if enough water collects in ethanol laden gas, the water and ethanol precipitate out of solution in a process called phase separation.
Many thanks: with the name of the problem it was easy to do some research. So:
E10 will absorb about a tablespoon of water per gallon, any more than this and the phase separation happens, producing the corrosive mix. This shouldn't be a problem in my 20L tank where my main worry was droplets of water, less than a teaspoon, when I should be able to have 4 tablespoons before the phase separation occurs.
It is a major problem with marine applications, so may be worse if you live near the sea or park your bike in the rain.
E10 has reduced shelf life compared to straight petrol, like a month compared to a year, so it's more important to use a fuel stabilizer if you're not using the bike for a while. It's also more important to avoid E10 when you're concerned that the pump fuel is old.
There's a lot of literature about how ethanol rots fuel system components, particularly in old machines, but we have the benefit of simplicity by design: as long as we've replaced the fuel hoses and carburetor seals in the last 20 years they should be able to cope with the ethanol.
The ethanol can also dissolve plastic, which seems to account for various varnishes and glug in small engines where there are plastic components.
Ethanol fuel has been known to corrode carburetor parts, however my bing float chambers corroded in the 1980s, pre-E10, so I'm not sure that this is any worse than water.
While I'm running the bike regularly, every few weeks, I'm not expecting a problem and am optimistic that using ethanol fuel will remove any water that has crept into the tank. If I was laying up the bike I'd be really careful to use non-ethanol fuel and probably add a stabilizer.