First Enphase Microinverter Issue?

[Siwastaja]

I am almost never considering battery storage, as all these inverters only support overpriced batteries.

The world is changing reasonably quickly, and becoming flooded with used EV batteries which can be repurposed for BESS, thus avoiding lock in to expensive proprietary batteries.
E.g. https://github.com/dalathegreat/Battery-Emulator/wiki

As always, there is a massive gap in price between DIY and turn-key solutions, it's no news.

Say you want a 10kWh battery storage system. So do you:
* Buy hybrid inverter + battery as turn-key, costing you 13000€,
* Buy the same stuff yourself and just hire an electrician to do the job legally with hourly contract, costing you maybe 5000€
* Buy the same stuff yourself and illegally install it (or legally if you happen to be a sparky), costing you maybe 4000€
* Find second-hand EV battery and build necessary electronics yourself, and get a 20 or 30 kWh battery instead for maybe just 2000-3000€

First case is hard to ever make pay for itself, even if you combine all possible sorts of spot price, PV buffer, frequency containment reserve market, peak power shavings and whatever multimarket optimizations you can find. Last case pays for itself possibly even just for PV buffer + some simplistic spot price algorithm alone.

The sad fact that we tinkerers seem to forget is that 99.9% of population are neither capable or willing to tinker (possibly illegal and dangerous stuff; heck, just commencing a home router is too much for many!) So in the big picture, turn-key solutions dominate, whatever their cost, and us tinkerers getting the same quality-of-life for fraction of cost is an exception to the rule.

[tszaboo]

I am almost never considering battery storage, as all these inverters only support overpriced batteries.

The world is changing reasonably quickly, and becoming flooded with used EV batteries which can be repurposed for BESS, thus avoiding lock in to expensive proprietary batteries.
E.g. https://github.com/dalathegreat/Battery-Emulator/wiki

On one hand I'm in theory an advocate for second life car batteries. On the other hand, you see that most car battery packs are weirdly shaped, and too large for in-home usage. Difficult to move without a forklift. Voltage of the pack is too large for typical inverters. With the exception of VW battery packs. They are ~48V and 32 kg, and look like a brick.
And from what I can see, off the shelf battery packs are often cheaper than these second life batteries, when you include BMS, cabling and all the other things you need to make them work. And then we'll have sodium batteries in no time, I expect them to be significantly cheaper than lithium.

[Siwastaja]

Voltage of the pack is too large for typical inverters. With the exception of VW battery packs. They are ~48V

You are looking at wrong "typical" inverters. SELV inverters are dead; good riddance. Well they are fine for light duty or hobby use, or low-power off-grid,  but for true home batteries, non-isolated, high voltage topology is obvious. It has significantly better efficiency and lower cost, and can supply much more power without requiring ridiculous wire gauges.

Look at any usual hybrid inverter brand. Like Kostal, Growatt, Sofar, Sungrow, Solis, Deye, Huawei, and so on and so on. All have HV battery options. They operate in the shared internal DC bus with the solar MPPT trackers (basically the battery is just one more "MPPT tracker"), which makes the architecture cheap and flexible (i.e., the voltage range is as large as the PV voltage range is, enabling flexible number of series battery modules). Coincidentally, from voltage perspective these are also compatible with almost any second-hand EV battery; all that is needed is a BMS translator (working as a bridge between the EV manufacturer's BMS protocol and the inverter's different BMS protocol).

I agree though that the packs have an awkward shape. It is obvious why home batteries are "towers". Second-hand EV battery takes a lot of floor space and a forklift is indeed needed to get it in.

[nctnico]

IMHO using old EV batteries for home storage batteries is a poorly conceived idea.

1) It is much better to use those batteries to keep existing EVs going. Sensible car manufacturers use modular battery packs which can be repaired partially. Good battery cells / modules have a much higher value on the used car parts market than to make home batteries from. What is leftover is just scrap.

2) As a home battery manufacturer it is much easier to work with a consistent and predictable influx of components than dealing with (what basically is) unknown scrap. Customers will want a warranty after all. How to provide that on used parts? At the end the price is just as high when using new cells. One (example) issue is to properly match cells that go into a pack and have the BMS tuned for the specific chemistry. In the past I have done some work (manufacturing systems and BMS development) for a Li-ion battery manufacturer and they even find rejects among brand new cells.

[Zucca]

You are looking at wrong "typical" inverters. SELV inverters are dead; good riddance. Well they are fine for light duty or hobby use, or low-power off-grid,  but for true home batteries, non-isolated, high voltage topology is obvious.

HV battery implies some serious safety.
48V give me some piece of mind in my DIY installation.

If you can afford the copper, 48V are perfectly fine.
My home runs on 48V, 57.3KWh DIY battery pack.

I am a happy camper with my Victron ESS 48V system.

I would have no problem to switch to an HV battery system and save some copper... but I would have to invest more in insulation stuff.

[Kjelt]

Growatt 4.2kW for 499 euro:
https://www.stralendgroen.nl/product/growatt-min-4200tl-xh-hybride/?v=1a13105b7e4e

This has support for battery storage and it doesn't need a gateway for commissioning.

This is the catch:

Let op: Minimaal 1x35A of 3x35A netaansluiting

This means an additional 700 euros per year out of your pocket to upgrade from 3x25A to 3x35A

For the microinverters being more point of failures, if one fails you still have n-1 working and producing energy.
When one panel in a string of panels fails it is over for the entire string.
Since my panels are 10+m high on a sloped roof I wait till more fail before going up there or call the maintenance company ($$$) 

[Siwastaja]

Let op: Minimaal 1x35A of 3x35A netaansluiting

This means an additional 700 euros per year out of your pocket to upgrade from 3x25A to 3x35A

This must be some kind of copy-paste error or similar brainfart. I have never heard about any requirement like this, it makes no sense. Of course there is no real requirement of a 3x35A supply. These Growatt inverters are installed in hundreds here, to our usual 3x25A contracts.

[nctnico]

Growatt 4.2kW for 499 euro:
https://www.stralendgroen.nl/product/growatt-min-4200tl-xh-hybride/?v=1a13105b7e4e

This has support for battery storage and it doesn't need a gateway for commissioning.

This is the catch:

Let op: Minimaal 1x35A of 3x35A netaansluiting

This means an additional 700 euros per year out of your pocket to upgrade from 3x25A to 3x35A

It is a single phase inverter for single phase systems. At these kind of power levels, you'd get a 3 phase inverter if you have a 3 phase supply. I'm on a single phase supply with a 40A primary fuse. IOW: it is not a problem at all but they have to specify that you have to have a suitable primary fuse rating for this inverter to operate. The inverter itself is supposed to connect to a 25A circuit and you'll need some room to have enough selectivity between the primary fuse and the fuse the inverter sits behind.

[David_AVD]

For the microinverters being more point of failures, if one fails you still have n-1 working and producing energy. When one panel in a string of panels fails it is over for the entire string.

AFAIK on mine only a disconnected cable between panels could cause the whole string to fail. I do have optimisers on them though so get per panel reporting.

[tszaboo]

Growatt 4.2kW for 499 euro:
https://www.stralendgroen.nl/product/growatt-min-4200tl-xh-hybride/?v=1a13105b7e4e

This has support for battery storage and it doesn't need a gateway for commissioning.

This is the catch:

Let op: Minimaal 1x35A of 3x35A netaansluiting

This means an additional 700 euros per year out of your pocket to upgrade from 3x25A to 3x35A

For the microinverters being more point of failures, if one fails you still have n-1 working and producing energy.
When one panel in a string of panels fails it is over for the entire string.
Since my panels are 10+m high on a sloped roof I wait till more fail before going up there or call the maintenance company ($$$)

A failure on the 230V side would bring down an Enphase system just the same way.
But the real failure point for the Enphase systems is software, the gateway, their terrible subscriptions, their mobile app. They sound and act like one of those companies that could push a FW update any time that will brick your device on a moment's notice. I don't trust them at all.

[Kjelt]

But the real failure point for the Enphase systems is software, the gateway, their terrible subscriptions, their mobile app. They sound and act like one of those companies that could push a FW update any time that will brick your device on a moment's notice. I don't trust them at all.

Mwah the gateway was optional so the owner can see the statistics. You did not have to buy the gateway, or even use it. Pull the plug of the gateway and the IQ7s remain operating and producing, don't see the fuzz.

[nctnico]

For the microinverters being more point of failures, if one fails you still have n-1 working and producing energy.
When one panel in a string of panels fails it is over for the entire string.
Since my panels are 10+m high on a sloped roof I wait till more fail before going up there or call the maintenance company ($$$)

Then again, a solar panel is a mostly passive device. Very unlikely to break down. As I wrote before: put as less as possible gear on the roof as it is difficult to access in most cases. Which gets to the point where you need to think about whether retrofitting roofs with solar panels is a good idea to begin with. For my own install I made sure I can access all the connectors from the windows in the roof. If there is a problem in one of the panels, I can disconnect / bypass it without needing to go on the roof.

[EEVblog]

I would have no problem to switch to an HV battery system and save some copper... but I would have to invest more in insulation stuff.

I'd only go for a HV system if I had a long run.
Copper is expensive, but it's not that expensive.

[Siwastaja]

Mwah the gateway was optional so the owner can see the statistics.

... which makes the thing unappealing bait-and-switch. "Here, buy this more expensive, more complex, less reliable premium system, but hey, it gives you fancy extra features like per-panel monitoring, which also helps with the unreliability!" ... "you can buy it without the monitoring!" This is like buying a Ferrari with 1.4-liter motor as a cheaper option, no one wants it, it makes no sense.

This same pattern is all over microinverters. "Buy this premium thing which produces more from the same light!" ... "By the way, it has worse efficiency, it has insufficient current rating so MPPT flats out in optimum conditions and you lose production, oh and it thermally throttles and you lose production."

All that is left is mental image of premium, with all premium really stripped off, because their manufacturers failed to keep improving their products. Even failed to keep them up-to-date. And I don't blame the designers, it's a very difficult engineering task.

During the same time, string inverters gained two or three MPPT trackers, remote monitoring, and nearly free-of-cost battery port. Just what people wanted and needed. And they do it for cheap.

Really, the competition is already over.

I can still see there is potential in distributed inverter topology but it needs some significant change in mindset how these things are designed; better integration and even more serious cost-cutting without undermining quality. A harsh equation.

[Kjelt]

I don't see it that black/white Siwastaja.
There are other advantages , you can place small setups like two or three panels standalone anywhere you want on your gardenhouse, on a small flat roof.
You can distribute easy over the phases, even unbalanced if you know one of the phases is heavier loaded. etc.
I choose for it two years ago not for premium but exactly low maintenance if one dies like this topic and more in the coming years.
The priceup in that time was 8% so yeah couple of hundreds euro's. Not a big difference as nowadays if I take your word for it.
Also not to have a large box somewhere in the house that needs replacement every ten years.
That big box might now be used for batterystorage, but I am thinking about building it myself and rather not have it in the house 

[David_AVD]

When I got quotes for Enphase micro inverter and Solar Edge string systems, the Enphase was more expensive despite being around 30% lower in output IIRC. The Solar Edge system also has optimisers on every panel.

[tszaboo]

When I got quotes for Enphase micro inverter and Solar Edge string systems, the Enphase was more expensive despite being around 30% lower in output IIRC. The Solar Edge system also has optimisers on every panel.

I have the SolarEdge for the past ~5 years. The system always works with optimizers, the inverter is a dumbed down version of regular inverters. The optimizers will generate a fixed voltage, so panels with shading will make 20V 5A for example, panels without shading will make 60V 5A. The total string will make 400V as I recall. Then the inverter only needs to go from this 400V fixed voltage to 230V AC, much simpler than MPPT. The optimizers according to teardowns don't have any electrolytic capacitors in them, and they are completely potted.  So no components in them with limited lifetime, the only way I see this broken is thermal cycling induced, or surges. And unlike the Enphase inverters, the optimizers are rated for more power than most panels, no clipping. Well, there could be some clipping, as I have a 3.5KW inverter on a 4KW nominal DC system, but thats rarely an issue for the Netherlands
It costs more, in my case I think the total system price was ~5% higher than string inverters. But then I have the piece of mind, and the per panel MPPT. If the system pays back in 5 years, it might take 3 months more for it to pay for itself, worst case. Considering that it generates more ,because I do have shading, mine probably paid for itself faster than a string. As far as compromises go, IMHO it's a good one.
But that was 5 years ago, panels are cheaper now, the economics have shifted.
If you are only after ROI, there are systems that are free here. Some people will give away their 5-10 year old system, you only need to go to disassemble it yourself from their roof. That's hard to beat in price, but I don't really feel like working on a 45 degree slanted roof.

[nctnico]

I don't see it that black/white Siwastaja.
There are other advantages , you can place small setups like two or three panels standalone anywhere you want on your gardenhouse, on a small flat roof.

Even in those situations a string inverter wins. A Growatt 600W inverter costs less than 100 euro.

[Kjelt]

Even in those situations a string inverter wins. A Growatt 600W inverter costs less than 100 euro.

Does it also have 25 years guarantee?

[Siwastaja]

Does it also have 25 years guarantee?

It is easy to give guarantees - another thing to respect them. Guarantees are marketing things. For example there is a long tradition that the cars known as being most reliable have shortest guarantees. A company which is a sinking ship has nothing to lose, and stuff like building illegal inverters saving costs of the mandatory isolation relay, or giving 25-year guarantees even when you know that you have huge return-% during that time, are last resort attempts to save the business. The very reason for the guarantee is that the whole industry "knows" that microinverters are unreliable. Probably the image of their reliability issues is exaggerated, they are not that bad really, but the image is very poor, affecting sales. What else can you do but to respond by a long guarantee?

Same thing as Korean car manufacturers respond with 7-year warranty when people think "they are not as reliable as the Japanese cars, which I buy anyway even if it only has a 3-year warranty".

Also not to have a large box somewhere in the house that needs replacement every ten years.

Sorry if I'm blunt, but that sounds like the weirdest mental acrobatics I have seen in a while. If we accept the premise that even a well cooled inverter is cumbersome because it fails and needs a replacement (in an easy access location) and that counts as a downside, then I have very bad news for microinverters.

To me and many others, the fact that the box is inside and replaceable are positive features.

But I have seen this a lot in microinverter marketing: take the most obvious disadvantages, and turn them around 180 degrees. "Better reliability" and "more easy replacement" are examples of claims I actually have seen before. But it is totally ridiculous from both logical viewpoint and practical experience, if you think about it.

[nctnico]

I fully agree about having the inverter inside in a place where it is easy to access. Replacing it takes like 10 minutes (assuming the new inverter has the same connectors). And it is not outside the realm of possibilities that future government regulations mandate replacing the inverter for a model with grid management features. I don't see that working well when you have micro-inverters on the roof.

For my solar install I planned having to replace the inverter for several reasons: defect, support for home storage or changing government (grid management) requirements.

I also agree with warranties are something to take with a grain of salt. Enphase's warranty only applies to the cost of the product itself. Not the cost for doing the actual replacement (which can easely add up to a multitude of the cost of a microinverter). In the end a warranty is an insurance you pay for. Enphase's bet is probably that few people claim the warranty beyond the warranty period from the installer because the cost for sending somebody up on the roof makes it financially unviable. Some people in this thread have already noted that they'll wait for several microinverters to fail before sending somebody up the roof due to cost.

[Jeroen3]

The Germans are already doing Paragraph 14a.

Meanwhile, of my inverters isn't reporting... 
The one on the highest part of the roof

[Siwastaja]

The Germans are already doing Paragraph 14a.

It's good to have some standardization. Hopefully it will be KISS. Because the current situation isn't at all that bad. At the moment, our box for example interfaces with seven different hybrid inverters and four different PV-only inverters. Introduction of support for new models has been completely my job in addition to everything else I am doing, not a huge workload at all, usually it takes a day or two with the Modbus documentation and testing per inverter.

So it's entirely possible to live with the current generation of devices - their controllability and communication features are already good enough - and small bits of driver layer modules. Compare to that, SunSpec already exists but the sole SunSpec compatible inverter we support also implements their own non-sunspec interface which is so much simpler that we use just that. SunSpec is a great example of failed standardization. We need something simpler which the Chinese hardware companies can implement without huge software teams.

Batteries specifically need some seriously intelligent control to make any financial sense. As PV buffers, they never pay for themselves. We do combined spot price + PV optimization which provides steady flow of savings but even that is too little. Now the big trend are the frequency containment reserve markets and for last 2 years quite lot of money has been on the table, but the market is already collapsing because it is so easily satisfied with a few larger battery sites. It never goes to zero however, so combined spot+PV+FCR optimization is now what we are chasing and which hopefully gives us competitive edge against those who believe that FCR will produce endless flow of money forever.

And in the meantime, 99% of batteries sold globally to consumer end-users end up being seriously underutilized, expensive and heavy piles of valuable metals sitting in the homes, doing nothing. That's really sad.

[EEVblog]

The sad fact that we tinkerers seem to forget is that 99.9% of population are neither capable or willing to tinker (possibly illegal and dangerous stuff; heck, just commencing a home router is too much for many!) So in the big picture, turn-key solutions dominate, whatever their cost, and us tinkerers getting the same quality-of-life for fraction of cost is an exception to the rule.

Yes, this is how the world works, and solar is no exception. Doesn't seem like Enphase are going bankrupt.