I’m not going into the science behind the debunking of the letter to the NEMJ. For an overview see Vapemestoopid on the subject and then read Dr. Farsalinos debunking the whole study. Meanwhile Clive Bates has something to say, as does Prof. Hajek. The ruthless fact checking of Tom Pruen is always worth a read and Fergus Mason writes on the whole fiasco here.
What I’m going to do is a simple test as a vaper who happens to have a cheap and cheerful 2 ohm CE4, some 50/50 12mg caramel juice, and an Itaste VV V3 to hand. This is somewhat similar to the setup used in the experiment. I wouldn’t normally use a CE4 on the itaste; I use protank style tanks on it. But for the sake of experiment, we’ll do it. I’m going to dial it down to 3.3v and vape it normally for half an hour with my after lunch pint of water – yes, water. I usually drink water. I’ll go up by 0.1 of a volt after each puff and see what I find. I can’t do the 100 second puff I read about somewhere as the iTaste has a ten second cut-off, so instead I’ll just vape it normally.
At 3.3v the vape is fine. At 3.6v I note that it’s probably at the “sweet spot” for me on that clearomiser. Turning it up to 3.9v the vape is harsh – I’d turn it down again at this point for preference. At 4.2v I really don’t like it; it isn’t tasting burned, but it is very harsh and unpleasant. That’s as high as I’m prepared to go, and I turn it all the way back to 3.6v. On a related aside the vapour production doesn’t really change. I can get lots of vapour from a high VG juice on a 1.6 ohm coil in a Genesis tank in a mechanical mod on batteries with a nominal voltage of 3.7v. Lots of vapour does not necessarily mean that you’re vaping high voltage or low ohm.
The moral of the story is this: CE4s work best used as intended. On little low power fixed V batteries. For a VV device, upgrade the tank. Or if you must put your CE4s on a VV device, dial that voltage back to minimum and then go up very gradually until you find the sweet spot. It’ll taste far better. These sorts of studies simply have to involve end users. Garbage data in will always give garbage results.