Power came back a little after the 24 hour mark. But I have some numbers to crunch.
First, I got up right at 5:20am to check the tank level as this was the 8 hour mark. This is what I saw.
I'm going to call this about half a tank but willing to round. So half a tank would be 2.1 and I'm willing to call it 2.25 used for 8 hours. So the consumption spec, including the extra 2 minutes, would be about 0.28 gal/hr. So my prediction was close. I estimated I'd use between 2 and 3 gallons all night and was hoping it would be closer to 2. It seems to have been. How about average load?
So over 8 hours and about 5 minutes, it used 3.48 kWh (works out to about $0.47/kWh or "only" 2.12x grid power cost - so efficient). This works out to an average load of 431W. Wow, that's close to my estimate of 450W. This works out to a 6% load, average, for the 8 hour overnight run. I should mention that at the 7.5 hour mark Jamie got up and started turning on lights and stuff (i.e. well pump), plus the furnace came on. Overall though I'd say this is a best case overnight scenario because I don't think the furnace ran at all other than when our programmable thermostat made the heat turn on first thing in the morning (we set it way back at night). But this is the full house load MINUS the dehumidifier which I did turn off at the breaker before bed. A colder night may have required the furnace. I have two refrigerators. Take that for what it's worth.
By the way, this extrapolates to a 15 hour runtime if I ran this tank dry at this load. That's pretty crazy on only 4.2 gallons.
I let it run another hour because I got back in bed. Here are the final specs before power came on and I shut it down.
So 9 hours. Tank definitely below half at this point and this last hour was definitely some increased loads. Actually I can do that math. The average load between hour 8 and 9 was about 900W or twice the overnight load. Still low though (12-13% of total capacity).
Final comparison to the Honda then. The Honda would have used about 0.95 gallons over this 8 hour run or about 0.12 gal/hr. I used an estimated 0.28 gal/hr which works out to about 2.4x the Honda fuel consumption. Put it in simple terms, the Honda would use "about a gallon" to get through a night. My generator uses 2.25, or an additional 1.3 gallons. You know what? That is really not bad at all considering it is a full size 7000W machine AND puts out pure sine just like the Honda. It would be cool to have a small one of course and the Chinese ones are pretty affordable. But from a strictly overnight perspective, on a multi-day outage, this means I will use an extra 1.3 gallons/night from not having a small inverter generator. Given that I have up to 23.25 gallons on hand, I feel like this is not a huge imposition.
Using the 30% rule between my generator and Bill's (again I think this is higher at light loads, this is for a 50% load which neither of our situations reflect), I will save about 6.6 gallons/day compared to Bill using a 24 hour duty cycle. Although, since actual loads will be much less than 50% sustained, the actual savings will be less. At 25% load, my savings might be more like 4.8 gallons/day over his setup. But even then, that MORE than makes up the fuel savings of not having a small inverter gen and saving one full 5 gal gas can per day is easy to visualize. I'm pretty pleased with this.
One final comparison case. Let's say Bill bought that Honda EU2200i and used it for the 8 hour overnight period, and his other generator for the other 16. Using 0.6 gal/hr for 16 day time hours (this is very reasonable since he claimed about 0.53 gal/hr for an entire 17 hour period which included a overnight period), that's 9.6 gallons. Add another 0.95 for the Honda to get you through one night and that's a total of 10.55 gallons.
I'll estimate mine at 0.45 gal/hr for day time use since that's the one data point I have on that and over 16 hours that's 7.2 gallons. Then add another 2.25 for overnight and that's a total of 9.45 gallons. So that means I would STILL use 1.1 gallons less per day even if Bill bought a Honda EU2200i for overnight use. I think this is also a conservative estimate (because I think I use more power on the generator than Bill does so normalizing for actual power use, the difference is likely larger) but willing to run with it. Very interesting.
For what it's worth that's comparing about $2400 worth of generators to my $811 generator, which means I could replace my generator THREE TIMES just to break even, and it would still be more fuel efficient overall. Now you can see why I was so excited for this deal.