SolarEdge Optimizer Mistrust - Fact Based?

[steveholtam]

Hi all. I've been around this board for a few months and have learned quite a bit about the competing technologies and such. I see some users that love the SE Optimizer product, and some that don't really like it. The normal fears being extra complications, costs, and the worry about the electronics being on the roof. I can understand some of these fears, but no so much with being on the roof.

Yeah, roofs are hot. But so is the environment under a car's hood, on a motherboard, inside of an oven, etc. There are plenty of hot man made environments where electronics have been built to withstand the conditions perform just fine for decades. Here in California they often place heater and A/C units on roof tops as well. Inside which is filled with all sorts of electronic parts.

I was just wondering since they have been mass producing optimizers since 2010 if any repair persons have seen large number of units failing? Is the issue a hypothetical myth? Six years is not 25, but there are some really punishing areas of the country that would see the first signs of heat fatigue before most of the US.

Thanks!

[J.P.M.]

1.) Facts seem harder to come by these days - at least the ones with proof behind them and from sources with no skin in the game. I've got no reason to believe failure data on micros/optimizers would be that much different/better.

2.) Reliability is sometimes a relative term. 1 failure out of 1,000 may be acceptable, but it's still 10 times worse than I failure in 10,000.

3.) Optimizers and micros may well prove to have acceptable reliability. However, the increased number of failure points seems to run counter to what I learned to be common sense with regard to design - more parts - more complication - -->> higher probability of failure, therefore not the best way to go, depending on the particulars of the application.

4.) Electronics in my garage - a semi controlled environment, may be no more robust than those of an optimizer sitting on my roof, but I bet the easier environment of the garage will serve to decrease the probability of failure a bit, and the greater simplicity of design will increase the reliability a bit.

5.) Some conjecture on my part, but it seems to me that after a few years, the bloom is probably off the solar rose for a lot of PV owners, who don't, as a general observation, seem to understand or even may much attention to system output a shot time after install. If so, and 1 or 2 optimizers or micros fail a few years down the road, I wonder how many owners will notice that decrease in output as readily as they'd notice a complete shutdown if a string inverter headed south, sort of like a built in alarm. Maybe an unintended or unconsidered advantage to the all/nothing output tendency of string inverters.

[solarix]

Ya, and none of those other consumer products are warranted for 25 years. Most all consumer electronic products are warranted for just 1 year. I'm sure the recent SolarEdge stuff is doing better, but (fact) I had a horrendous failure rate with their early stuff and am sticking with installing the one (1) brand of inverter that hasn't disappointed me.
I think the appeal of the module level electronics is that it feels like having a smartphone on each panel - trouble is, how many smartphones are meant to last 25 years?

[emartin00]

From what I understand, the electronics in an optimizer are far more simple than a micro inverter. SolarEdge's reliability has been excellent as far as I know.

[sdold]

From what I understand, the electronics in an optimizer are far more simple than a micro inverter. SolarEdge's reliability has been excellent as far as I know.

If true, and if they are more reliable, that's an upside to the optimizer, but if one (or the inverter itself) failed and needed to be replaced by the same part, it would be a bigger downside to me being tied to one manufacturer. So my choices would be string inverter, microinverters, or SE with optimizers, in that order. I've been around long enough and seen too many "solid" companies go away to put all of my eggs in the SE basket.

[jflorey2]

From what I understand, the electronics in an optimizer are far more simple than a micro inverter. SolarEdge's reliability has been excellent as far as I know.

Simpler, certainly. But "far more simple?" I would disagree there. The optimizer still needs an MPPT controller (which means a processor) a buck/boost, a housekeeping supply, communications, surge/ESD protection and at least four external connections. All the basic parts you need for a microinverter (caps, inductors, power switches, housekeeping, references etc) are there with the possible exception of the isolation transformer.

[DanS26]

Here's what the company says about their warranty reserves.....from their S-1 filing:

"Product Warranty

We provide a standard limited product warranty against defects in materials and workmanship under normal use and service conditions. Our standard warranty period is 25 years for our power optimizers and 12 years for our inverters. In certain cases, customers can purchase extended warranties for inverters that increase the warranty period to 25 years.

Our products are designed to meet the warranty periods and our reliability procedures cover component selection, design, accelerated life cycle tests and end of manufacturing line testing. However, since our history in selling power optimizers and inverters is substantially shorter than the warranty period, the calculation of warranty provisions is inherently uncertain.

We accrue for estimated warranty costs at the time of sale based on anticipated warranty claims and actual historical warranty claims experience. Warranty provisions are computed on a per-unit sold basis, are based on our best estimate of such costs and are included in our cost of revenues. The warranty obligation is determined based on actual and predicted failure rates of the products, cost of replacement and service and delivery costs incurred to correct a product failure. Our warranty obligation requires management to make assumptions regarding estimated failure rates and replacement costs."

Just watch the financials and compare the rate of growth of warranty expense to sales growth. Let the accountants tell you how good their products are in the long run. Everyone else is just guessing.

[steveholtam]

Thanks for everyone's thoughts. The S-1 filing is interesting, as even SE is guessing on the longevity. lol. And it seems like there is no real-world data to indicate issues at this point in time. Regarding cell phones lasting 25 years, for all we know they would last, but who wants a 25 year old cell phone. Yuck.

I agree with the "hunch" factor about the electronic on the roof, the other benefits to me such as PowerWall compatibility and Panel Monitoring, outweighed my hunch during my final deliberations.

One final though, like JPM, I have my inverter in the garage. But I've seen a ton of them on YouTube mounted outside. Seems like that is a bad idea to me as well. Wonder why they do that.

[dennis461]

You started the post with a question, statement, or myth regarding reliability of a particular electronic product with no data to support it.
You asked, "...any repair persons have seen large number of units failing?".

Of course, no one responded to that, possibly because the product is too new, or there is no way to quantify the number of units sold for repair vs new installations.
The same might apply to micro-inverters as well.
I suspect, the electronics is very similar, (optimizer vs micro-inverter), I've opened up a microinverter and it seems as complicated as a desktop PC.

As for replacing a product under a panel, most installations are 3-4 rows on the short side, Removing two panels to reach the third is not a big deal (for a person with experience.)

[steveholtam]

I would hate to be the person fixing my bottom row middle panel.

Panels1.jpg

[J.P.M.]

Here's what the company says about their warranty reserves.....from their S-1 filing:

"Product Warranty

We provide a standard limited product warranty against defects in materials and workmanship under normal use and service conditions. Our standard warranty period is 25 years for our power optimizers and 12 years for our inverters. In certain cases, customers can purchase extended warranties for inverters that increase the warranty period to 25 years.

Our products are designed to meet the warranty periods and our reliability procedures cover component selection, design, accelerated life cycle tests and end of manufacturing line testing. However, since our history in selling power optimizers and inverters is substantially shorter than the warranty period, the calculation of warranty provisions is inherently uncertain.

We accrue for estimated warranty costs at the time of sale based on anticipated warranty claims and actual historical warranty claims experience. Warranty provisions are computed on a per-unit sold basis, are based on our best estimate of such costs and are included in our cost of revenues. The warranty obligation is determined based on actual and predicted failure rates of the products, cost of replacement and service and delivery costs incurred to correct a product failure. Our warranty obligation requires management to make assumptions regarding estimated failure rates and replacement costs."

Just watch the financials and compare the rate of growth of warranty expense to sales growth. Let the accountants tell you how good their products are in the long run. Everyone else is just guessing.

That seems a pretty objective way to get a number. Haven't you suggested something similar in the past ? Wonder how many folks will do it ?

BTW, who covers the labor and shipping ?

[J.P.M.]

One final though, like JPM, I have my inverter in the garage. But I've seen a ton of them on YouTube mounted outside. Seems like that is a bad idea to me as well. Wonder why they do that.

One reason might well be it's easier for the installer. Another might be that many in CA and the western U.S. are outside is because basements are nowhere near as common. Sooo... garages fill up with the junk that basements usually get. Soooo.. Space is at a premium. Or, I've heard folks complain that they don't want to look at an ugly inverter in their garage (next to the lawn mower and water heater I suppose).

As for the garage being hotter than the outside and thus bad for an inverter, and strictly anecdotally, my inverter runs about 30 F. warmer than the garage amb. temp. Also, and not that it matters for the garage to inverter temp., but the garage ambient temp. is usually close to the outside ambient temp. Also not important but FWIW, my super spiffy and very high tech window fan cooling arrangement for the inverter drops the inverter to gar. amb. temp. diff. to ~ 20 F.

Unless building codes are involved, my money's on the installer making life easier for installation and access for a lot of the outside mountings. I wonder how many disappear at night ?

[solarix]

The SE warranty may well cover replacement of the optimizers, but doesn't cover labor (at least, I'm still waiting for them to compensate us for all the warranty work we've done for them).
This roof-mounted electronics is going to bankrupt a lot of dealers down the road as they have to go out and replace this stuff one at a time)

[DanS26]

That seems a pretty objective way to get a number. Haven't you suggested something similar in the past ? Wonder how many folks will do it ?

BTW, who covers the labor and shipping ?

J.P.M.......very observant. Yes I have offered this advice in the past, and it was during the large run-up of Enphase inverter installations. The earlier generations of Enphase products saw a disproportionate increase in warranty reserves compared to sales increase of the product. A red flag for anyone studying the financials of a company.

Then Enphase changed their accounting method for accounting for warranty reserves. Another red flag. Companies do this to hide real problems. The accountants seldom lie....but marketing, manufacturing and management will lie like a rug.

[J.P.M.]

J.P.M.......very observant. Yes I have offered this advice in the past, and it was during the large run-up of Enphase inverter installations. The earlier generations of Enphase products saw a disproportionate increase in warranty reserves compared to sales increase of the product. A red flag for anyone studying the financials of a company.

Then Enphase changed their accounting method for accounting for warranty reserves. Another red flag. Companies do this to hide real problems. The accountants seldom lie....but marketing, manufacturing and management will lie like a rug.

I remembered the simple elegance of the logic.

On accountant's veracity: My first wife was a C.P.A. One of her favorite stories was about an accounting mgr., a marketing mgr. and a mfg. mgr., all in the GM's office discussing ways to explain a cost overrun on a gov. contract. The GM asked the mfg. mgr. how he'd explain it. The response seemed plausible. The GM asked the same question to the marketing mgr. That response also seemed believable. The GM then asked the accounting mgr. the same question. The chief accountant simply looked at the GM over the top of her glasses and said: "Tell me which one you like better and I'll make that one work".

The best humor has a kernel of truth in it.