In article ,
Tony Sivori wrote:

I recently acquired a small portable back up generator for nuisance
outages. Only paid $200 new. It is 2500 watts continuous, 3250 surge rated
and no 220 volt outlet, so I won't be hooking it up to a transfer switch.

Even though I'm going the extension cord method, I decided to properly
ground it, and use an inline GFI and surge protector. I'm willing to risk
using a back up computer with it, and I am hopeful that a surge protector
might clean up the "dirty" generator power.

Now the problem. I drove an 8 foot ground rod into typical soil, used a
correct copper clamp (nothing improvised), and connected the rod to the
generator ground lug with a single strand 12 gauge copper wire.

Trouble is, the generator outlets still tests as ungrounded.

I immediately suspected miswired outlets on the generator. So I tried
checking the resistance from the power receptacle outlet ground to both
the generator frame and the generator ground lug. Both tests resulted in
zero ohms of resistance. I'm thinking that would eliminate miswiring on
the generator as the cause.

I then removed both ends of the ground wire connection, checked and
reconnected them.

Same result, tests as open ground with a three prong plug in tester. I
also tested the tester, and it shows grounded outlets in my house as
properly grounded.

I checked the resistance from the generator end of the ground wire to the
ground rod itself. It was also zero ohms.

So it seems to be a case of it has to work, but it doesn't work. Any
suggestions are welcome.

It sounds as though the generator's neutral leg is not tied to ground at
the generator. The outlet tester isn't actually testing that the ground
prong is anywhere near ground, merely that current can flow between the
hot and the ground leg. (In a standard residential setup, the neutral
leg is tied to ground at the main breaker box, and all neutral and
ground wires go to the common neutral/ground bar. The outlet tester
would still show the outlet is wired properly even if the ground leg
were not actually connected to ground.)

By the way, the surge suppressor will do pretty much nothing to clean up
the generator output power. It's also rather likely that the output is
reasonably clean to begin with; certainly far better than a typical
"modified sine wave" inverter output, which is a combination of a couple
of square waves. A typical computer, with a switching power supply, is
not picky at all about power quality, as the first stage in the power
supply is a rectifier and reservoir capacitor, changing whatever the
input is into (unregulated) high voltage DC power. They run just fine
from the above mentioned modified sine wave inverters, potentially even
a bit more efficiently than from sinusoidal inputs.

Anyhow, you could tie the "neutral" leg of the generator output to
ground at the generator, which would be safe provided the generator
frame is actually grounded, and which would make the outlet tester
happier. Safer, IMHO, is just using the GFI and not worrying about it
more. Indeed, code permits GFIs to be used on existing ungrounded
circuits to allow three-pronged devices to be plugged in despite the
lack of a safety ground conductor (albeit with certain cautionary
labeling required).

--
Andrew Erickson

"He is no fool who gives what he cannot keep to gain that which he cannot
lose." -- Jim Elliot