Building a simple inexpensive 1N4148WS diode array to prove for yourself that diodes produce a DC voltage:
It should not take that many 1N4148WS diodes in-series to produce 1 mV. By my calculations, it would take ~ just 17 of the 1N4148WS undisturbed diodes connected in-series to produce 1 mV. The price of the 1N4148WS diode at a quantity of 17 is just $0.04 (4 cents) each. I would predict that these diodes should recover in ~ one month. Don't get me wrong, if you produce a relative amount of current in the diodes (relative to what the diodes produce, which is ~ 10 pA), then I have no idea how long the diodes will take to reach the undisturbed state. What is meant by "disturbing" the diodes is on the order of a few nano amps.
When your diodes arrive by mail, immediately solder them together in-series. *PLEASE* only *tap* the soldering iron on the diode pins just long enough to get a basic solder connection. No need to enter you diode array in a beauty contest or take them four wheel driving. So a very quick solder job is sufficient to hold the diode pins together in-series. Make sure the diodes are soldered in the same polarity/direction. Heat quickly disturbs diodes, which is why it is important for best results to keep the soldering iron time to a minimum on the diodes. Also, do *NOT* place the diodes on a meter. The manufacturer has already tested the diodes. The odds of the diode not working is almost non-existent. The only meter you want to place on these is an electrometer. Please read the following blog post for details -->
How to replicate the diode experiments
Created on 2009-07-01 16:34:11 by EnergyMover
Diode replications, FE diodes, FE Misc devices, Free energy, Free energy devices, Science, Testing procedures, Diode, Diode replication, Free energy, Testing procedures