Yep, that's why it's more accurate to blend by weight. But suitable blends have all been found by trial and error over the years by mixing by volume because everyone's got some sort of container handy and it's soooooo simple .

Take for instance FAI fuel which is 80/20 methanol/castor by volume (or 4 litres of methanol to 1 litre of oil). If it was mixed by weight then you'd have 4 kg (or pounds, whatever) of methanol and 1 kg of oil. Convert that to volume and (to a close approximation) it'd be 5 litres of methanol to 1 kgm of oil or a volume mix with 16.7% oil which is how it appears that Byron mix their fuels. However, if Byron used weight to get a final result that matched a mix by volume then they'd take 3.2 kg of methanol and mix in 1 kg of oil. That would give a consistent fuel blend no matter what temperature it was mixed at and be what you called the double translation.

From photos I've seen of commercial fuel manufacturer's plants, the bulk methanol etc is stored in huge tanks outside so there'll definitely be some expansion/contraction between winter and summer in which case mixing by weight (using that double translation) ensures a constant blend.

With that "frosty flakes" thing, I think that's an indication that the chemically extracted oil has been blended with the pressed oil. We don't have those extremes of temps here in Oz but a few times I've had fuel in my fridge freezer down to around -10C (14F) with Castrol M and I've never seen any flakes at all. Not that there's anything wrong with the chemically extracted oil other than the flakes could block a filter if anyone is silly enough to go flying in weather like that .