“Vanilla Ice Cream” on page 37 of Van Leeuwen's Artisan Ice Cream Book by Laura O'Neill, Ben Van Leeuwen, Pete Van Leeuwen, and Olga Massov.
- The online recipe can be found here. (You'll need to scroll down a bit, and click on the image to enlarge).
- My other vanilla ice cream reviews can be found here.
I made this ice cream FOUR times before I felt like it was at all reasonable to review it. The first two times were failures, and despite all my research, I still don't know why. The third time was a failure for a different reason, one that was entirely my fault, and the 4th time was an unmitigated success producing a wonderfully rich, smooth ice cream.
This ice cream uses eight egg yolks for a quart-sized batch. Yes, you read that right - EIGHT. The ice cream is very nearly yellow in color. The recipe lacks both a liquid sugar or any sort of stabilizer, but I suspect the extreme number of yolks may keep the ice cream smooth despite that, though I'm curious to see if it remains smooth after a couple of weeks in the freezer.
For the first three tries, I was still using up older stock of Madagascar vanilla beans, and for the fourth try I finally got to use my newer Vanilla Bean Kings stash.
First try (failure): I used the no-temper technique (placing all ingredients while still cold into the pan) in a sauce pan using medium heat (as normal) and whisked nearly continuously until it reached 170F/77C and when it was done cooking it was nice and smooth, but once the custard cooled, it had turned grainy (before going into the churn). I don't think the problem could be with leaving the bean pod in the custard overnight - I always do that, and it's never caused graininess before. The ice cream was an abject failure. Good flavors, but absolutely terrible, grainy texture. Overall, it was a 3 out of 10. Other than giving a few samples away to family, I tossed it.
Second try (failure): Same as above, but I cooked the custard on medium-low and used not one but TWO thermometers - a probe-style clipped to the pan, and also regular checks with the instant-read. This, too, was smooth while hot and grainy once cool. I rescued this one by running it through the Vitamix before churning, and that raised its score to maybe a 7/10. Still not where it should be, but definitely worth eating.
Third try (failure): For this one, I followed the recipe EXACTLY (except for leaving the bean pod in the custard overnight), even using a double-boiler. I did notice that the quality of the creme anglaise was better than usual when the cooking was complete. The graininess was still there the next morning, but MASSIVELY reduced, like by 90%. I also re-read the sidebar which suggests blending the bean pod in with the custard, and I unthinkingly did so, which spoiled the experiment by adding a confounding variable). I must not have blended it long enough because the texture was just ... odd. Almost, powdery, I guess? It was scoopable, though, so I don't think it was an emulsion failure, and the odd texture seemed to be from not-quite-fully processed vanilla bean pod bits. I don't think the vanilla flavor was improved by blending the bean, so I won't be doing that in the future. I threw away the second half of the quart.
Fourth try (success): I used a bean from Vanilla Bean Kings (is it possible that the problems with the previous tries were with the beans I was using?), and I again followed the recipe as written (though I left the bean pod in the custard during the chilling process, removing it just before churning). This time, I turned out perfectly and was delicious.
For whatever reason, this custard seems far more finicky than others, but that might be an unfair assessment, and I will be doing some testing to see what caused the graininess. Like, can I get away with just a saucepan (instead of a double-boiler) if I add some liquid sugar and/or stabilizers? Or what if I use the double-boiler - is that enough to use the no-temper method even without the liquid sugar or stabilizers? Or did I simply overcook it in the saucepan? (My instant read thermometer seems to register a cooler temperature than the candy thermometer.) I am really not sure what the tradeoffs will be. It'll be interesting to find out.
Substitutions and Techniques:
- Turbinado sugar instead of white sugar (always) as I prefer the flavor for the first two batches.
- I used white sugar for the third and fourth batches (as I'm trying to figure out the cause of the graininess).
- This recipe uses only one vanilla bean, and no extract (I think I prefer vanilla ice cream to have both), but in the first batch, I accidentally used two vanilla beans, so the flavor was nice and strong despite the lack of extract. How did I accidentally use two? Before putting the milk/cream for this recipe in the freezer (I buy milk and cream in larger quantities, then freeze it in ready-to-go amounts), I split a vanilla bean and dropped it into the milk. When I was ready to make this ice cream, I set the jar of frozen dairy in a pan of warm water to thaw, then started gathering my other ingredients. Forgetting that there was a bean frozen into the milk, I got another one ready to go and added it to the pan. Then I dumped the milk into the pan ... along with the second bean.
- For the second batch, I used only one bean.
- For the third batch, I used only one bean, but I blended it into the custard.
- For the first two batches, I used the-throw-all-cold ingredients into the pan and cook until it hits 170F/77C technique (ie, no-temper technique), but for some reason, the custards were grainy.
- For the third and fourth batches, I used a double-boiler, and did temper the eggs.
Results:
- Custard 1:
- Same day: Grainy. Good flavor.
- Next day: Terrible texture. Dry and very, very hard and grainy. The flavor was good. But it was bad enough to throw it away.
- Custard 2 (blended before churning):
- Same day: Good flavor, acceptable texture.
- Next day: Acceptable texture, reasonably scoopable, Flavor was good. Much better than the previous batch, but definitely far from my best.
- Custard 3 (radically reduced graininess that might just be the vanilla seeds, blended pod into custard before churning). Unfortunately, I don't think I blended the pod ENOUGH because it made it a little chalky.
- Same day: kind of chalky from the blended pod. Not sure I like that.
- Next day: It's reasonably scoopable, even frozen hard. Texture is still off, but I think that's the fault of the pod, which I apparently didn't blend fully. I also don't think the flavor is especially improved by blending the pod, so doubt I'll bother with that again.
- Custard 4: Delicious. Just perfect both days.
Uses:
- I added dark chocolate chips, candied orange peel, and drizzled homemade orange syrup into the ice cream to make a ripple of sorts in two batches. I think ginger snap chunks would taste great with this combination, or maybe chocolate wafers (like Oreos, but better).
- With the third batch, I didn't add any stir-ins as I'm having guests that weekend, and having vanilla ice cream on hand would be a good idea.
- With the fourth batch, I added Oreo cookies and raspberry ripple.
