I am trying to make the caramel ice cream from TPPB.  I have made the vanilla ice cream many times and it is super good do I thought I would try the caramel.  I have tried 4 times now.  I can get the sugar and water to the amber stage but when I add the milk, it curdles and separtes and no matter how much heating I do it is not smooth, it is gross.  I have made caramel before and it is usually whipping cream added to the sugar syrup.  The recipe says to add the milk.  Is that the problem?  Thanks!

moderncake, I made this ice cream a few months ago and I did find that there were clumps of milk proteins in the caramel that didn’t dissolve. I wasn’t surprised because when the milk was first scalded, it separated to some degree and formed a skin while I was keeping it warm. It was pretty gross but I wasn’t too concerned about it - though in hindsight, maybe I should have been - because the solids were strained out when I sieved the mixture.

Is the curdling you’re describing more severe than this?

It was pretty gross and there was a lot of it.  I tried again with more water because I found it was really hard to keep the sugar from crystalizing with so little water and but is still didn’t turn out.  Almost a litre of milk, 4 cups of sugar later, I decided to just make the vanilla.  It turned out fine.  I will try again another day with the caramel.  This ice cream is for my daughter’s birthday party and I don’t want to take any chances.  Thanks for your input!

Hi, moderncake!

I haven’t made this ice cream (so I don’t know how much sugar/water is used), but two things I do when I have to caramelize sugar are (1) grind it fine in the food processor (like a minute or so), and (2) the day before, put it in the pan with the water, covered (so it won’t evaporate).  The time and grinding helps it sort of “pre-dissolve.”

As with our caramel mousseline topic, I wonder how vanilla ice cream, but made with brown sugar, would approximate caramel ice cream?

—ak

Hi moderncake!
I have no experience with this particular sitaution, but I looked up the errata for TPPB, and didn’t see anything.  But…I did come across this post by Julie which may be of some help to you.  Perhaps she will chime in with advice. 

This is a question by sue:

I am a trained pastry chef and a great admirer of your books. All your recipes have been exacting and fool proof until recently. I made the caramel ice cream out of the pie and pastry bible and had a great deal of trouble incorporating the hot milk into the cooked sugar. Much to my dismay, the mixure kept curdling. I strained it but there was lots of residue. I then made the anglaise and tempered it with the caramel that had been strained out. I still wasn’t happy,...the mixture didn’t have the caramel flavor I wanted and there was still a lot of curdled residue in the strainer. I then poured the entire mixture into the blender and voila, everything came together. I restrained the mixture and there was very little residue this time. Finally, I added the caramel custard mixture to the cold heavy cream that had vanilla bean seeds in it. The recipe was written so that the hot milk was added into the cooked sugar and reduced to 3/4 cup. The problem is that milk can’t stand up to this kind of cooking. Wouldn’t it have been better to add part of the heavy cream to the caramel instead and then add all that to the anglaise and finally the heavy cream (minus 1 cup cream/add 1 cup milk) steeped with vanilla bean seeds. Anyhow, that’s how I would make it on my next try.
Julie’s response: 

Sue, I’m not Rose, but was able to solve a similar problem with curdling caramel creme anglaise. In my case, the problem turned out to be that the caramelized sugar was becoming too acidic, so that adding the hot milk to the acidic sugar resulted in curdling.

For me, the acidity was both from using blonde organic sugar (which has a bit of residual molasses), and from taking the caramel a bit too dark (the darker the caramelized sugar, the more acidic it becomes).

When I switched to refined white sugar and used a thermometer to make sure it didn’t exceed the recommeded temperature, the problem disappeared.

Hope that helps!

The rest of the discussion is here:
http://www.realbakingwithrose.com/2007/07/corrections_the_pie_and_pastry.html#comments

Hopefully this is of use to you.

Sherrie, thank you for finding that- just what I was thinking!

Milk heated in the prescence of an acid curdles every time, so if the milk is curdling, the mixture is too acidic. 

Possible reasons:
- caramel too dark (even white sugar, when caramelized, becomes more acidic the darker it is)
- sugar with molasses
- milk gets more acidic as it ages, a result of bacterial activity.  Milk doesn’t stay bacteria-free until suddently it spoils, there is a slow progression as bacteria increase.  It’s not unlike culturing yogurt from milk, in that bacteria are making the milk more acidic.  Except that with aging/ spoilage, the result is not desirable! 

Good luck if you try it again.

Thank you so much everyone!!!  I have all of Rose’s books and have taken for granted that they always work out so I was really frustrated when the ice cream didn’t.  The sugar was really dark (the recipe says to allow it to get to I think, 380 degrees) so it was probably super acidic.  I will look at the suggestions and try it again and report back.