[RE-wrenches] To Megger or not to Megger

Mon Apr 27 20:08:28 PDT 2009

Hi Keith,

Both hands up here. From day 1 in PV.

Every single field-installed current carrying conductor to the point of
termination. AC & DC. BEFORE energizing them! One megger lead on the
conductor, the other on the raceway or ground-wire. Megger @ 1kV. Must be
greater than 250 Megohms to pass. Record the values on your commissioning
sheet.

I only meggered the actual arrays on projects where it was a requirement and
I only did that following specific procedures provided by the module
manufacturer. No procedure from the mfr = No array megger.

I believe it's a critical thing to do as part of the commissioning. But then
again, I don't like ground faults or fires.

I'll never forget a couple of head scratchers that the megger found.... One
was a perfectly fine, 6' long piece of Orange #12 XHHW-2 that would only
pull 40 Megohms installed in EMT. Pulled it out and replaced it. Inspected
it for any visible flaws. None found. Not a scratch or scuff. Used it as a
training tool for my guys. Hung it on the wall over my desk as a handy
reminder. If anybody wants a picture...

The other was interesting, too. The very first PV system I ever installed.
Was shooting the DC circuits from the inverter all the way to the connector
on the output jumper cables at the modules. Megger showed dead short on one
wire. Fluke 79 on ohms showed clear. Hit it with the megger again. Dead
short. I'm known for being anal about conduit and wire installation, so this
was quite unexpected, not to mention a little embarrassing. This happened to
be a Solarex Millennia Integra system with an Omnion inverter and the array
happened to be scattered all over this roof. For those of you who who have
dealt with that situation, you'll understand what a cluster-*$#! that is...
Especially being my very first PV system and all... I wasn't anywhere near
confident that I had a clue by that point. All ready to see my first install
fire up and this happens.

It was my first install and I had 3 people getting paid prevailing wages to
learn... I was learning too, but I wasn't making anywhere near prevailing
wages, let me tell ya. Well, it was early January, late in the day and foggy
as foggy gets, but we went back onto the roof anyway. Yeah... Overtime for
everybody but me. We opened up the condulets and J-boxes for inspection.
Pulled all the wires up out of the J-boxes. Pulled the endcaps and ground
clips off (unique components of the Integra product). Couldn't find anything
suspicious. Shot it with the megger again, and it was all clear. But no sign
of anything that could have been shorted. We put everything back together
and shot it again. Dead short! Aiyeeeee!

Call it a day. Go home. I couldn't get back to that site until about the
same time the next day. Foggy again. Only took 1 helper with me this time.
We completely repeated the entire exercise, with exactly the same results.
By now, I was pretty sure this solar stuff wasn't my calling. I've always
been regarded as a gifted trouble-shooter and all I could say at this point
was that I didn't know _____! All signs pointed to it being associated with
one of the J-boxes or the endcaps or the ground clips, but there was NOT a
thing we could see. Oh, well. I would have to come back another day. When I
went back, I took the same helper and made sure I had all day if necessary
to fix the problem. No matter what it turned out to be. We planned to take
it apart 1 piece at a time and re-shoot it with the megger after each step.
I left my helper on the ground with the megger while I took it apart on the
roof, 1 screw at a time. Again.

Well, lo and behold, he hollers up that "It's clear" when I pulled the cover
off one of the 2-gang bell boxes. Hadn't moved a wire. Just took the cover
off. Let me tell ya, I stuck my nose and four eyes in that box real close,
but didn't touch any wires. The box was a tight fit with those big,
sealant-filled blue wire nuts that came with the Solarex Kit, but I still
couldn't see anything wierd.

"Shoot it again."

"All clear, boss."

I put the cover back on and installed all the screws. "Shoot it again."

"Dead short." We had a talk later about shouting things like "Dead short" on
the jobsite, but I won't get into all that right now.

I took out one of the cover screws. "Shoot it again." Same answer from the
ground.

I took out a second screw. "Shoot it again." This time the reply was
different. I took out the third screw and we repeated the process. Still
clear. And the fourth. Still clear. WTF? I lifted the cover for what seemed
like the hundredth time. I studied the nested wires, all neatly and
systematically put into what I had thought would be their final resting
place... Days before and time and time again at this point. 

I began gently lifting the wires out so they all stood up like dandelions in
a spring lawn. I inspected them for the nth time. Looking for a stray strand
outside its protective hat. Nope. Overtightened wire nut with the spring or
a conductor coming out the tip? Nope. But what's this little, nearly
imperceptible dent in the side of this wire nut? As I looked closer, I could
see that it was round, flat bottomed and about the same diameter as a 6-32
screw. Hmmmm. I rolled the wires back into their resting position. Uh, huh.
This particular wire nut's natural home was directly under the cover screw
which had cleared the fault when removed and the dent lined up perfectly.
The screw could just barely be long enough to pierce the plastic and contact
the wire-spring. 

I replaced the wire nut, repositioned the conductors to make sure we
wouldn't have a repeat, and buttoned up the J-box. I put the wirenut in my
pocket for later reference. One more megger shot proved the system to be all
clear. We could now energize the DC and do our Voc & Isc testing prior to
startup. Thank goodness the sun was shining. I wasn't happy about this whole
ordeal, but it was a good thing we caught it with the megger before we put
the power to it.

That wire nut became the first exhibit in what became my collection of
training materials for the "what can go wrong and why we do these
procedures" trainings. Verification that cover and mounting screws cannot
come in contact with energized equipment was officially put in the
checklist. All cover & mounting screws in all types of enclosures. There
would be no exceptions. There would be no excuses. I actually had to write
one guy up for a violation of this at one point. Good thing we caught that
one with the megger before putting the power to it.

I would posit that at least one of the recent PV fires would absolutely have
been prevented if the installer followed the procedure described above. Too
bad most don't. Kudos to those of you who do. It's only a matter of time
until more and more of these AVOIDABLE problems surface.

If you don't have a megger, get one. If you have one, use it! The Fluke 1520
is a nice unit. I recommend it over analog models. I cut my teeth with
analog gear, but really like my 1520. A lot! Rugged instrument that gives
you an actual number to write on your commissioning sheet. In addition to
Megohms, it also does VAC, Lo Ohms, Continuity, has a display backlight,
Lock and Zero functions and my favorite.... Battery Check!  Reads out an
actual % value for your battery condition. Nice! Uses 4 "C" cells.

I hope more hands go up on this topic. Thanks for asking, Keith.

Peace and Palm Trees everybody,

Matt Lafferty

  _____  

From: re-wrenches-bounces at lists.re-wrenches.org
[mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Keith Cronin
Sent: Monday, April 27, 2009 2:37 PM
To: RE-Wrenches
Subject: [RE-wrenches] To Megger or not to Megger

Hi

I was wondering, by a show of hands, how many of you megger every project?

Do you have a cut off- like if it is "x" sized system, you will or decide to
opt out of performing this task?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20090427/0476f21c/attachment-0003.html>