Yes, it can! Using electrical reactance, dead batteries can synthesize electricity and charge themselves if they are not already damaged beyond their ability to hold a charge after charging is turned off.... http://vinyasi.info/energy/precharged-with-75-volt-spikes-in-less-than-9-minutes-from-a-cold-start.jpg Shortcut to image, above... http :// is . gd / canadeadbatterychargeitself Circuit simulated in a modified version of Paul Falstad's simulator... http :// is . gd / deadbatterybewley NiMH (nickel metal hydride) is an excellent choice for battery chemistry since it is easy to revive. That's why Matsushitsa was sued by Chevron (who possessed the controlling share for its patent) to kill the electric car put out by Toyota, the RAV4EV built during the years from 1998 until 2003.... http :// is . gd / whokilledev Run the simulation until you reach your target voltage. So, let's assume the following data... Technical Specifications for a nine volt nimh battery... input voltage: 11.4V Charge current: 100mA voltage vs current = Ohms of resistance adjacent to the batteries 114 to 1 or 160 to 1 current vs voltage 0.00877 to 1 or 0.006 to 1 charge voltage vs battery rated voltage between (1.2 and 1.267) to 1 So, a nine volt battery is charged with a voltage which is 1.267 times greater than its nine volts which yields 11.4V. As little as 1.2 charge to battery voltage is possible. Meanwhile, the charging amperage can be one fourth of the ampHours of the battery. For a nine volt nihm, this could be 280mAH / 4 = 70mA. When 11.4V is divided by 70mA, the result is a resistance of ~163 Ohms adjacent to the battery to reduce the charging current. Yet, it's dangerous to surge this simulation with that much resistance which can raise the surging voltage spikes to intolerable levels. So, it's best to run the surge mode while the resistors are very small. They're preset to 1.67 Ohms. But once you reach a voltage which is a fraction representing the appropriate current, then stop the simulation and adjust the resistors to an elevated value by hovering your mouse cursor over these resistors (adjacent to each of the two batteries) and roll the mouse wheel downwards to increase their resistance, turn off the surge switch, right-click over the oscilloscope windows and select: "RESET", and then restart the simulation. The trick is in regulating this process so as to remain within safe limits of tolerance (which the batteries can withstand), for it can easily get out of hand without the assistance of the clock timers governing the switches. The ring of four capacitors, inside an outer ring of two opposing coils, are a key ingredient in amplifying voltage. The heavy-duty 10 kilo Farad capacitors are merely there to stabilize the operation (slow down the synthesis or dissipation of charge) so as not to deteriorate (ie, dissipate energy) or accumulate energy too quickly. Without their presence, the 10 Farad capacitors would quickly amplify the storage of voltage and also accelerate its dissipation. A longer duration of duty cycle is therefore needed to alleviate the large size of the 10 kilo Farad capacitors. This enlargement of duration of duty cycle can be accomplished in either of two ways... 1. Slow down the Hz of the clock timer associated with the switches governing the duration in which the four-ring-capacitor is kept closed, or... 2. Increase the percentage of the duty cycle. #2 is limited while #1 is not. So, by shortening the Hz to smaller values (option #1), the wavelength of each cycle is lengthened allowing the capacitors to store more energy before they are discharged at the beginning of the next cycle when the ring is rejoined. By the way, 140mHz equals 7 and 1/7th seconds per cycle. 500mHz is equal to a two second cycle. So, 140mHz stores more energy than does 500mHz allowing a larger surge to amass before it is released when the ring of four capacitors is rejoined. The irony is that the rejoining of the capacitors does more than simply discharge them. It also allows for the build up of charge depending upon how long the ring formation is maintained. Thus, I like to call this: Bewley's Archetype,... http://vinyasi.info/energy/bewley-archetype.jpg Simulation... http :// is . gd / bewleyarchetype ...and in LTSpice... http://vinyasi.info/circuitjs1/texts/Eric%20Dollard/bewley-archetype-in-ltspice.jpg LTSpice simulation file... http :// is . gd / bewleyarchetypeltspice ...derived from scanning through LV Bewley's texts on the subject of "Travelling Waves on Transmission Lines" and going through all of his network analogs of transmission lines to see if I can create from them an archetypal module - similar to Eric Dollard's analog computer in LMD mode (longitudinal magneto-dielectric). Eric cites Bewley as being instrumental for Eric's education into the realm of electrical engineering when Eric was still in High School. It was mind blowing for me. I don't understand engineering stuff too well, but I can pick out patterns and form them into new ones. So, I took one of Bewley's diagrams and used one edge (the ground plane) as a fold mark to mirror (duplicate in reverse orientation) what was above ground (the transmission line) into another transmission line below ground mirroring the one above and joining them together to create a circuit module. http :// is . gd / ericbewley Eric actually uses this creation of mine in his lectures and papers so I didn't just come up with it, myself. The master (Eric), himself, already came up with it long before I rediscovered it on my own... Graphic... http://vinyasi.info/circuitjs1/graphs/closeup%20of%20a%20refined%20LMD.jpg Enhanced... http://vinyasi.info/circuitjs1/graphs/enhanced-LMD.jpg And mentioned in this YouTube video... https://www.youtube.com/watch?v=cCJcU7INwnU&feature=youtu.be&t=3h5m15s Shortcut to YouTube video, above... http :// is . gd / refinedlmdyt And contained on PDF page 110 of his power point presentation depicting an analog to an "element of coil"... http :// is . gd / refinedlmd It's pleasing to know that I am picking up a few crumbs which Eric has dropped as he walked along his path... http :// is . gd / ringharmonics This answer is mirrored, here... https :// is . gd / ericquora Please see... http :// is . gd / paratexts ...for more information.