Have you had a Sol-Ark Failure?

[HighTechLab]

I am trying to get a better feel for the failure rate of Sol-Ark inverters - my company sells a lot of them and I have only seen 1 out-of-box (DOA) failure, and I saw one guy on YouTube that had his PV shutoff switch go bad and Sol-Ark sent him a new one

One of my engineers keeps telling me they are high frequency and prone to failure under inductive loads, but the statistics don't agree with him. I've personally blown a 12kw Sigineer LF inverter up 3 times - but never a Sol-Ark.

Just wondering if anyone else has made one die, and what it took to make that happen.

 

[Jason.Blier]

I had a 15K fail - but it was due to a faulty bus connection on the grid relay. It burnt up and caused the grid connection to fail. Everything else still worked. Sol Ark sent a new 15K to me immediately and provided an RMA label. 0$ out of my pocket and the new unit hasn’t missed a beat.

-Jason

 

[jasonojordan]

No issues in 3 years with my 12k Solark. Wonderful support all around as well. Either someone available when I call or call backs within the hour.

 

[Oldphile]

I had a glitch, but not a failure. First week after PTO, Sol-Ark 15K (without a battery) would not return to bypass after a power failure. I was able to get it back by power cycling. Sol-Ark did a software update and it hasn't happened since. Prior to this, I had noticed the voltage and frequency trip settings were all wrong or not set (I forget). After the software update they matched the ISO New England requirements.

 

[2TrevorJ]

Only 6 months in but 13.6 mWh produced so far...no Sol-Ark failures. Running my 4k sq ft house day and night (self-consumption).

 

[Lt.Dan]

10 months and 12.1mWh through my Sol-Ark 12k that I got through @HighTechLab and loving it! No issues here other than minor software issues that were fixed with updates.

@2TrevorJ I better start using more power to catch up... lol

 

[niktak11]

13 months and 12MWh PV with a sol-ark 15k and no major issues yet. There was an issue that I reported early on that was fixed by an update (low output voltage with high loads when off grid).

 

[Nobodybusiness]

No problems with my Sol-Ark 12ks or Sol-Ark 15ks in 2 years.

 

[LydMekk]

8.4MWh so far - no issues on 3 8k units.

 

[Lt.Dan]

8.4MWh so far - no issues on 3 8k units.

Do you have those Paralleled? I didn't think that was available under the 12k.

 

[LydMekk]

In 3-phase config, but sure can also be paralleled. No probs. as per the manual.
Must have the exact same firmware though in both cases.

 

[Lt.Dan]

In 3-phase config, but sure can also be paralleled. No probs. as per the manual.
Must have the exact same firmware though in both cases.

I've always seen/read the 8k not being parallelable. Not even for 3 phase. Is this a new update they released that allowed it?

 

[mvonw]

My 12k has been perfect for over a year and half. You better not Jinx me with this posting..
I have a backup in a box. I may actually give it to a friend because it will like be superseded by the time I need a new one…

 

[solar8484]

my company sells a lot of them and I have only seen 1 out-of-box (DOA) failure

What's a lot? Is it a meaningful sample pool?

 

[HighTechLab]

What's a lot? Is it a meaningful sample pool?

Tipple digits per year

 

[niktak11]

Tipple digits per year

Do you stock spare parts or are those only available directly from sol-ark?

 

[Hedges]

Tipple digits per year

Dexter, have you been drinking again?!

 

[Hedges]

0.1% ~ < 1.0% failure after a year, pretty good in my book.

If I'm not mistaken, One out of just 100 after 1 year (if sales distributed uniformly over a year, 50 unit-years of operation), could mean 50 year MTBF.
Except, his experience was one DOA which doesn't count. So not really any MTBF data from his experience. Others had some failures, so there's a top line, but we don't have bottom line number.

 

[HighTechLab]

Dexter, have you been drinking again?!

Only on Fridays after 5. Sadly, it's Monday.

We are a direct distributor.

 

[solar8484]

100-999 units with 0.1-1% failure rate after 1 year is certainly does not indicate a bad initial quality but nowhere near 6 sigma either. Let's see how they hold up at 3 and 5 years.

 

[Nobodybusiness]

100-999 units with 0.1-1% failure rate after 1 year is certainly does not indicate a bad initial quality but nowhere near 6 sigma either. Let's see how they hold up at 3 and 5 years.

Certainly enough data for that with the 12ks anyway.

 

[Hedges]

100-999 units with 0.1-1% failure rate after 1 year is certainly does not indicate a bad initial quality but nowhere near 6 sigma either. Let's see how they hold up at 3 and 5 years.

I think 6 sigma is a process itself more than a quality level, and that quality level might be expected of a component or a process step like SMT.
I don't think it is an expected quality level for a complete assembly like this inverter.
(6 sigma process to allow drift of mean and achieve 4.5 sigma one-sided, which is 3.4 defects per million opportunities.)

Six Sigma - Wikipedia

en.wikipedia.org

Got 1000 solder joints in a SolArk? make that 3400 defects per million inverters, 0.34% (right in the middle of the range we're estimating, although with wide uncertainty bands.)

Of course, some bad solder joints should be found in test, but some will escape and be field failures.

10 DPMO soldering was world class a few years ago, not sure where the numbers stand now.
When we had some issues with protype PCB from a local house, I challenged them by asking what DPMO they delivered.
Sales guy (pathological liar) claimed 10 DPMO. I called BS. Not gonna happen out of the gate with our layout, small quantity, design not properly reviewed for DFx.
There were wrong parts hand placed because out of SMT, some had been dropped, and operator picked them up and hand placed in wrong spot.
I've been an insider before, showed a board to the guy in charge and asked him what defect rate. "30 DPMO". I flipped the board over showing all the red arrows I had applied and said, "3000 DPMO".

 

[Bluedog225]

Sending data to china seems like a fail. ?

 

[robby]

2 years of ownership covering several power outages and Numerous brown outs and she, has not skipped a beat.
its like my house is connected to one big UPS.

 

[ArthurEld]

Sol-Ark 12K, 2.5 years no failures. Damn good thing because I don't know what I am doing and I don't have any help except Sol-Ark.
I never call Sol-Ark. The Sol-Ark is easy to set up and it just works.
I was getting some annoying fault codes when I first set it up but they went away. I never figured out what caused them.
There has been no significant loss of solar production in the 2.5 years. And no significant loss of power to our house either.
I bought it because it was the easiest to set up option. I have not been disappointed.