side of any pros/cons to get the clear idea of where you need to budget to get both reliable and affordable batteries.
Personally I need a min of roughly 30kwh’s of batteries to power my house.
AC / dehumidifier
you know most inverters can do export limiting.
so if you had a premises with a large electrical draw you could still have a battery and inverter sized for your load and still be grid tied as you have the export limit set to the amount the distributor will allow.
the excess is sent to ground if your battery is charged.
Export limit is not always the only limiting factor. In Vic, under Ausnet, single phase installations are limited to 10kw inverter, regardless of export. I would like to have a 10kw inverter, and a 5kw for winter usage, but, ausnet says no. I’m on a 5kw export limit. 10kw combined max for the property.
last i looked with Ergon the default offers only care about inverter capacity they don’t acknowledge export limiting as an option.
for any of that you need to apply for a personalised Negotiated Connection.
—
p.s to staff it appears that the character limit for comments includes your name as my previous comment kept erroring out until i removed more letters.
One thing I’d look at is the power rating of the inverter/charger in the AC battery. If it’s a bit piddling, then two or more AC batteries might be needed to meet loads, and take available array output when charging. But, yes, as energy loss goes, there’s mostly not more than one more PV panel in it, if you really want to make it up. Then that’s the cost *for life*.
My setup is deliberately 50% AC coupled. The PV inverters AC charge the BEV at 7.2 kW, slightly *more* efficiently, I think, than the DC coupled solar, which goes through an extra voltage conversion. When the system is pumping 14 kW of photons, 7 kW is AC to the BEV, the rest into the battery or across it, then back out through the battery inverters to power the domestic AC loads. (These two equate to an AC battery, but with DC access for MPPT charging – maybe 2% efficiency gain on charging, so it’s much more for off-grid Black Start.)
Roofspace permitting, an extra panel or two helps cool the roof, I figure.
Am surprised that the grid feed in restriction, 5kw, means the dc coupled battery can only charge at a maximum 5kw.
I have 13kw of modules, 10kw inverter and a battery with Max charge discharge of 10kw.
I regularly see battery charging at 10kw while also powering home loads of 2-3kw.
I understood that this could happen only because the battery is dc coupled, so that 10kw dc is not subjected to the capacity of the inverter.
My feed in restriction is 5kw set by Energex, yet my battery can charge at full noise, 10kw.
Why is it so?
Feed-in is your EXPORT to the grid, from the inverter.
Anything behind the meter is irrelevant in that case. You could have 30kW of panels, a 20kW inverter and a 50kWh battery with no restrictions for personal use…
But kind of not in QLD… Anything 10kW or more (why most 10 inverters are rated 9.999) now require a back stop device to be installed. Even if you’re not exporting anything and self consuming 20kW, they will still shut down your inverter during excessive solar events.
I have to rip out my old inverter (which luckily has paid for itself) and DC couple the panels to the batteries so don’t exceed the DSNP rules. I would have been more than happy to AC couple as it would mean I had a fallback if something went wrong with the battery system.
Anyone want to buy a secondhand 8.2kW inverter (still on the CEC approved list)?
Hey Luke,
What’s a back stop device? And why do you have to it?
Thanks.
Bob
Backstop is the distributor/market mechanism that protects the overall grid. In victoria, it is now legislated to be functional on all new installations.
It’s basically internet control of the inverter, they (the distributor) have control over your export ability and can send out a command to throttle it back to 1.5KW instead of whatever the export agreement allows for.
It’s never used, but, it’s there, for the extra sunny summer day, just in case.
The Queensland version is controversial due to legislating ancient 60s style tech that just takes a analog signal from the grid and hard shutdowns your system.
Read all about it here – to be avoided if at all possible as it totally shuts down your system and you dont know you are using grid power.
https://www.solarquotes.com.au/blog/qld-dumb-solar-shutdown/
Energex FAQ here:
https://www.energex.com.au/our-services/connections/residential-and-commercial-connections/solar-connections-and-other-technologies/emergency-backstop-mechanism
If an existing string system is 133% of inverter capacity it might be getting flat topped in the middle of the day. DC coupling it might allow you to generate a bit more and charge the battery faster. But that might be a marginal benefit too.
What follows is based on my own gear….which is a statistically relevant sample of 1……
I have 19kw of panels and a 12kw inverter. I also have a 12.5kw DC EV charger. I can often be pulling 15+kW from the panels and still be nowhere near the inverters limit because the DC charger operates directly with the panels due DC coupling between these components rather than forcing everything through an AC backbone.
Like most of this renewable stuff the engineering is complex and unfortunately exposed to the end user. I have an electronics background and see that it’s generally complicated at a user level. Get it wrong and it’ll only be apparent for 20+ years. Perhaps there is a market for a user’s advocate where people who don’t understand it and who can’t ever be expected to understand it can go.
Andy
“Perhaps there is a market for a user’s advocate where people who don’t understand it and who can’t ever be expected to understand it can go.”
It’s not a bad idea. Another idea would be to have standardised configurations. I know, I know, it wouldn’t work for some (a minority of?) users who understand how it all works and you’ll get people calling into talk back radio complaining of government overreach, but for the majority of users standardised configurations could be a good idea.
If you want solar: you have these configuration options.
If you want solar + battery: you have these configuration options
If you want solar + battery + BEV: you have these standardised configuration options.
If you want to change your configuration from Option x to Option y: I don’t know, I haven’t thought about that and I d/k how everything works now.
Standards won’t work for some, but they could avoid badly planned out installs and would make it easier for an end user to plan and understand.
a big issue is the layman doesn’t understand things like the difference between kilowatt and kilowatt hours.
they don’t understand the basics of ohms law
or the difference between kva and kilowatt
aparent power and actual power.
they don’t understand that 5 kilowatt of solar array produces around 4.5 kilowatts of energy for every hour of peak sun so on average it produces around 24 kilowatt hours everyday with more in the summer months.
and of course people don’t see the other side of the coin about smart and efficient use of energy.
yes we can produce huge amounts of renewable energy but if we don’t use it wisely we won’t have enough.
Andy, what is the brand & model of DC EV charger – it sounds like a good’n. My 7.2 kW AC charger is OK, but a second one ought to be beefier, I think. (It’s in summer I have visitors, making two chargers worthwhile, and solar yield isn’t an issue then.)
Consumers half way up the learning curve can climb further with questions on fora, some youtube wallowing, and reading datasheets, but so many are unavoidably starting at the bottom, and don’t know what they don’t know, so have a devil of a job building off a zero base. It is hard to paint on knowledge – research does a better job, slow and painful though it is. Solarquotes blogs can provide some short answers, and Whirlpool is worth a try, I think.
SIgenergy gear, 5 modules, 12kw inverter/controller, 12.5kw DC EV Charger and 3 x 8kwh battery modules.
As I understand it Sigenergy have just dropped prices on the DC modules quite significantly. The battery modules are now 9kwh I think instead of 8
These dont sit at the cheaper end of the pricing spectrum, but as I said originally I can regularly exceed inverter output limitations when using eth EV charger and then again for the house battery charging as well
Sorry, but discussion of Sigenergy and the benefits of DC coupling on efficiency is misleading!
Anyone who has looked at the Sigenergy in any detail would understand the missing component not discussed here. Yes, there are efficiency losses for AC to DC, and from DC to AC. HOWEVER this focus completely ignores the fact that there are also efficiency losses with DC to DC conversions as well. The architecture of the Sigenergy system has an extra DC to DC conversions being used both into and out of the battery what most other systems do not have. This makes the architecture extremely flexible in a way that most other batteries are not. But this extra DC -DC conversion all adds up to the Sigenstor really not being very efficient at all. Despite many people calling is a DC coupled system.
My Sigenstor uses about 360w to sit there and do absolutely nothing. And for my real world use, if I get to 80% efficiency, I will be doing pretty well.
energy is neither created or destroyed it just changes form.
every time a conversion happens there is a bit of loss.
this is entropy and the laws of thermodynamics.
so the more often you convert it the more losses you accumulate.
all these control modules use electrons and they turn those electrons into heat.
this has to be disapated via heat sinks and fans.
so really they may not be as efficient as people think.
when you try to make a Swiss army knife its always a compromise.
your always better off with individual tools specifically tailored for the job.
DC coupled very effective when you have a decent PV oversize and a home with reasonable loads I.E HVAC and EVSE etc . Assuming good sunlight, In morning the PVES with BESS can be running loads up to the inverter’s capacity and charging its mostly discharged battery with the oversize component of the PVES. I also think its cheaper and easier to set up a PVES DC coupled BESS as an islanded microgrid for times of grid outages, precautionary islanding during electrical storms or just for whatever reason you want your home to run independently from the grid.
had me confused until i read your footnote, my battery sits around 400 volts, I know the strings can be running a lot different voltage than that!!
Bit in the losses argument – wouldn’t that DC to DC conversion also have some degree of loss as well ?
Yes, at the end of the day losses in any form will be present as heat/sound. if it has fans running and or is warm to touch then there are losses and thats equally true of AC and DC. But as Finn suggests its losses in terms of free energy anyway…its why I personally never bought into the heat pump hot water thing…if I use abundant PV energy to heat my water then why do I care if its efficient or not. When I have to forgo something else so I can heat the water then my interests in efficiency will rise. Simple with no moving parts is IMHO desirable, and even then the heating element will from time to time fail due thermal cycling… that they are only $30 to replace is a bonus…
yep, my system basically just idles after about 9:30 when my battery is full as exports are limited to 5kWh in Qld. Plenty of excess power.
only by default. but many couldn’t be bothered to fill out the application for higher amounts.
Are you referring to a dynamic connection? Aren’t you concerned about the potential limiting down to 1.5 kW?
Thanks Finn, for uncluttering this issue – very useful – reduces confusion and spin.
What concerns me most though, from a more fundamental point of view, is that a minister in charge of a whole state’s power system could suggest something as daft as converting the state’s power system to DC!?!
This is a discussion/ debate/ contest that was decided more than 100 years ago!
Back then, Edison had DC systems, and Tesla had AC systems. Eventually AC won out, because transformers made it easy to convert to higher or lower voltages, which makes AC suitable for long-distance transmission.
Advances in power electronics now makes DC-DC conversion easier, but certainly doesn’t make converting a whole state’s electrical system to DC even vaguely feasible.
The two quotes I received from SolarQuote rated WA top 2 installers (for a SigEnergy battery) basically asked me to get rid of my existing Fronius 5W Primo despite my requests to retrofit via AC coupling and keeping my existing inverter and panels. My take is that they just want an easy installation with the current demand and dont want to deal with AC couple installations. I didnt proceed and choose to wait until the rush and madness dies down; I am in no hurry and will wait for an installer who is amenable to what I am after.
I think it depends on how old the system is. If replacing old panels can claim the rebate again on those plus the wiring needs to be brought up to new regulations. Extra labour is a big factor here.
Hi Andrew, Maybe try asking around the area you are in and try to get a referral from people you trust. If you have an electrician you have used a bit, ask them if they have a battery and solar, and who put it in.
You could ring and talk to a few locally. You will soon get an idea if they are just sales people using subcontractors. You can ask if they have in house installers, but that will not necessarily be truthfully answered. Smaller the better from my experience.
Solar quotes have great articles and informative discussions. As to getting quotes, I have unfortunately had two complete fails and one satisfactory result using the service.
its about money.
and yes they want to make as much as they can.
and my point is this why rip out what is still working perfectly.
it doesnt make sense
what happened to the customer is right and you quote what they bloody asked for!! and yes you can quote the other option too. and give the customer the choice.
The problem is that the customer then expects the new installer to deal with an issues with the original components, as the installer responsible for the warranty for the old system will probably wash their hands of it as you have modified it.
Or even just the possibility of arguments of what is actually causing the problems etc.
The organisation responsible for the warranty for the new battery may even (will probably) try and claim any issue was caused by the old components.
I wouldn’t add to an existing system either, way to much new work going at the moment without giving your self a potential headache.
Dnsp inverter limits (eg . ausnet in Vic) are the problem when AC coupling. I’m sitting on a 2 year old enphase system (20xIQ8 micros + 2x5p batteries) and at the 15kw single phase limit. If I upgrade to 3 phase, I now need to decide between going just 4 X 5p AC batteries (max out rebates and inverter limits of 30kw for 3 phase) or get a combo of some more panels and fewer batteries. Even then when v2h/g becomes real ausnet will probably limit me on the v2h inverter output. If dnsps just relied on export limiting and let us do whatever we wanted for self consumption then the whole AC or DC coupled wouldn’t matter. But the way it stands, at least with dnsp like ausnet, you’re better off DC coupled with something like sigenergy as you can get away with a lot more battery capacity and even DC EV chargers and never hit the silly inverter limits.
I have a question about DC coupled batteries.
I’m on single phase with 13kw of panels. If I switched over to a DC coupled battery with a 10kw inverter e.g. sigenergy. Added another 7kw of panels for a 200% oversize, could the battery now potentially charge at 20kw?
If so, can I also use/export up to 10kw (inverter limit) while also charging with any excess capacity above 10kw?
I realise it’s probably not super practical, but down here in Tassie we get a lot of overcast days where you might only get 1 hour of good sun and winter never gets close to the inverter 10kw max with 13kw panels.
With Sigenergy gear Inverter capacity doesn’t seem to be related at all to battery charging rates. I have 12kw inverter and 19kw of panels. I can recharge house battery at 12kw (0.5 x battery capacity, in my case 24kwh) and still have full inverter capacity available if i have enough PV that is producing to do so. In my case with 19kw i wont be able to max charge and max out the inverter, what happens in that case is user defined on the app, in my case the inverter gets first call and the battery recharge scales back. My DNSP limits export on a single phase to 5kw, I can max out export and still be consuming 7kw+ in the house. If house consumption grows beyond 7kw then export drops to accomodate
Andy
I’m in this bind currently. Only a 3yo high quality system but to AC couple I need a battery plus another inverter plus controller for blackout etc and the cost to do that was actually more than just going DC coupled and deleting the perfectly good exisiting inverter. Feels wasteful and loose a bit of redundancy but if it costs about the same to ac couple (possibly more) then what’s the point
So dont think in terms of whole system replacements, think more in terms of inverter only replacements. If the pv array is only a few years old then it should be ok to re use. Make sure the new inverter has as many seperate MPPT inputs as possible, at least 3 and better 4. Add new arrays across some of the MPPT inputs and the old across the other. If the array is old enough that you have an on roof isolator switch then pay to have that removed. You may need to have the existing array resized, but probably not because the voltage and current limitations ( more the former) if you stick to single or 3 phase as per the existing inverter then wont have changed.
Inverters have useful lifes of between 10 and 15 years. Good PV should be 25 to 30years. No reason to replace b4 needed. In my case my original PV array was done 2009. Annual losses back then were significant. It was worth it in my case to replace as the watts per m^2 weren’t comparable.
I’m not replacing the whole system, but my inverter (just ) predates being a hybrid one. In which case if I have to buy a second inverter to AC couple the battery then I might as well just replace the existing inverter with a hybrid one. Other than redundancy then I can’t see the advantage in that case
Does the calculus change if you’re wanting to charge an EV? I need to charge 30kwh on average every day. In the winter my additional consumption is around 15kwh. Does the principle remain the same – I’m best sticking with AC if retrofitting?
If you have an AC BEV charger, such as a Level 2 (7kW) unit, then having a PV inverter allows panel output to go more directly to AC generation, granting a few percent greater efficiency than the longer trip through an MPPT and a battery inverter.
But e.g. Sigenstor has a (high power) DC charger option. With that, DC coupling will be optimal, again by a few percent.
My system is 50;50 AC/DC coupled, with both PV inverters for straight feed to the BEV charger, HWS, and aircon, and MPPTs for battery charging, followed by battery inverters, mostly for night-time & clouds.
The battery inverters must usually be higher power rated than the AC inverters, to grid-form in a Black Start, off-grid. Victron handles only 1:1, Selectronic is beefier. A grid isolating (change-over) switch allows you to go off-grid in a power failure. It’s indescribably dufus to allow your system to black out just because the grid does, if you have a battery. Demand a competent installation.
Thanks for posting. My current PV is 6.6kW with a 5kW inverter (Huawei). I am looking to get ~25kWh batteries and thought it would be best to DC-couple (ie. remove existing inverter) because I was looking to join Amber. If I understand correctly, if I were to AC-couple, I would not be able to curtail excess solar during negative prices once the battery is full. I figure with so much solar going into the grid, the negative/low pool prices in the middle of the day are only going to get worse (noting that in the long term, more batteries will soak up the excess and this might be less of an issue some years down the track).
no reason why you can’t. if the inverter is capable of that programming it doesnt matter.
remember solar pv and battery systems are behind the meter.
the meter doesnt know what provided the energy it only knows the current flow direction and the amount.
Thanks – this is why this thread is useful. I am looking into the capabilities of my inverter and software. Huawei SUN2000-5KTL-L1. I have a smart meter, and reading into the FusionSolar documentation, there is a functionality for “enabling negative rate optimisation”, so I am hoping that this would work in my case.
I have emailed the original installer (Solargain), so hopefully they can shed some light. If anyone is aware of someone in Australia setting this up, please point me in the right direction.
Hi Jose,
You’re on the right track with SolarGain. Also have a look at iStore, that’s the brand available in Australia now.
DC PV oversizing to help charge the panels with extra unused PV harvested power is a plus. Exactly what I’m doing with my upgrade. Also, if Solaredge ever release the DC EV charger that they showed off 2 years ago then the DC coupled side of things is great if you’re already in the ecosystem.
There is too much focus of efficiency losses for DC to AC for another reason not mentioned. If your architecture has DC to DC conversions there is also efficiency losses for this. So for people to focus on only DC to AC and ignore any DC to DC conversion, misses half the equation.
For example, you only need to look at something like the Sigenstor. Many YouTubers have talked about its efficiency and sited the DC to DC architecture. But they have obviously not measured the actual efficiency which is disappointing. The reason for this is to maximize the flexibility of the architecture, they have put a DC to DC converter in each battery module. And each of these conversions add to efficiency losses.
So to talk about DC coupled solutions being efficient like it is some sort of rule of physics and ignore the rest of the architecture is misleading. What IS relevant is the round trip efficiency of the whole system. Something that is often hidden from us.
Thanks Finn for bringing back memories of my father’s (an Electrical/Mechanical engineer) 1960’s tutorials to me when I was learning “Science” @ secondary school. Both his knowledge and yours seem surprisingly simple and straightforward. I went on to become an accountant (don’t hold that against me) which has given me a “ second string to my bow” for critically evaluating most proposals regarding solar/battery installations. Keep up the clear and thought provoking articles.
I found my way here because I have had several potential Sigen installers refusing to connect my existing solar array to the Sigenstor (DC coupled). Indeed one told me that the Sigen could black start even while leaving the existing AC coupled inverter in place. I’m a little gobsmacked tbh.
The existing inverter is connected to the grid. If there is no grid, it won’t work. full stop. I also live in an area where days-long blackouts are not rare, which is why I want a system that will black start from the solar array. In mid January last year a massive storm caused widespread blackouts. The next day the sun came out. I was without power for 3 days. My existing inverter sat there doing nothing because there was no grid.
Yet I had an otherwise respected installer telling me no, I can keep the current AC coupled inverter and the proposed Sigen would be able to black start. What gives?
Hi Potential,
DC coupling is best because makes blackstart fundamentally easier no matter what brand.
Generally you can take existing solar and hook it straight up but some installers don’t want the warranty liability.
Some systems like Fronius have superior current inputs, so depending on panel layout/stringing, Sig can’t cope with parallel arrays. ie you need to retain the Fronius as AC coupled and honestly I’d just get a Gen24 to go with it.
Some systems like Sungrow can run AC coupled during an outage, not sure about black start.
Others like Tesla PW2 could restart AC coupled solar in the morning, provided you had a sufficient reserve set.
In any case you must document what you want the system to do before signing a contract.
I’ve had several installers come back to me offering to re-work (re-orient, re-wire) my existing solar array. After having been explicitly told I want to keep the array as-is.
If your current inverter will not provide power up, the Sigen will not black start. It’s that simple.
I’ve not investigated all the attributes of other brands, except GoodWe (which wasn’t suitable). If someone can recommend a good alternative to Sigen, which provides whole home backup, DC coupling and black start, then I’d be keen to know.
I’d like to know if Sigen is telling its installers that they can’t DC couple an existing array, or if its just some installers applying their own interpretations. I’ve had several installers try to push me toward a new array, despite being explicitly told no. Even a brand new array can develop faults and Sigen’s MPPTs are protected from various faults (as you would expect them to be).
Hi Potential,
Many installers are risk averse and simply don’t want to deal with somebody else’s work or products.
Some have experienced unforseen issues when lifting existing arrays for rewiring, like broken/seized/unobtainable hardware, so they just quote whole new jobs. Wasteful but they consider it quality control.
Sungrow offer “whole home” 63a pass through with backup & multiple DC inputs.
Fronius have parallel connection for whole home backup. They only have two inputs but superior current capacity for parallel DC arrays.
Tesla PW3 still makes sense for single phase properties.
It’s impossible for me to judge without much more detail sorry.
But that’s the thing. I don’t want my existing solar panels lifted, re-wired or otherwise touched. I’m happy for someone to inspect the array and if it has any defects or actual safety hazards I want it fixed. But I’ve only had one installer actually go and look at the array (and he said it was fine). The installers who didn’t want to re-use the array never looked at it. The array was installed by a professional installer (a major company) only 4 years ago.
Which Sungrow model are you talking about?
I took a look at the Tesla PW3. It requires the PW and 2 expansion packs. Didn’t make sense cost wise. Plus I don’t fully understand how they handle backup.
On the general theme of DC coupling…
Apart from the issue of black start, one advantage of DC coupling is you bypass the limitations of your existing inverter. My inverter has a strict 5kW limit (more exactly, it’s a current limit). I have a notionally 6.6kW array and this is being curtailed by the inverter. The Sigen doesn’t have this limitation.
Also, when charging, A Sigen alongside an AC coupled inverter is limited to the rating of its inverter. A Sigen DC coupled can use power from both the mains and the solar array to charge the battery. In this case its a 10kW inverter and the battery has a limit of 16kW.
One other subtle detail is that the Sigen has a global power point tracker, which may deliver a small incremental increase in efficiency.
Yes, the 10% you lose going through AC coupling doesn’t matter if you’re designing a system from scratch and have room on the roof to add another panel or three. But it does make a difference if you want the most of an existing array.
In your example the difference is relatively small (though not insignificant) , in mine I have a 12kw single phase inverter and 19kw of PV on the roof. I also have a 12.5kw DC Charger. I regularly see the car charging at full tilt (priority 2) , batterys in the home charging above that( priority 3) and my AC needs in the house also fully met (as Priority 1) . To be clear Ive seen 18.6kw being pulled out of the arrays on the roof and nothing, other than the EV charger running at full capacity. DC coupling in my situation is a beautiful thing. If it was all AC coupled then the car charger would be it, the inverter couldn’t provide another watt beyond….
Andy
Potential,
In the absence of an installer with sufficient initiative to document the work he does for you, thus limiting the scope of his warranty liability, you may need to document the scope of the new work required, and explicitly exclude warranty of the pre-existing system? Keep it *short and simple*.
It is a matter of negotiating, in the most patient and accommodating manner compatible with eventual success. The job has to be worth the installer’s effort to understand and deal with complexity and risk. Optimise reward/risk, as serially whacking in rows of standard new installs is better for the business.
With $5B/0.3 = $16.7B newly thrown at battery installs, the “wire in old array, warranty excluded” needs to be part of a quick profitable zero risk job, I figure.
There will be more of this, as folk incrementally transition, as falling cost, and technology improvements, meet increasing need.
The array must be V & current compatible with a new MPPT string input, though.
Yes, after having dealt with a number of potential installers, I do get the feeling that the subsidy has created a market where installers are happy to take on the easier jobs – to put things politely.
As for the service/warranty issue. What are the risks to a battery installer that come from an existing solar array? Yes, there’s various ways that a solar array can fail. But it’s my solar array and if it fails, that’s my problem. Can a solar array fail in a way that damages the MPPT it is connected to? I can’t absolutely rule that out, but as an electronics designer myself, I find it rather unlikely, particularly if the electronics is well designed – as you might expect from a quality product like Sigen.
Yes, my current array is compatible with the Sigen’s MPPT.
Easily if an array fails in a particularly bad manner it can outright melt/catch fire at the inverter.
The voltage/current of a solar array should not be underestimated.
For the liability its the fact that they could be liable for the other hardware if it gets damaged by the array was well as the usual of the idiots ruining it for everyone else with there being people who will treat it like they’re responsible for damage caused by existing issues in the old install that shows up later by failing in a destructive manner.
Leaving them to either go along with it at their expense or face likely reputational damage from their review/word of mouth.
Though can you even claim the rebate keeping an existing array? I thought it either had to be kept separate from the new system or replaced.
Though that could be outdated info from the 44c feedin period when i heard it.
By what physical process or mechanism can a failure in an array cause damage to the MPPT tracker in an inverter? Solar cells are a series array of diode junctions. They are inherently voltage limited. Similarly, they are current limited by virtue of their physics.
In any case any fault current isn’t going to go through the MPPT in the inverter, it’s going to heat something local to the array or its wiring. So yeah, something might catch fire, but it won’t be the MPPT in the inverter.
A well designed MPPT will have surge protection aimed at dealing with the transmission line effects (and lightning induced fields). A good MPPT has arc fault detection and I’m pretty sure the Sigen has that. There’s also earth leakage detection which will eventually trigger an alarm and I’m pretty sure the Sigen has that too.
My advice from several installers is yes you can include an existing array in the battery subsidy. You just don’t get any special rebates for new panels if you do.
Each MPPT has maximum safe voltage and current specifications. If an array fault current could exceed the MPPT’s limits, then fuses/breakers are required, as the fault current certainly could flow through the MPPT.
On the voltage side, an installer would have to take into account the higher array voltage in cold weather, to ensure the new MPPT has adequate margin.
Each installer knows his own stuff. Pulling panels off to read specs on the back, and calculating cold weather voltage peak, is extra brain work, unfamiliar territory, risk of making mistakes, and just awkward botheration. And on top, “Damnation, we’re behind on installations, ‘cos the rebate’s grown like topsy.”. Nah, I think I can understand why it’s more profitable to do the easy ones.
When the rebate runs out, you’ll find more takers, don’t you think?