Canelé

[Anonymous Modernist 8324]

First off, I'd like to express my gratitude for the MC book. It's water in the desert for me. Up till now, my only trustworthy reading material was Harold Mcgee's and Hervé This' books. Exquisite job, people!

My question involves the gelatinisation of starch in flour.

A bit of context first.

In attempting to make the notoriously capricious little canelé, the constant problem I have is that the batter keeps popping out of the molds. Whether it's silicon or copper, it won't stay put. This gives an uneven color and even burns at times. I've done the resting period of 24,48,72 hours and they would still "misbehave". This got me to thinking about why the batter needs resting in the first place. I've looked at many recipes and found variations of quantity in ingredients, but the constant in all of them was the resting period. The problem is that nobody explains why this is so. One person mentioned fermentation, but I highly doubt that even factors into the why of the resting. This is where Harold Mcgee came in handy. He talks about gelatinisation of the starch in flour.

When I looked at the ingredients (whole milk, cake flour, sugar, rhum, eggs, vanilla) I was trying to pick which ones would expand with heat to a significant degree. If I factored in the inevitable small air pockets found in the batter, what else was going to contribute to the expansion. When flour is mixed with water, the starch starts to swell but only to about 30% of it's own weight. Add heat to this and the absorption rate increases making them bigger. So when the batter is resting in the fridge the starch is swelling, but it's limited. I imagine this is why there's a thicker sediment at the bottom and this requires a stirring once in a while. So after the resting period I pop them into the oven. I'm guessing the heat is increasing the swelling of the starch and may be contributing to the pushing out of the molds. How do I get around this?

Cook the flour.

I figured if I make a kind of béchamel and then mix in the rest of the ingredients, that would prevent the batter from expanding too much and making them pop out of the molds. I usually stop cooking until I get the thickest results before adding the rest of the ingredients. Seems to work, but I haven't done enough tests to confirm this method. My question is: Does cooking the flour into a sort of béchamel in order to get maximum gelatinisation of starch, compromise structure/texture/taste at the baking phase? Am I even on the right track as far as solving the expansion issue?

Thanks,

Robert