You're a glass is half empty kind of guy, aren't you?
Actually I'm an engineer with both nuclear power experience, practical experience in running water electrolysis machinery, and experience in solar array design for spacecraft.
What are you?
Let's start with the bomb. You know, the electrolysis machine that'll pull the hydrogen out of the water you put in. Got any idea why submariners call those things 'the bomb'? It's because incorrect operation can make them explode.
You think that just anyone can run his own bomb? You're talking about a nation of people who've made a running joke out of the complexity of setting the clock on their VCR suddenly being able to safely operate a sophisticated water purification and electrolysis gadget. ( and no, you can't run tap water in your bomb, the impurities get left behind and before long you're trying to separate hydrogen from sludge).
Let's talk about solar cells.
The solar constant is 1300 w/m2 squared times to cosine of your latitude. That doesn't take into account inclination of the earth's axis to the ecliptic, which introduces a 24 degree annual variance in solar angle. That's something spacecraft don't have to contend with but it's fairly negligible hit, anyway. Spacecraft power systems aren't affected by cloudy days, either. That's probably a 80% efficiency.
Night times make things difficult, too. Cut by one half.
Silicon crystal cells have an efficiency on the order of 25%. Higher efficiency cells are available. Higher efficiency cells cost even more and don't help cost effectiveness.
Naturally, spacecraft don't have to worry about dust, dirt, or streaks, either, but your house does. Let's give that a 90% efficiency hit.
Did I mention that the sun is only shining straight on the panel for a little time each day because, well, the earth rotates and the sun moves in the sky? So the peak production hours are, well, when most people aren't home. How convenient. So you have to toss another reduction factor on your power curve, to come up with an average output. For more than six hours a day the sun is more than 45 degrees off axis. Let's make that a 70% hit.
So one square meter of cells will produce, ummm,
1300 w/m x cos 45 (latitude) x 1/2 (it ain't daytime all day) x 0.25 (efficiency) x .90 (dirt) x .80 (clouds) x .70 (sun moves)
equals about 60 watts of DC power per square meter of collector surface.
The average electric bill is about 1200 kw/hr per month, or 40 kw/hr per day. To collect this energy during the peak six hours in the middle of the day would require 110 square meters of cells, which is a fair sized roof.
In other words, you need a lot of collector area, and oh, yeah, a LOT of batteries to save that power for when you're home to use it. You know, environmentally sound lead acid batteries or maybe nice nickel-cadmiums, or any of the more exotic things, right there in your house with you. And the charging and battery maintenance gear and the static inverter. And most electric gear more complicated than toasters and lamps aren't as fond of the taste of the squarish wave of statically inverted power as compared to the nice smooth taste of proper AC generated on a real turbine. Oh, and all this gear WILL require skilled people to come play with often.
And you think that will compete with the $.0016 per kw/hr of projected nuclear generated electrical costs anytime in the near future, or the remote future?
Current costs of this gem of a system are on the order of a hundred grand.
At the average retail cost of electricity of about $.25 per kw/hr, it would take 27 years to pay for it in electrical savings...except the systems are only good for about ten years.
Who the hell is going to buy that?
And I haven't even started making fun of hydrogen yet.