Solar death spiral?

4 05 2015

Every day, I’m bombarded with information about some place somewhere going 100% renewables by (insert your favourite date here), and lately Tesla’s new battery wall being a ‘game changer’.  Let me tell you, from where I sit, I predict a renewables death spiral to match the grid’s.

This feeling started in ernest recently when our newest grid tied inverter died for the fourth time in 4½ years.  If you are new here, you may not know I was already complaining about our Kinglong/Sun Teams inverter right from its initial installation.  I took it down, drove it to Solazone’s office from where it was inspected and found to have been ‘storm damaged’.  Except there was no storm.  It just stopped.  The other inverter was fine, nothing else was damaged in the house, this was just plain BS to get out of replacing it.  Kinglong no longer operate in Australia, tough luck…  I didn’t want to replace it anyway, I was over those Chinese pieces of crap, and I didn’t care if it was replaced by a used device, as long as it was a reputable brand I could trust.  And believe me, they are hard to find now…..

Pallet load of dead Kinglong inverters

Pallet load of dead Kinglong inverters

I started looking on eBay and Gumtree for second hand inverters (I simply cannot afford a new one) and found heaps of them.  It seems every second person in Australia is upgrading from their small initial investment to the full 5kW, or even bigger.  Trouble is, they are nearly all Chinese rubbish…..  Many of the inverters for sale were branded with names I was unfamiliar with, and googling for info on them soon revealed so many problems I was blown away.  I even found pallet loads of dead inverters available ‘for parts’ on eBay, including lots that were the same as my defunct one……  and these failures are driving solar installers to bankruptcy.

I eventually found a used Xantrex, the brand I am now kicking myself for not buying in the first place, and installed it yesterday.  It’s almost 5 years old, but according to its original owner was only in service for about 2½ years.  Mind you, he ‘works’ in the solar industry, and I would not trust him any further than I could throw him, and he was big and heavy!  So far so good, the Xantrex works, and we’re feeding power back into the grid from the second array after a break of at least two months.  It will have cost us $200 at least in lost revenue, but there you go.  It had to be fixed, could not sell the house without it.

Xantrex inverters, however, are no longer sold in Australia.  It seems the bigger 5kW ones had problems.  Xantrex is a Xantrex GT 2.8 - 2800 Watt 208/240 Volt Grid Tie Inverter - Xantrex ...Canadian company with a reputation for quality; they make all sorts of other things solar related including battery chargers and other assorted peripherals, but the inverter I bought is ‘assembled’ in China.  Is this a problem?  Is it the problem??

I like these devices because of their huge heat sink at the front, rather than the normal ‘sandwiched between the hot inverter and the wall’ manner.  Plus they have two MPPTs (Maximum Power Point Trackers) which means whoever buys our place and wants to add another kW can easily do so by plugging it into the unused input plugs.

The guy who sold me the inverter had the gall to tell me about ‘all the cowboys’ in the industry closing up shop when they could not keep up with warranty claims…  when clearly he is one of them, he’s just new to the solar Matrix, and his turn will come.  Then he’ll have to sell the Beemers parked in his garage.

I have now met so many people with failures it’s bewildering.  Our neighbour up the hill had a Sharp inverter, installed by his electricity supplier no less, which failed within months.  He too upgraded to 5kW with a Dutch made Nedap inverter from a company in Nambour which no longer exists….  I just checked!  Another neighbour, the one with 10kW I’ve mentioned before had a SMA installed which had problems right from the start and was quickly replaced, proving that even the best are not infallible.  The lady at the back of us had a Solazone system installed (I sold it to her) and her Sunteams inverter died too.

Because all my tools are now parked in a shed in Tasmania, I had to borrow a cordless drill from a friend to finish my installation.  When I told him I was replacing a dead inverter, he told me his neighbour was also having problems and maybe I could have a job helping her out as she’d had two ‘cowboys’ come to have a look who told her completely different stories…. and these are all within walking distance from here.  You can even find lists of defunct solar power companies like this one.  It only goes to 2013, I’m sure the list is an accelerating one…

Go to whirlpool forums, and you’ll find more info about inverter failures than you can poke a stick at….

As the industry matures, failures will smartly pick up.  It might be a good thing for the industry as they sell more and more junk with inbuilt obsolescence, but you have to feel sorry for the people who unlike me have no idea how to repair their systems themselves and have to fork out maybe $2000 or more just to keep their solar systems going.

John Michael Greer recently published an article about the demise of the internet not due to technical problems, but rather economic ones.  I find it difficult to not feel the same way about the solar industry.  Methinks all the optimism about ‘100% renewables’ is highly overcharged.  Except of course we will one day be running on 100% renewables…..  because there won’t be any fossil fuels left to burn!  Which will of course mean the closure of all renewables factories and the return to the one true solar god, photosynthesis….




10 responses

4 05 2015
Chris Harries

This is very relevant, Mike.

The lifespan of solar panels, for calculation purposes, is generally put at 30 years (albeit with about 10 decline in output by that time), though many solar enthusiasts insist that they will last much longer than this.

I’ve come to a conclusion that when calculating the lifespan of any technology it’s best to use conservative figures. The stainless steel kitchen sink is a good yardstick. With a theoretical lifespan of over 1 million years what can go wrong with a kitchen sink? Yet they are thrown out in their tens of thousands when only 20 years old. Owners decide that they want a double sink, or the existing one doesn’t fit into decor and so forth.

The reasons for obsolescence are different for every technology. In the case of solar panels a proportion will come to an end as a result of natural disasters and house fires and this can be built into the statistics. Mike’s focus on the much shorter lifespan of the inverter in the main one. Even if the inverter problem is eventually overcome for new systems, all of the current stocks of millions of solar panels will be encumbered with this limit.

What will owners do in, say, 15 years time when they have an inverter failure and will be confronted with having to buy and install a new inverter? And how will this factor into their nominal payback of their initial solar investment? And will they use that failure to make another choice… say, to install newer solar panel technology that will have been developed by then?

In the next ten yours our galvanised roof will need to be replaced and that would entail the solar panels being removed and re-installed and I’m wondering what the then owner (if not us) would decide at that point. More than likely these older style panels would be considered like an old steam train and they may be sent to the recyclers. With the never ending introduction of newer, smarter technology older technologies tend to be jettisoned long before their theoretical use-by date.

When calculating the actual lifespan of a technology there are so many negative factors at play. It’s best to use conservative estimates.

5 05 2015

Kitchen sink life depends on fashion. I don’t think fashion will have too much effect on solar power system life, although I think that fashion may have had some small effect on its takeup rate.

5 05 2015
Chris Harries

Agreed, Don.

What I said was : “The reasons for obsolescence are different for every technology.” I used kitchen sinks simply as an illustration.

There are several factors that will shorten solar panels’ life span. Aside from the inverter problem, above, I think the main one (as is true for most electronic technologies) is rapid design obsolescence. By 30 years existing solar panels will be seen like steam trains and typewriters. We’ll want to shift to the latest technology on offer, as people do every day with computers. What will precipitate this more so will be inverters failing because people will have to do something to keep their old system going.

I expect some owners will just leave them up there and so numbers of panels will simply fall into disuse, because most people will still be on mains power (unless they go off-grid) and it takes a bit of effort for busy people to hunt around for a replacement inverter and get someone to install it retrospectively.

The point I was making is that theoretical life spans almost never measure up to reality, for any technology. It’s best to use conservative figures to be real.

4 05 2015

Unless solar panels are poorly made, and this happened to early BPs which suffered from corrosion due to water ingress, PVs should last an awful long time. NASA has currently 50 year old PVs that still work. The fact they put out less power than when new is largely irrelevant, we will one day use much less anyway!

PVs are also not that prone to natural disaster damage. My old boss at Solazone (the one who let me down with this latest failure) likes to tell the story of a customer whose roof was so badly damaged by hail it had to be replaced. He got the job (as the original installer) of removing the panels….. and putting them back up when the roof was fixed because they were perfectly fine!

I’ve now had to replace five inverters in ten years, if you count all the warranty replacements. In retrospect, they were all junk, even the first German made one. Which proves provenance doesn’t have much to do with anything.

Inverter failures will happen at far shorter intervals than 15 years Chris…. On average, they are under five years from what I can tell. Unlike PVs that just sit there doing basically nothing, inverters have a hard life continuously adjusting outputs for clouds, weather, seasons, and relentlessly looking for the maximum power point using sophisticated electronics that get hot and are exposed to the weather. It’s a recipe for failure if you ask me… though motor car electronics seem to be amazingly reliable by comparison.

I’ve seen entire second hand 2kW systems for sale on the internet, for as low as $600. One had a Sunnyroo inverter that almost certainly didn’t work, they were just about the very worst model out there with their promoters doing a runner last year… A friend of mine who bought a 1kW system six years ago when even more cowboys were ‘giving them away for free’ in exchange for the rebates (when they were ridiculously high) has just bought a duplicate system from Gumtree to tack onto the 2kW inverter he had the presence of mind to pay for thinking he’d upgrade later. It’s an SMA, so hopefully will last another few years yet, but he picked up the panels for peanuts…

He knows enough, and I’ve helped him, that it will be finished properly; but I wonder how many of these second hand systems will end up in the hands of idiots with no idea of what they are doing, even possibly installed illegally and with no regard to safety?

4 05 2015
Chris Harries

Yep, Mike. That’s all too true. Solar panels won’t survive a house fire that demolishes a house, of wild storm damage that sends them flying around the country-side. Agreed that these will be lesser factors but they will add to the sums when all factors are take into account.

I guess you are saying that the inverter problem aside, the 30 year plus lifespan is probably quite sound? I’m happy to stand corrected. Was just speculating that it’s best to be conservative when making estimates of technology lifespan.

I hear that new technology will very soon overcome the inverter deficiency, in which case solar systems from now on may indeed end up with 50 year plus lifespans – though my mind boggles as to what will transpire in the next 50 years, that will change everything.

4 05 2015

I haven’t heard of this new inverter technology….. have you got more info?

4 05 2015

Just found this……

However, if you have a 20x250W 5kW array and one of these fails, you may not notice, for YEARS if ever. Even if 2 or 3 did, you might just put it down to age related derating. Then there’s the issue of panels getting very hot in the sun, and electronics attached to these might not last as long as they think. After all, NO ONE knows how long anything will last until it’s actually in service, and then long enough for failures to occur.

4 05 2015
Chris Harries

Yes, that’s true too. A couple of years ago I made an interception when a school groundsman, observing the school meter, decided that the enormous school pv system wasn’t delivering as it should. The principal brushed him off so he gave me the readings and I concluded the same. Consequently the solar provider was brought back in to check it out. Sure enough only half of the solar system was feeding, the other half being installed incorrectly.

The lesson here is that if it wasn’t for the (unusually observant) school groundsman the pv system may have operated for decades without anybody noticing. Hard to imagine a coal fired power station or a big hydro-electric system reduced output by 50% and nobody noticing!

4 05 2015

Tesla is to blame. He convinced Westinghouse that AC was superior to Edison’s DC-based power system and we have ended up with this inverter legacy. I don’t know if the micro-inverter is the ultimate solution, but at least replacing a small inverter has to be a lot cheaper than the one big one.

Maybe the solution is getting rid of the inverter completely and going to a DC system. That has its own problems, of course, but they are solvable. They tend to arc when an appliance is plugged in, for instance, but a current inrush limiter can solve that problem. DC appliances are mainly targeting the RV market today.

Or, maybe the solution is two electrical circuits, one DC and one AC for those appliances which are too expensive for DC, or just have to have AC. That would trim the needed inverter size and cost down.

4 05 2015

My two bob…
I’m with Robert on this one, go DC as much as you possibly can, and get off the grid! I reckon 24v is the sweet spot – v drop is less of an issue than 12v and you can still find ample 24v appliances as opposed to 48v (eg. Laptop chargers, Led light bulbs, etc). I’ve got a small 24v 250ah Gel set-up with a cheap Jaycar 30A MPPT controller that runs DC lights, radio, chest fridge, chest freezer, laptops. Then for the bigger AC loads (washing machine, kettle, electric lawn mower) I’ve got a 5kw PowerJack (chinese) inverter thats simply switched on when needed. If you get in touch with the supplier on e-bay and pay a fair price, these are simply the best value inverters on earth IMHO. Great boards, beefy toroid tranny’s, decent cooling and replacement boards can be bought/self installed if need be. Don’t snipe a cheap bid though, or you’ll be shipped one of the dodgey night-shift units with screws still rattling around in it and under rated tranny’s! 😛

You should seriously consider one of these when you get to Tassie Mike. There’s a bloke on Flinders Island who loves em too, and does his own mods. Look up OzTules on the Anotherpower forum and you’ll find an active thread all about em.

Longer term though, I want bigger Ni-Fe batteries (I think they need to get much cheaper though, exchange rate doesn’t help either), a MidNite MPPT Controller, and a bigger 8Kw PowerJack Inverter.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s