Generac iX2000

[Robert Green]

<stuff snipped>

<The inverter type is a unit that varies voltage by rpm, it can run as
low as 900 rpm to power a light bulb.. Regular gens run at 3600 rpm,
its a big waste of evergy to power a small load and engine life
[depending on quality] is about 300-3000 hours, Inverter gens by honda
run with a few hundred watts are known to last 11000 hours. RPM
directly relates to engine life, this is the main reason a 1800 rpm
deisel lasts 4x as long as a 3600 gas engine, and 1100 rpm Lister
generators can go 100,000 hours, and deisel boat engines that maybe
run at 100 rpm run nearly forever. The Honda inverter also has the
Generator as part of the motor, it cuts about 30% of the weight out
making for a compact lightweight unit. So for long term low load at a
remote site such as a cabin, if the load is kept to a minimum you can
with an Inverter get 9-12000 hours life vs 3-350 for the cheap non OHV
B&S engine. Inverters 2.5x premium in cost can be worth it for someone
living for example on a boat or camper for long periods. Plus Hondas
inverters have cleaner power than your power co and are so quiet you
cant hear them from 10 ft away. Generac probably stole the design but
his batch looks to be infected with a defective part. I would not
exchange it for a unit of the same batch. Once I got a cheap Generac
power washer, it blew in 2 hrs, the replacement lasted 1 hr, it was
the same batch and had the same defect. Inverter technology is great,
but there are a few others on the market that have proven reliability
like Yamaha>

Thanks for the explanation. That Generac unit looked cute and portable
enough to use on camping trips. But since I've owned Honda cars since 1980
with great satisfaction, I think I'll see what they have to offer.

 

[Dan Lanciani]

| About two years back, I bought a cheap 2.5KVA petrol(gasoline)
| generator from my local Aldi store to supplement the 2KVA APC
| SmartUPS2000 that I use to provide a protected supply to my computer
| kit. After much testing, it turned out that it was basically useless for
| this purpose since even a modest 4.7 microfarad capacitor was sufficient
| to make the generator output jump from its regulated 230v to some 275
| volts.

I ran into this problem with a 9kW (derated to 6.5kW on natural gas)
generator and a couple of SmartUPS1000s. The generator also gave my
Sub-Zero refrigerator fits and would not run my cat's drinking fountain at
all (it just hummed). I was lucky enough to pick up an 8kVA MicroPower II
isolator/regulator from a locally decommissioned computer center and it
largely solved my problems. (I still have to set the UPS sensitivity one
level down, but that's not so bad.) As an added benefit, the isolator
draws only from the 240V output so the generator always sees a perfectly
balanced load. These and similar units are often available on eBay, but
shipping may be expensive...

Dan Lanciani
ddl@danlan.*com

 

[Dan Lanciani]

| On Jun 22, 2:22=A0pm, Johnny B Good <[email protected]>
| wrote:
| >...
|
| Have you played with a ferroresonant Constant Voltage Transformer amid
| this mish-mash? I'm curious if a UPS works with them but don't want to
| risk destroying anything.

One problem with ferroresonant transformers is that they are rather
inefficient (they get very hot). They also aren't a panacea. The
Best online UPS I mentioned earlier in this thread uses a ferroresonant
transformer and the flicker at transfer is terrible. (The MicroPower II
is not a ferroresonant transformer; it uses many taps on a conventional
transformer and selects the most appropriate one on a cycle-by-cycle
basis.)

Dan Lanciani
ddl@danlan.*com

 

[Tony]

By "stepped" you mean a square wave, no?

No, it's all choppy. Taking a bunch of little square waves to make up
one complete cycle. On a scope it looks like steps going up and then
down. The more steps the closer it gets to a sine wave. Not just one
up and one down like a square wave. It sounds good in theory, but
unless it's really a good one, it isn't worth the money.

I've run a few different computers and other electronic equipment off my
plain Jane generator (alternator) without any problems. No electronics,
just a cheap generator (alternator) at about 60 Hz. Most electronic
equipment is marked for 50 to 60 Hz showing that it isn't as sensitive
to frequency as one might think.

 

[Johnny B Good]

| About two years back, I bought a cheap 2.5KVA petrol(gasoline)
| generator from my local Aldi store to supplement the 2KVA APC
| SmartUPS2000 that I use to provide a protected supply to my computer
| kit. After much testing, it turned out that it was basically useless for
| this purpose since even a modest 4.7 microfarad capacitor was sufficient
| to make the generator output jump from its regulated 230v to some 275
| volts.

I ran into this problem with a 9kW (derated to 6.5kW on natural gas)
generator and a couple of SmartUPS1000s. The generator also gave my
Sub-Zero refrigerator fits and would not run my cat's drinking fountain at
all (it just hummed). I was lucky enough to pick up an 8kVA MicroPower II
isolator/regulator from a locally decommissioned computer center and it
largely solved my problems. (I still have to set the UPS sensitivity one
level down, but that's not so bad.) As an added benefit, the isolator
draws only from the 240V output so the generator always sees a perfectly
balanced load. These and similar units are often available on eBay, but
shipping may be expensive...

I've got a small transformer based stabilizer (aka, power vampire) but
it's only rated at 250VA. I tried it but it didn't seem to have any
effect. There used to be a couple of local "Government Surplus" shops
which would deal in this sort of "Junk", but they've long since gone out
of business. Unless the proprieters had family to carry on the business,
usually mutating into an electronics store, the business would die with
them.

Any such solution is not likely to be as effective on generator power
as it would be on PSU power due to the wider variations of supply
frequency since these utilise resonance and core saturation to function
properly.

The variations in frequency with load are usually within the tolerance
range of most UPS, provided you retrim the governor to centre the speed
at half load on the nominal 50 or 60Hz they're supposed to be set for
(they're typically set on the high side in the factory).

It's a sad fact of life that modest amounts of excess capacitive
loading will mess up the voltage regulation of cheap (and, not so cheap)
generators. The sine inverter types are immune to this effect and have
the advantage of allowing a permanent magnet alternator to be used at
variable revs to suit the actual demand. The extra losses in the
inverter usually being negated by the efficiency gains in the PM
generator (no 10% loss to the field windings and VR circuit).

Until I actually bought a cheap 'n' cheerful 2.5KVA (continuous, 2.8KVA
peak) generator, I was woefully ignorant of this tiny but oh so serious
deficiency when used with certain classes of electronic gear.

Regarding the SmartUPS2000, I guess it probably uses a near identical
schematic diagram to its SmartUPS1000 cousins, switching either a single
4.7 microfarad (or, in the SmartUPS 2000 case, a pair of them in
parallel[1]) across the line when in pass through standby mode,
disconnecting them when running off the battery.

Since the design incorporates a current transformer to monitor the
current flowing through this capacitor, disconnection is not an option
since it is very likely that the UPS will not function properly in their
absence.

What muddied the issue was the fact that I could only select between
high and mid sensitivity on the rear panel dip switch. The lowest
sensitivity setting is only available via the Powerchute management
software (or equivilent open source alternatives) and it turned out that
the interface firmware in my particular example has been broken in such
a way as to preclude this option (although I could interrogate some of
its operational parameters using the open source management software).

However, in view of the real problem being the extreme voltage boost
effect, perhaps it's just as well I couldn't persuade it to accept the
generator power. I have another SmartUPS which I was able to persuade
into accepting the generator voltage but this was only a SmartUPS700 so
may not have used anywhere near the same amount of capacitive loading
(if any).

I've not had any cause to examine a circuit diagram for the SmartUPS700
since the three sensitivity levels can be set with a rear panel
pushbutton switch and its software interface works perfectly ok, so I
don't know whether they even use a similar capacitive circuit or how
much capaciticance even if it is the same circuit (but I'd expect a
proportionally smaller value if it was).

The cheap generators designed for european/uk use tend to use a two
pole 3000rpm (50Hz) generator head with split stator windings to allow
either 115v or 230v operation. They don't have an option to supply both
voltages at the same time since there isn't such a need outside of North
America.

[1] I'm guessing that you're located un the US of A so it's quite likely
that those 115v versions of SmartUPS1000s might[2] _still_ need a pair
of 4.7microfarad caps (a 115v version of the SmartUPS2000 would need to
almost double up on the two caps already fitted in the 230/240v versions
all else being equal[2]).

[2] Possibly, the capacitor value might not be too critical since the
frequency is higher and the monitoring software can simply be
recalibrated to compensate for the different
voltage/frequency/capacitance ratio that would result if they were
simply sticking to a common capacitor value to simplify their parts
inventory.

I could probably check the circuit diagrams if I could be arsed enough
to search them out. If anyone else wants to research this, they're free
to google for a source of schematics (they're out there, or they were
when I last looked).

 

[Johnny B Good]

The message <[email protected]>

====snip====

holy crap.

When I moved back up the creek full off grid back in 2k or so My bro gave
me a UPS to help with the computers .. thinking I could stay alive while
switching from solar/batteries to generator (this is before my hydro) --
the damn UPS wouldn't do just that using my old generator. Switch to
charge, kill the generator, switch back to UPS .. the generator would
pick back up, then it'd go back to charge and kill it again in a very
distrubing loop with everything sounding like it was going to blow up.

Now I have an explanation!!!

Well, only if I interpret the "switch to charge" as meaning the UPS has
switched from battery power back to line power to power the load from
the generator and, incidently, recharge its battery pack. If I've
interpreted this wrongly, then no, you don't have an explanation.

I just had my bro return it and exchanged it for a nice video card at the
time.

I'm wondering if now that i'm on the inverter type gens I could actually
have a UPS again. Might have to see if I can pick one up.

At the time I thought the UPS charging was drawing more amps than my gen
was able to produce with the other loads (though at the time it didn't
seem like too much), so it would overload the generator. But clearly
there may have been another reason for the evil power UPS+ Gen loop.

The only type of UPS that would do that would be worth at least 10 to
20 of the most expensive, latest, all singing, all dancing top of the
range graphics adapters, say £5k($7 or 8K?) and up.

The usual line interactive types don't present a heavy battery charging
load when line power is restored after an outage. My SmartUPS2000 only
draws a maximum of 190 watts to charge the battery after an extended
outage (a test run, usually) and this normally drops to less than 140W
after the first hour or so if the battery has been tested to exhaustion.

If that UPS were of the expensive kind, it could represent a heavy
battery charging load after a relatively short outage since it would be
float charging its battery pack _and_ powering the constant run inverter
feeding the protected load which would only ever see inverter power
regardless of the state of the utility power.

If such a UPS has a load rating of 2KVA it's very likely that its
charger will be capable of drawing 2.5 to 3 KW so as to be able to
provide a net charge to the battery when power returns when under
maximum load. When the protected load is only half the maximum, the
charging demand after a few minutes outage could very well still peak to
the same maximum for a minute or two before settling back to provide a
more sedate topping up charge rate plus load demand.

To reiterate, my situation was this:

I'd fire up the generator, disconnect the UPS from the PSU power which
made it transfer to battery power. I would then connect it to the
generator and the UPS would monitor the line power, see a supply within
frequency and voltage tolerance, wait for its own battery generated AC
power to get in sync and then switch its load (and those damned
capacitors!) back to line power which immediately caused the genny
voltage to jump to 270 volts which, in turn, caused the UPS to swiftly
transfer back to battery power to protect the load from the overvolt
event. The genny voltage would then return to normal and the cycle would
repeat.

If this describes the situation you had with that UPS and your
generator, then a suitably rated sinewave inverter model should solve
the problem.

HTH

 

[Dan Lanciani]

| The message <[email protected]>
| from ddl@danlan.*com (Dan Lanciani) contains these words:
|
| > In article <[email protected]>,
| > [email protected] (Johnny B Good) writes:
|
| > | About two years back, I bought a cheap 2.5KVA petrol(gasoline)
| > | generator from my local Aldi store to supplement the 2KVA APC
| > | SmartUPS2000 that I use to provide a protected supply to my computer
| > | kit. After much testing, it turned out that it was basically useless for
| > | this purpose since even a modest 4.7 microfarad capacitor was sufficient
| > | to make the generator output jump from its regulated 230v to some 275
| > | volts.
|
| > I ran into this problem with a 9kW (derated to 6.5kW on natural gas)
| > generator and a couple of SmartUPS1000s. The generator also gave my
| > Sub-Zero refrigerator fits and would not run my cat's drinking fountain at
| > all (it just hummed). I was lucky enough to pick up an 8kVA MicroPower II
| > isolator/regulator from a locally decommissioned computer center and it
| > largely solved my problems. (I still have to set the UPS sensitivity one
| > level down, but that's not so bad.) As an added benefit, the isolator
| > draws only from the 240V output so the generator always sees a perfectly
| > balanced load. These and similar units are often available on eBay, but
| > shipping may be expensive...
|
| I've got a small transformer based stabilizer (aka, power vampire) but
| it's only rated at 250VA. I tried it but it didn't seem to have any
| effect. There used to be a couple of local "Government Surplus" shops
| which would deal in this sort of "Junk", but they've long since gone out
| of business. Unless the proprieters had family to carry on the business,
| usually mutating into an electronics store, the business would die with
| them.
|
| Any such solution is not likely to be as effective on generator power
| as it would be on PSU power due to the wider variations of supply
| frequency since these utilise resonance and core saturation to function
| properly.

The MicroPower II (and similar) is not a ferroresonant "power vampire."
It uses multiple taps on a conventional transformer and selects the
most appropriate one on a cycle-by-cycle basis. I also have a 3.1kVA
ferroresonant transformer in the form of a Best UPS, but I don't find
it particularly useful. In any case, the MicroPower II works great
with my generator and solves the SmartUPS problem.

| Regarding the SmartUPS2000, I guess it probably uses a near identical
| schematic diagram to its SmartUPS1000 cousins, switching either a single
| 4.7 microfarad (or, in the SmartUPS 2000 case, a pair of them in
| parallel[1]) across the line when in pass through standby mode,
| disconnecting them when running off the battery.

The 1000 is probably closer in design to the 700, but it does have the
capacitor. It kills X10 signals so I have to run it throught a filter.

| What muddied the issue was the fact that I could only select between
| high and mid sensitivity on the rear panel dip switch.

The 1000 has the push button like the 700 to cycle through three settings.

| The lowest
| sensitivity setting is only available via the Powerchute management
| software (or equivilent open source alternatives) and it turned out that
| the interface firmware in my particular example has been broken in such
| a way as to preclude this option (although I could interrogate some of
| its operational parameters using the open source management software).

I was never able to get any of the free software to work so I wrote my
own trivial program to handle shutdown. For setting parameters I use
a terminal program and do it manually. There is great variation in
the semantics of the command set making reliable control a bit tricky.
Even the SU1000 and SUA1000 differ in, e.g, the operation of the "@"
command.

Dan Lanciani
ddl@danlan.*com

 

[Johnny B Good]

[email protected] (Johnny B Good) writes:
====snip====

| I've got a small transformer based stabilizer (aka, power vampire) but
| it's only rated at 250VA. I tried it but it didn't seem to have any
| effect. There used to be a couple of local "Government Surplus" shops
| which would deal in this sort of "Junk", but they've long since gone out
| of business. Unless the proprieters had family to carry on the business,
| usually mutating into an electronics store, the business would die with
| them.
|
| Any such solution is not likely to be as effective on generator power
| as it would be on PSU power due to the wider variations of supply
| frequency since these utilise resonance and core saturation to function
| properly.

The MicroPower II (and similar) is not a ferroresonant "power vampire."
It uses multiple taps on a conventional transformer and selects the
most appropriate one on a cycle-by-cycle basis. I also have a 3.1kVA
ferroresonant transformer in the form of a Best UPS, but I don't find
it particularly useful. In any case, the MicroPower II works great
with my generator and solves the SmartUPS problem.

My bad, I did see that description, but only after I'd sent my reply. I
wish I could get my hands on something like that, but only at its scrap
value price point.

| Regarding the SmartUPS2000, I guess it probably uses a near identical
| schematic diagram to its SmartUPS1000 cousins, switching either a single
| 4.7 microfarad (or, in the SmartUPS 2000 case, a pair of them in
| parallel[1]) across the line when in pass through standby mode,
| disconnecting them when running off the battery.

The 1000 is probably closer in design to the 700, but it does have the
capacitor. It kills X10 signals so I have to run it throught a filter.

| What muddied the issue was the fact that I could only select between
| high and mid sensitivity on the rear panel dip switch.

The 1000 has the push button like the 700 to cycle through three settings.

In which case, it's rather a worry that the 700 could have been
producing a similar, if less extreme overvolting effect.

| The lowest
| sensitivity setting is only available via the Powerchute management
| software (or equivilent open source alternatives) and it turned out that
| the interface firmware in my particular example has been broken in such
| a way as to preclude this option (although I could interrogate some of
| its operational parameters using the open source management software).

I was never able to get any of the free software to work so I wrote my
own trivial program to handle shutdown. For setting parameters I use
a terminal program and do it manually. There is great variation in
the semantics of the command set making reliable control a bit tricky.
Even the SU1000 and SUA1000 differ in, e.g, the operation of the "@"
command.

The only motivation I had for tracking down a copy of APC's management
software that would recognise that ancient SmartUPS2000 (and subsequent
searches for open source software) was simply to reprogram it to its
least sensitivity setting. In the end, this proved a total washout since
it looks as though a previous user may have screwed the serial interface
control firmware, which may explain why it found its way onto a mobile
radioham rally trader's table about 6 or 7 years back where I picked it
up, sans battery box, for £35 (about $50).

Apart from the mystery of the broken interface, it has performed its
primary function perfectly since then, once I'd set it up with an
external 48v battery pack.

The only benefit of my efforts to make it compatable with that cheap
generator was that it developed a show stopping fault which had started
as an intermittent failure to pass line power through to the load. The
benefit being that I had, by then, acquired the circuit schematics and
given them some study so was able to quickly identify the source of the
trouble which turned out to be the mains transfer relay itself (rather
than some obscure fault within the complex control electronics).

One of the base pins had cracked just within the relay base and gone
open circuit. The pin in question being one of the relay contacts. I
knew that trying to make a solder butt joint was going to be doomed to
certain failure but careful drilling with a 0.4mm dia drill alongside
the contact wire in the relay base allowed me to thread a short piece of
copper wire to act as a replacemt soldering pin that I could positively
anchor and make a much more durable solder joint onto the remains of the
original contact pin within the base.

That repair was made about 12 months ago and I've not had any other
problems since so it looks like I've thwarted Sod's Law of unwarranted
premature failure of high tech equipment by a low tech fault. If I
hadn't already invested my time in studying the schematics, it would
probably have ended up as landfill by now. As the more accurate[1]
saying goes: "All's well that ends." ;-)

[1] I find that the second "well" in the more common version of that
saying is often simply superfluous wordage in most cases.