From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Sinemaster Inverter Generator:  Anyone have one?
Date: Tue, 15 Mar 2005 22:20:36 -0500
Message-ID: <d49f3190r5d9d07ghpp7mj0us8qg0m887u@4ax.com>

On Sun, 13 Mar 2005 22:13:34 -0800, "Dave Thompson"
<dave.usenet@bdthompson.net> wrote:


>No.  There are a couple of ways to parallel AC generators.  It's either done
>with a control unit to manage governors or sense circuitry to determine to
>amperage load on each generator and adjusting power output accordingly.  In
>addition, the generators (inverters in the Honda case) also have to adjust
>speeds so the sine wave outputs are synchronized.

Nope, not at all with an inverter generator.  Each generator runs at
whatever speed it must to supply its share of the load.  The Honda
inverter is designed to look at the output when coming online.  If
there is already voltage present, the inverter syncs with it.
Otherwise it starts up on its own.


>Improperly balanced AC generators running in parallel can have one actually
>turn the slower/weaker machine into a motor with nasty results.

Again, irrelevant to the Honda inverter generator.  For any inverter
generator not designed for paralleling, the most likely result of
slamming them together out of phase will be the destruction of the
output semiconductors.  Since the output frequency does NOT change
with speed and because an inverter cannot be "pulled" into sync unless
it is designed that way, there is no practical way to synchronize two
inverter generators not designed for syncing.

I have, in fact, run an inverter and a regular generator in parallel
for a little while.  The inverter is the frequency master and the
regular one is governed to match phase.  With any buildup of governor
error, such as significantly changing the load, the inverter quickly
became unhappy and tripped.  IOW, not something that's actually
practical to do.

John

From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Sinemaster Inverter Generator:  Anyone have one?
Date: Wed, 16 Mar 2005 03:18:42 -0500
Message-ID: <m9qf31hqvkc5791e8b4845fo225n13gcp5@4ax.com>

On Tue, 15 Mar 2005 22:36:17 -0800, altar wrote:

>On Tue, 15 Mar 2005 22:30:32 -0500, Neon John
><johngdDONTYOUDARE@bellsouth.net> wrote:
>
>>BTW, I don't see any reason at all why an EU2000 and an EU1000 cannot
>>be run in parallel.  Each should share the load proportionally.
>>
>>John
>
>Now THAT is something I would have an interest in. Wonder how a person
>would go about that.
>
>Tom

If I had one of each sitting here, this would be my approach.

I'd make up a jumper cord with a light bulb socket in series with each
lead.  I'd install the jumper and start off with small bulbs, say, 100
watts.  I'd crank the gennys and apply a load to one.  If the
generators synced properly the bulbs should burn at a steady and equal
but dim level.  The steady and equal part is the important part.

If that worked I'd shut 'em down and replace the bulbs with, say, 300
watt bulbs.  Duplicate the test.  Depending on the load (which should
be near full load for the generator it is attached to), the bulbs will
burn dimmer but still equal and steady.

If that worked, I'd remove the bulbs and replace them with 1000 watt
screw-in heater elements.  Repeat the test.  At this point if the
heaters get about equally hot and both generators stay online then I
would conclude that it is safe to parallel them.

The operating principle here is to add sufficient resistance to the
paralleling cable that if the generators fail to sync, the circulating
current will be less than full load; i.e, less than what it takes to
let the blue smoke out.

In reality, I would expect one of two possible results.  That it would
work properly or that one or both would trip more or less instantly
when the second one comes online.  I would not expect them to
oscillate or display unequal brightness which would indicate ground
current.

I'm having a hard time imagining a scenario where the two would NOT
sync.  I'm less sure that they'd split the load proportionally.  I
suspect that they would but that would require testing.  At worst, the
smaller one would simply trip on overload.