The COP is a fudgy number when it comes to LENR. Does it refer to some ratio of energy-in/energy-out? (Usually, in one way of calculating or another.) Does it refer to the power-in/power-out? (I have seen it used this way.) What does it mean for the COP to be "X" when, if you could chain a series of reactors undergoing a reliable reaction together, you could get a COP of 2X, 5X, 100X or 1,000,000 X?
Keeping that in mind, it is interesting to learn that Torkel Nydberg is quoted in E-Cat World concerning unnamed sources who have visited Rossi's customer's plant and who report a COP ~ 20-80 [1]: I know first hand from very reliable sources that themselves have visited > the Rossi/Industrial Heat E-Cat customer that the plant works very well. > This has been verified both by measurements made by the customer and by > significantly reduced electricity bills. The plant seems to be able to > produce heat from electricity with a COP in the range of 20-80 depending on > the level of self-sustain-mode applied. I guess that is what Rossi is > working on right now. Medium and long-timers will be aware of the caveats that go with secondhand reports of this kind. Eric [1] http://www.e-catworld.com/2015/04/22/1-mw-e-cat-plant-watch-thread-update-1-rossi-production-cost-in-kw-is-very-competitive/

