I don't see how voltage and phase come into the picture since all you are trying to do is to reduce current to the meter to a minimum. At worst, additional current probes will be required. Who knows, maybe there are issues here I don't yet see/understand.

Interesting figures re efficiency - I didn't know that efficiencies could get that low. I keep hearing the efficiencies of inverters being around 95%. Sure the extra electronics will take some power but I would be surprised if efficiencies go below 90% (based on full load of course).
Hmmm, thinking about it, the efficiency figure that I would worry about would be only the portion after the generator and before the utility meter. Everything else doesn't really count in my case of very low cost fuel.

This system really isn't as complicated as you think it is. A small processor collecting all the data points, a software PID loop and an output to the SCR's. The only thing that makes it a bit more complex is the high power stuff - it's usually a bit tricky until all the bugs (due to transients) are worked out and the smoke has settled down.

Jens

Because you can't tell if you are using or generating 10 amps into your utility feed... Say, for the sake of argument, you are currently generating 40 amps, and you know for sure that the meter is showing you consuming 10 amps from the utility service.

Then, you turn of the A/C (or electric heater, or irrigation pump, etc.) that was consuming 20 amps... Bang, you are now pumping 10 amps out the meter instead... And your control circuit is still happy.

The only way you could catch this problem with a non-directional current clamp would be to dither the generator output... Say drop its output by 5 amps... If the meter current goes up, then you are a net consumer, if the meter current goes down, then, you are net generator...

If the meter current stays zero then you were a net generator by 2.5 amps (unless something turned on/off during this period)...

But, in the end, if you are using non-directional current clamps--then you could measure separate currents--one for your loads, the other for your generator... Assuming the loads are always consumption, and the generator is always a source, then a simple difference between the two probes would give you the answer (assuming all is calibrated, nothing is broke, etc.).

Will still be the equivalent of GT and need permits. In the end, I don't know of any GT inverter/interface that supports real-time energy output control. Not that it could not be done (albeit with a bit more complexity), but that a person would still need all of the permits and approvals of a "real" GT system--so it does not make too much sense.

Regarding efficiencies, the 77% is based on Solar PV panel name plate wattage and system losses... Probably 10-15% of that number is simply the way STC (standard test conditions) are defined for solar PV which over estimates output (of panels in typical installations). The rest are based on controller, wiring, and dust on panel type losses. Dropping from 77% to ~52-54% are the 20% losses with flooded cell batteries and typical inverter inefficiencies (80%*85%=68% overall efficiency as an example).

Lastly, just controlling pulse width to the SCRs (if that is your plan)--is no longer acceptable for AC attached equipment in the US and Europe... Doing this changes the pure sine wave form into a "broken" AC sine wave which ruins power factor... I believe that the sine wave must have 5% or less harmonics for a GT attached inverter/device.

-Bill

As you say, when I mentioned 'at worst, additional current probes are required' I was referring to seperate current probes for stuff known to be a pure load and stuff known to be a pure source. Probes at the meter could be used as a safety double check.

Yes, contemplating this some more, it seems that the utility company probably won't be all that happy with this sort of arrangement and could claim that since there is the potential for feeding the grid, a net metering contract will be required

Re the controllling SCR pulse width - the SCR's would be between the generator and the Sunny Boy (call it a fancy lamp dimmer). The 5% harmonics you refer to is what the Sunny Boy needs to put out to the grid and that is taken care off by the SB itself.

Well, I think I need to think this over some more. As you say, the utility company might look dimly on this and require a net metering contract in which case this is a lot of effort for nothing. I am also starting to get a bit concerned about our economy at this time so the entire thing will likely be shelved for a year or two and who knows what equipment will be out by then.

Thank for your participation in this discussion.

Regards
Jens

OK, I'm definitely not an expert here. And I'll admit I don't understand much of this. But, from what it sounds like, there is a diesel engine that you're able to run, basically for free, in order to get heat for the house. You could also be getting electricity from the diesel, thus reducing the electric bill and also making the diesel run better because of the load on it. In order to put the load on it you need the inverter / battery setup, but you can't just run the AC from the generator because it isn't stable enough.

So what I don't understand is why have some complex setup connected to the grid when you don't really want to be connected to the grid? Why not have the inverter feed one or two electrical lines with a refrigerator / freezer attached to it and maybe a TV also? Then while the generator is running all the time in the winter you are able to reduce the electric bill for items which consume a noticable amount of electricity, and would also put the load on the generator. If you don't need to run the generator all the time the refrigerator would still be OK for many hours until the generator started again, in fact it might work out better because then the refrigerator would be putting the load on the generator immediately.

The water heater consumes the most electricity from what I can see, about 0.5 kw / hour, but I think it might draw too much power for the inverter to supply, but I'm not sure. However, if you had it so only one element in the water heater worked it might be OK that way.

sub3marathonman:
First paragraph: Yes, that is correct.
Second paragraph: I made several attempts at explaining myself but nothing really explains things properly. I would suggest that you email me directly (jensatpacificsundotca) if you want to discuss things. For now, the entire project is on hold due to financial issues.

Jens