Guys:
   Just using junk to make these cells - will not work for long.  As John B mentioned, you have to use the best purest metals, carbon or what ever is used needs to be of good quality, otherwise the cells loses power and you got nothing, or almost nothing.
  I now see that the semiconductor protective layer is really important, but I'm not sold on watering the cells to produce output. At least that is not the direction that I'm taking.
John B has also mentioned that Mg is not affected by oxidation when the protective layer is used, but aluminum is affected, if I understood him correctly.
  I do think that the baked ceramic glass like electrolyte layer looks like it would be the best way to go, and if presure is applied it also may help to increase the output, for whatever reasons.

Last night I took the biggest capacitor that i had bought in the 20 I bought yesterday.It turned out to be rated 16v2200uf,I found a cell producing about .85 volts and hooked it up in parallel with it(pos to pos and neg to neg).After a few seconds I took the capacitor off and used the voltmeter on it.The capacitor had a small voltage of about 5/1000s volt on it.Convinced I had it hooked up ok I left it on there an hour.
After an hour the capacitor showed a voltage of .75 volts and the cell had dropped down to .48 volts.I unhooked everything.An Hour later I came back and measured the same cell,its voltage had popped back up to.80 volts.So I hooked up the capacitor again and left it on overnight for 6.5 hours.I read it this morning and its voltage was .91 volts(the capacitor).I was able to discharge the capacitor thru an old speaker and get a small burst of static both times.Once again the cell was down to about .5 volts.triffid

So I got to thinking ,these cells remind me of what they call a seep well.A seep well works by collecting water from a trickle of water coming into it.Once it fills up you might have a lot of water.But use it faster than its coming in and the water level drops again.Leave it alone for a while and the well fills up again with water.




So that said these cells make electricity from zero point but not very well(yet).Once we get a gusher then we will be rolling.triffid

Pretty much junk is all I ever had to work with.I once needed a 200 mesh screen stainless steel screen and found out a square inch of it was going to cost me about $125.00 per sq in.I went to a garage sale and found more than enough brass screen of the same size for $3.00.Brass was ok for my project so I bought it.And used it.I'm not worried too much if it will last 20 years or not.Will it last me long enough to complete my experiments?I will let the giant corporations worry about the cost of the finest materials and the top grades of steel etc.ButI will use what is at hand.I will use what I have.Or I will never do the project cuz I'm waiting for the money that never comes.triffid

  These cells will charge a capacitor or a dead AA battery to their own voltage level, but not to the batteries normal working current levels. As these cells have no current.  So, you can charge a capacitor, but when you connect an led to the same capacitor is will only light for a few seconds, if that.  So, what use is a capacitor or battery that is charged with 0.8 volts (or higher voltages), and no current?  It won't light anything, not even an led.  It won't run an oscillator other that low or blinking led. I don't know how many volts it would take to light an led, with  hardly no current, without the use of an oscillator circuit.  I don't see any useful benifit in blinking or dim leds when the oscillators are used. 
  It is normal that the cells drop to lower voltage levels due to their high impedance when under a load, and also normal that they return to the standing voltage levels once disconnected from the load.  High impedance is the hardest thing to overcome.