Well, no, that's still wrong. The >100% efficiency comes from using two slights of hand: 1. Counting only the electrical energy input. But you also need to s

Well, no, that's still wrong. The >100% efficiency comes from using two slights of hand: 1. Counting only the electrical energy input. But you also need to supply heat at the same time. That's the whole point of this product--use heat ~~instead of~~ (edit: in addition to) electricity to split water. Critically, they are assuming you get "free" (as in beer) waste heat at 800C... which is fairly silly. 800C heat is high-grade energy, not waste heat. Exhaust gas from a combined-cycle NG power plant, for example, is more like 250-300C. Thermal steam plants operate with a Th of ~550C (so the waste heat is much, much cooler). 2. Ignoring the fact that you need to heat the reactant (water) to 800C before you can have the discussion in #1. Now, the problem is that the heat capacity of water is much higher than an equal mass of H2 and O2. So, you can't even use the hot as-produced gas to heat the incoming water to anywhere near the operating temperature of the electrolyzer. The efficiency in a practical application is probably half to two-thirds of what they are claiming.