Ah, curdling. The bane of my ice cream making existence. I can't count how many custards have been sadly ruined after several minutes of infusing, carefully tasting and adjusting, tempering, slowly heating/stirring, and then... #^$*! One second too many turns perfectly smooth, thickened custard into egg drop soup. This is probably a testament to my somewhat reckless attitude towards cooking, as I'm sure this rarely happens to other ice cream making people. Or at least, multiple times.
Sure, people say you can "save" a custard by blending it back into shape, which I did once to some ginger ice cream (it was a holiday emergency), but it doesn't come out as good as it could be. So what's the idea behind curdling anyway? And how can we prevent curdling of any custard?
The Inquisitive Cook (Accidental Scientist) states that "In both cooked and stirred custards, setting happens at temperatures well below boiling. Cooking at too high a temperature or for too long toughens proteins and squeezes out liquid. This makes a baked custard "weep," and a stirred custard curdle--both signs of overcooking."
So temperature matters, as one would expect. But they also claim that it is the rate of heating that matters: "When egg proteins are heated quickly, there's a very small temperature difference (just a few degrees) between thickening and overcooking, so that custards seem to curdle instantly. When heated slowly, this range widens to 10° F or more." This is why some recipes call for use of double boilers for stirred custards, since they allow for a longer "grace period" before reaching the curdling point. Stirring is important for even heating throughout the custard, and not allowing one region (say the bottom) become overheated.
Ok, so now I have my double boiler and low heat. Wouldn't it be great if, with the help of my trusty thermometer, I knew exactly what temperature to look out for to prevent overcooking? This website says I can, with a greatly oversimplified formula. Unfortunately, application of this rule does not work. As described in Experimental Cookery, From the Chemical and Physical Standpoint, "The temperature at which coagulation (thickening) starts varies with the varying proportion of ingredients of the custard and the rate of cooking." So what works for my recipes won't work for someone else's, in other words.
On a related note, a few weeks ago I was reading an ice cream recipe on a popular blog that said to heat the custard to exactly 84° C (183° F). Of course, this was before I read about curdling, and so I thought "Oh great, a guideline for what temperature to heat my custard up to!" Next thing I knew, I had a curdled mess on my hands. What does help is keeping a thermometer in and taking note of approximately when thickening occurs. This way, from that point on you'll have a rough guideline (assuming the same proportions are used) of when to stop.
And lastly, with this newfound knowledge I made a delicious and beautifully pink strawberry basil ice cream! Simply use this recipe by Emeril (I cut it in half), but infuse the milk and cream with bunches of chopped fresh basil leaves for one hour. I also added a pinch of salt and half a packet of gelatin as described here. For presentation, I sprinkled the top with crushed dehydrated strawberries picked out a box of cereal, hehe. Wrong season for this flavor I know, and in summer it would be twice as good and refreshing, but keep it in mind until then!