We get precious little detail about the QuarkX from Andrea Rossi, but yesterday there was a question and answer on the Journal of Nuclear Physics that might shed a bit of light on the battery requirements:
May 10, 2017 at 1:59 PM
Dear Dr Andrea Rossi,
Which is the voltage you need from the batteries to use them as a power source?
May 10, 2017 at 4:38 PM
It’s not possible to determine input power from volts alone, and we have been told nothing regarding the control system that is used with the QuarkX, so this information does not reveal very much; however, it does indicate that as a practical matter, the battery input would be easy to arrange given the wide availability of batteries of that voltage range.
According to Andrea Rossi there is no difference in the performance of the QuarkX whether the input is AC (grid) or DC (battery), and he stated again recently that his experiments using simple calorimetry confirm the same results as reported in the paper he authored with Carl-Oscar Gullstrom where the power output was calculated using Boltzmann and Wien equations (since temperature was determined by spectroscopy). In that case the calculated COP (based on the reported Watts in v. Watts out) was over 20,000.