A real stumper--Grid-tied system no longer grid-ties [RE-wrenches]

Mon Feb 21 14:51:19 PST 2005

Your free subscription is supported by today's sponsor:
-------------------------------------------------------------------
Your opinion counts! We're giving away ten $100 cash prizes!
A major food company wants to know what you think about a product 
& package design. Hurry, only ten $100 prizes left!
http://click.topica.com/caadbR9bz8Qcsbz9JC9a/Val Rad Survey
-------------------------------------------------------------------

Tom/ Wrenches:

The Fac_Bfr error is technically a frequency error meaning the inverter
is not seeing the proper frequency.
We all know the likelyhood of the grid having real frequency problems
is close to zero so it is an indicator of something else.

Since it is an 1800 I would look for a loose neutral, neutral bonding
jumper, or possibly a loose neutral connection behind the meter (bummer,
I know).
The neutral connections are often neglected and can loosen over time.
As the power comes up you can get get a small arc in the loose
connection and the inverter will interpret this as a high frequency
grid.
It will work intermittently until the carbon build up makes the
resistance of the connection too high then it will not run at all.  

I once has a very similar preplexing problem in Davis.
After several unsuccessful attempts to solve the problem remotely I
went down to the site to look myself.
Everything would be measuring fine with my DVM then the inverters would
randomly trip offline with an Fac_Bfr error.
I was completely stumped until I happened to lean up against the
customer installed kWh meter used to measure production.
I heard a small "Bzzzt" sound and the inverters tripped ofline.
It was repeatable - lean on meter, Bzzzt sound, inverter error.
Ultimately, the customer determined that the culprit was a bad rivet
joint in the bus work of the meter socket.
The Bzzt sound was a small arcing fault that took the voltage to zero
until the zero crossing extinguished it.
The inverter correctly interpreted this as a short cycle, i.e. high
frequency, error and tripped offline.

Another cause for mystery Fac_Bfr errors is when you have 2 2500's at
one location (obviously not the case here).
Measuring L1 to ground = 120 Vac and  L2 to ground = 120 Vac so it all
seems right but the inverters give waiting or Fac_Bfr.
The problem is (2) L1's going to one inverter and (2) L2's going to the
other inverter.
Seems funny after the fact but not when you are in the thick of it.

I hope this helps, please keep me posted.

Best Regards,

John Berdner 

>>> tom at regrid.com 2/18/2005 05:59:25 PM >>>

Wrenches,

I have a real stumper and could use your collective wisdom and ideas.
Here's the situation:

-- A residential grid-tied system consisting of 12 Matrix 165's (12V)
in a
single string feeding a SMA 1800U.  PV voltage and string parameters
are
fully within spec for the inverter.  Typ. PV input voltage (mp) is
around
242V, so there appear to be no problems on the PV side of the system.

-- The system operated without problems or errors for 9 months.  Then,
suddenly it quit running with the 1800U display showing either
"waiting" or
"Fac-Bfr" errors indicating that it cannot connect to the grid.  It
doesn't
attempt to enter the "grid mon" phase.

-- The grid voltage measured at the inverter terminals is stable at
around
122V, completely within spec for the inverter.  A Fluke meter with
Min/Max
settings shows the grid voltage varying very little (between 122.3V
and
122.9V).  The meter shows the frequency locked on at 60.00Hz. and only
occasionally varying up to 60.01Hz. (again, via Min/Max).

-- We swapped out the inverter with a new 1800U, and the new inverter
displays exactly the same errors.  The old inverter that was sent back
to
SMA worked just fine on their bench, and they couldn't find anything
wrong
with it.

-- We replaced the circuit breaker inside the load center so we could
rule
that out as a variable.  The wire length from the inverter to the load
center is 3 ft., and the conductors are #12 solid copper as they should
be
for the inverter's output current.  The voltages measured inside the
load
center are identical to those measured inside the inverter.  In short,
the
grid seems completely stable and within its normal voltage and
frequency
parameters.

-- The system is properly grounded via a driven ground rod, and DC and
AC
grounds are bonded together.  The neutral and ground bond inside the
load
center is in place and solid.  All wire connections have been checked
and
are tight.

-- A solar system on a house two doors down from this system is
functioning
just fine on the same utility grid power, which would appear to
indicate
that the grid is stable and there are no micro-spikes on the line.

-- The same errors occur regardless of whether or not we turn off
every
breaker in the load center (except the solar breaker), or the
homeowner
turns on every load in the house (in other words, whether or not there
is a
load through the meter).

In short, everything about this system seems completely normal and
according
to Hoyle, and yet two different 1800 inverters that work fine at
another
location still cannot connect to the grid.  We have replaced every
variable
in the system except the wiring itself.

Have any of you gurus out there ever run into anything like this, and
what
ideas might you have to help us figure this one out?

Many thanks for any suggestions any of you may have.

Tom McCalmont
NABCEP-certified PV Installer
REgrid Power, Inc.
(408) 656-2944 (cell)
tom at regrid.com 

Your free subscription is supported by today's sponsor:
-------------------------------------------------------------------
2-Day Tax Refund Loans TaxBrain Online Tax Service Get Started 
Now!
http://click.topica.com/caac8vhbz8Qcsbz9JC9f/TaxBrain
-------------------------------------------------------------------

- - - -
To send a message: RE-wrenches at topica.com

Archive of previous messages: http://lists.topica.com/lists/RE-wrenches/read

List rules & etiquette: www.mrsharkey.com/wrenches/etiquette.php

Check out participant bios: www.mrsharkey.com/wrenches/

Hosted by Home Power magazine

Moderator: michael.welch at homepower.com
--^----------------------------------------------------------------
This email was sent to: michael.welch at homepower.com

EASY UNSUBSCRIBE click here: http://topica.com/u/?bz8Qcs.bz9JC9.bWljaGFl
Or send an email to: RE-wrenches-unsubscribe at topica.com

For Topica's complete suite of email marketing solutions visit:
http://www.topica.com/?p=TEXFOOTER
--^----------------------------------------------------------------