Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Solar panel circuit



On 06/01/2010 09:27 PM, scitch@verizon.net wrote:

I connected the solar cell (12 V 75W) to a gel cell car battery, the car
battery to an inverter, and the inverter to a power strip. The power strip
is complaining (via a red LED) that the circuit is not "grounded" as verified
by my GFC tester. Of course it is correct, the circuit was completely
isolated from ground.

So far so good.

So, I "grounded" the frame of the solar cell to the metal surrounding the
circuit breaker box

There's no reason why the frame should be part of the
circuit, and evidently it is not, which is consistent
with the observation that "grounding" it doesn't do
anything.

It doesn't do anything good, but it doesn't do anything
bad, either.

and the negative pole of the inverter to the ground
junction on one of the garage lights.

By itself, that dramatically decreases the safety.
See below.

The power strip is still not happy.
Plugging the outlet tester directly into the inverter also shows that the
circuit is not "grounded". Grounding the case of the inverter to the house
ac ground does not solve the problem either.

Any ideas for how to ground this thing? I don't want anybody to get hurt.
Everything works fine when plugged into the system, but that red light on
the power strip is still on.

The idea of "grounding equals safety" is one of those ideas
that lots of folks take for granted, even though it is a
grievously flawed idea.
-- "grounding equals safety"
-- "dQ = T dS"
-- "significant figures"
-- "Lorentz contraction"
-- "filled Lewis octets in molecules"
-- etc. etc. etc.

There are three possibilities for the amount of grounding:
a) no grounding at all; floating circuit;
this includes transformer isolation as well as the
optical isolation we are discussing today.
b) cold side (white wire) of inverter (or transformer) "grounded";
nothing else.
c) cold side (white wire) of inverter (or transformer) "grounded";
chassis of appliance "grounded" via "third prong" (green wire).

It is common knowledge -- and actually true -- that (c) is safer
than (b). However, that does not mean that more grounding is
better, and in fact (a) is vastly safer than either (b) or (c)
for indoor applications. Outdoors there are lightning protection
issues that I don't want to get into at the moment.

In a floating circuit, it would take TWO faults before anybody
could get into trouble. See diagram below.

======

As a related matter, the plug-in type "circuit tester" does not
actually check (and could not possibly check) whether the circuit
is "grounded" in the sense of being connected to mother earth.
How could it? All it does is check to see whether there is a
low-impedance path from the cold side of the line to the "third
prong" (green wire). The red light on the outlet strip undoubtedly
is making the same check. It is a third-wire check; calling it
a "grounding" check is a gross misnomer.

So, if you just want to make the red light go away, connect the
cold side (white wire) of the inverter to the "third prong" (green
wire). This makes the warning go away, even though the circuit
is still floating with respect to mother earth. This is the
safest configuration, if you are not worried about lightning.
This is the configuration used for e.g. EKG machines, which are
a rather demanding application.

If you want a configuration more like "standard" wiring, you can
connect a wire to the exact same point where you tied the white
wire to the green wire, and then connect the other end of that
wire to "ground" somewhere.

++++++++++++++++++++
. +
----CB------------------------------ black +
| . +
inv + +
| . +
------------------------------------ white +
| . +
----x-----------------------+ green +
/ + +
y ++++++++++++++++++++
/ | chassis
/ |
"ground"..............."ground"-----person


I'm not saying this configuration is optimal, but it is "standard"
and it evidently provides an "acceptable" level of safety, as
evidenced by the fact that millions of people accept it every
day.

As mentioned above, the worst possible thing would be to have
the green wire open at a point such as x in the diagram but
with the inverter still "grounded" via y. In normal usage we
expect a person to be touching "ground" and touching the chassis
at the same time. Now a single fault, namely the black wire
touching the chassis, will zap the person, with no protection
from the green wire.
-- if the circuit is open at x it must be open at y
(floating configuration, red light on)
-- it is better to have closed at x and open at y
(floating, no red light, single fault blows the CB)
-- if the circuit is "grounded" via y it must be "grounded" via x.
("standard" "grounded" configuration)

===============================

In general, always draw the circuit diagram.

Always draw the circuit diagram.

Students think that if they are smart they can skip drawing the
circuit diagram. They can't. Professionals always draw the
diagram.

Draw the diagram already.

Include the fault paths on the diagram (e.g. the person in my
diagram above).