Back of the envelope calc for fuel cell production

Let's say you had a fuel cell that could produce 3.5 kwh continuously.  If you ran it for five hours, that would produce 17.5 kwh of power.  Plenty enough to run a household.

The amount of hydrogen needed for that could be about a kilogram.  The reason for that assumption is that a fuel cell vehicle can get about 60 miles on a kilogram of hydrogen.

An electric car can go about 4 miles on a kwh of electricity.  For a Tesla, that's about 50 kwh for a vehicle for 200 mile range.

So, roughly about the same as above: 15 kwh of electricity gets you about as much as a kilogram of hydrogen from a fuel cell.

So, the above fuel cell could use about one kilogram, as stated.

It takes about nine kilograms of water to produce the hydrogen for one kilogram hydrogen.

That's about twenty pounds of water, or about two and half gallons of water.  The fuel cell would produce not only electricity, but two and a half gallons of drinking quality water.

Now for the interesting part:  It would take about one and a third pounds of smelling salts to yield a pound of water.  That means that twenty pound of water would require about twenty six or a little more of smelling salts.

The smelling salts could be broken down to ammonia and carbon dioxide.  The ammonia can break down to hydrogen for the fuel cell.

The tech exists that could do all that.  Economics?  Not sure, but there's an outfit that can make the ammonia cracker and of course there's companies that make smelling salts.  The last piece is to collect the ammonia from the smelling salts.

I think it is all feasible.