FWIW, I see how they determined the panels are 350W. From the specs on the eBay site, they have used the used the "max power voltage and the max power current". Without knowing more about the testing regime used or seeing proper output graphs, I am not sure this is a legitimate method of calculating output power power although a first glance, it seems
appropriate. Unfortunately the output power determined in this manner doesn't tally with other ways of determine the output power and it also doesn't tally with what you are seeing.
Others more versed in solar panel technology might know more; I'm pushing at the limits of my understanding given the scant (and conflicting) data available.
I can't think of a simple DIY
way of testing the panel to replicate the stated output V and A. Ideally a 1 ohm, 400W resistor placed directly on the panel would test the stated specs. Point the panel (clean and cool) directly at the sun at midday (on a clear day) and measure the V and A. Look up the solar radiation for your location (using previous website) and adjust for the difference between the actual radiation and the ideal radiation (i.e. 1000W). Getting a DIY
1 ohm 400W resistor is the sticking point, maybe Raymond's bicycle spoke could be cobbled into some useful.
My gut feeling is the numbers on the eBay site need to be taken with a good of salt
but ATM, I can't think of a sure way method of disproving them.
@RaymondR, you are a wealth of DIY left field knowledge, what are your thoughts for finding a load that replicate 18V and 19.4A?