Black Swan Event Final(?) Report

Hmmm, two posts sans images in one week.

I’m not sure where exactly I left this tale of my formerly dysfunctional hybrid power system for the barn, but following the replacement of the solar controller guts by the manufacturer after they found ants had shorted out the main circuit board, the reconditioned unit was returned to me.  As we were about to leave town for several days I hurriedly installed it to give it a test drive.

It would not even turn on.  I stewed about that for several days.

On our return after traveling I reached out to the manufacturer’s support tekkies and related the situation, with resolute firmness and precise language.  After a brief silence from them they sent me a shipping label and I sent the unit to them.  Again.

Two weeks later it arrived back with the cryptic note that a disconnected fitting had been connected.  So much for the assertion that the unit had been previously tested, don’t you think?

Again we were on the cusp of leaving town for a few days, but at dusk I installed the newly re-repaired unit.  At least this time it turned on!  There was great joy in Mudville.  That it would not perform any controller function was not especially disconcerting since the evening was fast approaching and the unit normally puts itself to sleep for the night once the photon levels get lower than net operating power.

Assuming that the system would wake up with the morning sunshine I left everything status quo and left town.

Big mistake, but then you know what assuming does.

On our return three days later I discovered that not only was the solar system not turned on and functioning well, the entire system had shut down because the batteries had been drained to the point where, well, the system turns itself off in order to protect the batteries from harm.  Now, this is not a cluster of AA batteries.  These are four monster huge batteries, each weighing about 150 pounds.  Something was amiss.

Side note – when the troubles first began I tested the circuit from the solar panels to the control input terminals and noted the voltage.  It was fine (~100 volts at solar noon on a clear day).  In the follow up testing I was finding voltage variations not unusual for solar systems given that the voltage output varies with the intensity of the sunlight.   Just keep that in mind for future reference.

I contacted the tech weasels again, and spoke to them with increased fervor.  I was given a series of diagnostic exercises which I executed while I insisted the tekkie remained on the line, waiting for me to walk to the system, make the test and return from the power house fifty yards away.  Zilch, zero, nada.  The unit would not perform its functions even though it powered itself up.  The input voltage numbers were a bit low (~60), but certainly enough to jolt the system to action.  But it was not responding no matter what the tekkie told me to try.

“Would you like to return it to us for an further evaluation?” I was asked.

“What I want is for you to send me a unit that actually works,” I replied.

After a few minutes on hold I was told that a new unit was being sent to replace the old one.  It arrived a week later, smack dab in the middle of the ripple molding soiree (we had been using the hydro power and gas generator for that).  Anxiously I installed the unit at high noon on a brilliant sunny day, checking and double checking my wiring connections.  I threw the switches in anticipation of, something.

The unit turned on but refused to perform its function.

To say I was disappointed is to gloss over my mindset.  I took a couple hours to gather my thoughts, called the engineer who had helped design and install the system, dismantled the breaker box and re-took the circuit readings.  But something weird was happening with the readings.  They varied wildly and continued to drop regardless of the sun shine.  40 volts.  32 volts.  26 volts.  50 volts.  20 volts.  48 volts.  24 volts.  39 volts. 22 volts.

Second side note – the buried cable from the solar panel array to the power house was a type specifically certified for direct burial, no conduit required.

Third side note — when digging the trench for the cable with a rented trencher, the trencher broke in less than a minute due to the rocky soil.

With my friend Brint’s help we took some cable and bypassed the buried cable to connect the solar array bus on the side of the cabin directly to the power controller.

It read ~90-100 volts.

Something, somehow, the circuit had been breached, and through trial and error we determined that it was somewhere in the 75-foot buried section, not in the open cable that was suspended underneath the bridge over the creek.  We grafted in the new cable to replace the buried cable, this time enclosing all of it entirely in conduit sealed from the fuse bus to the power house.  This will be buried as time and weather permit.  The new circuit worked perfectly and at solar noon the next day the panels were cranking out over 1300 watts, pretty astounding given that it was September and the theoretical capacity rating for the panel array is 1410 watts.

So now I have a fully functioning doubly redundant power system for the barn; hydro turbine, solar panels, and gas generator.  As a friend once quipped, “The problem with being your own power company is that you are your own power company.”  Every part of it  must be maintained and attended to, but through it all my appreciation for the aggregate utility grid is immense.  Although this has been a supremely frustrating episode I find that my understanding of every part of my system has been enhanced immeasurably.

Last side note — in retrospective contemplations we have arrived at the un-provable conclusion that somewhere in the original underground cable a sharp rock had encountered the cable and through essentially micro-seismic vibrations had eventually breached the sheath of that cable.  Not enough to cut the circuit entirely, but enough to ground it, the amount of the grounding discharge probably dependent on temperature and soil moisture.  As I said that is un-provable but does explain a lot; varying voltage, draining the battery bank, failure to wake up, etc.