Jump to content
  • Welcome to the eG Forums, a service of the eGullet Society for Culinary Arts & Letters. The Society is a 501(c)3 not-for-profit organization dedicated to the advancement of the culinary arts. These advertising-free forums are provided free of charge through donations from Society members. Anyone may read the forums, but to post you must create a free account.

Home Made Ice Cream (2015– )

Recommended Posts

Which reminds me, earlier in this thread I did a cooked milk tasting experiment at paulraphael's suggestion.  The idea of "cooked milk taste" at much less than boiling temperature is nonsense and a non-issue.


The corollary to this is that boiling milk for ice cream is a bad idea.


Again, this has nothing to say about resulting texture.

Link to post
Share on other sites

I was surprised to discover this, not just because of all the lore, but because Jeni Britton Bauer (of Jeni's Splendid) cooks her base at 75C for something like 90 minutes, saying she likes the flavor of cooked milk ... which she describes as sweeter. Possibly this is a marketing statement. Her other reason for the long cook is turning milk proteins into stabilizers and emulsifiers. 


I'm playing around with this idea also. So far I haven't noticed a textural differences with different amounts of cooking. But Jeni must be on to something, since she uses neither eggs nor hydrocolloids in her mix. 


[edited to add: cooked egg flavor is likely a different story. My tests weren't designed to learn anything about this, since I use so little egg. But for anyone making a more traditional french style ice cream (4 or more yolks per 1000g) it may be worth testing. However ... with this much egg, you're already getting a lot of stabilization and emulsification from the custard and lecithin.]

Edited by paulraphael (log)

Notes from the underbelly

Link to post
Share on other sites
  • 3 weeks later...

Hi fellow scoopers - Long time reader / first time poster!


I was wondering how you guys used the ice cream calculator for recipes that involve adding a large qty of non dairy into the mixture (canned pumpkin, fruit purees, etc) ?


For Thanksgiving, I attempted to prepare a Pumpkin ice cream w/ a Honey Molasses Swirl but the ice cream came out too firm w/ it being difficult to scoop and chunky - (I think it needed more sugar / possibly more stabilizer?) 


Recipe adopted from a Williams Sonoma one found here:


and idea for honey molasses swirl found here:



Instead of using the heavy cream / sugar / yolks recommended on the site, I utilized the numbers that JoNorvelleWalker had been using to arrive at a recipe of


1 cup (244g pumpkin puree + tblspn vanilla paste - reserve


1000g heavy whipping cream

112 brown sugar (subtracted 8 due to 8g sugar content in puree is this an equal substitute to white?)

6 yolks

1/2 tsp cinnamon

1/2 tsp ginger

pinch nutmeg

pinch salt

5g Cremodan to help offset puree




1/4 cup honey

1/4 cup dark karo


I have adopted the RuebenPorto method with my Thermomix just like JoNorvelleWalker has done w/ her KitchenAid Heating Mixer.


To prepare:

I mixed the base contents in a Thermomix @ 80C setting for an hour (instant thermometer reads 69-73 so it's pretty on point) and had a reduction in base of about 80-90g.  I then used an immersion blender to homogenize for 5 minutes before chilling in an ice bath.  I aged for about a day before adding the reserved puree mixture and churning in my Musso Pola for about 9-10m.


For the swirl, I used some honey and some dark karo corn syrup (as I could not source the gingersnap liqueur in time).  That turned out okay but froze a bit into a hardened block - will try w/ the liqueur next time for a similar swirl to see what happens.



About me:

Total cookbook junkie - have made or attempted ice cream recipes from Fat Duck, Frozen Desserts, Sweet Cream and Sugar Cones, Jeni's Splendid, Lebowitz's Perfect Scoop.  Incidentally have also made recipes from Rueben's IceCreamScience blog and PaulRaphael's Vanilla base (didn't realize till I saw the underbelly link), and currently working with JoNorvelle's recipe!

Link to post
Share on other sites

Would 1 Tblspoon of alcohol have had that much effect on the base to counteract freezing?  Moreso than the stabilizer that I added? 


I still have some remaining so I'm going to try to melt the entire batch down (honey/molasses syrup) included to see if I can reduce remaining water content / melt in sugar content and see if that rescues it.  I'll know in the future to heat up the puree to remove water content and possibly recompensate w/ more sugar.  I had really great luck w/ the base recipe to make a Cranberries and Cream batch for another Thanksgiving party that turned out to be a hit!

Link to post
Share on other sites

Compared to glycerides, ethanol is a small molecule.  It tastes better too.


I may be wrong, never having tried it, but I don't think Cremodan does much for freezing point depression.  Seriously, I'd increase the sugar till you get the sweetness that you like, then add bourbon or your ethanol of choice.  I doubt it would take more than three tablespoons at most.


Also I'd add the alcohol just before spinning, not while the mix is cooking.  I used to rely on alcohol for freezing point depression before I discovered Ruben's method.  Now alcohol makes my ice cream too soft.  But you have all that water in the puree to counteract.

Link to post
Share on other sites

Alcohol does soften ice cream at a given temperature, but it doesn't help texture in any other way, and can actually promote iciness. I think it's better to use sugars and other dissolved solids to control freezing point. If the sweetness level is where you want it, just substitute dextrose for a portion of the sugar. Between table sugar, dextrose (less sweet, greater freezing suppression) and trimoline (sweeter, greater freezing suppression) you can finagle any combination of sweetness and hardness. (As Jo said, stabilizers do not affect freezing point).


Milk solids are generally important ... they suppress freezing point, add body, and encourage a smoother texture by a number of mechanisms. Solids content is traditionally increased by adding nonfat dry milk, although people here have been experimenting with reducing the milk themselves. 


If you have a lot of solids from another source, like a puree or nut paste you can compensate by reducing the milk solids. Of course things like pumpkin puree add water, also. It's important to take into account the added water as well as the added solids so you don't completely mess up the texture.


I've been playing with nut butter ice creams. These add a lot of solids and a lot of fats, so I'm compensating by eliminating cream, and by eliminating dry milk (not sure how the final recipes will be, but probably in that direction).


I'd suggest with the pumpkin puree that if you're making it yourself, you cook it by roasting, so it will have as low a water content as possible.

Edited by paulraphael (log)

Notes from the underbelly

Link to post
Share on other sites
  • 3 weeks later...

There have been a few mentions of salted and/or burnt caramel ice cream in this thread, but no actual details on how to make it.  I want to use Ruben Porto's basic recipe and method.  I assume the amount of sugar would have to be adjusted downward, as would the amount of liquid, but I don't know how to go about this.  David Lebovitz has a recipe that he says is great, but fitting his method and ingredients into Ruben's is the issue.  I think I would start by substituting the (liquid) caramel for some of the sugar in the recipe (though "burnt" caramel would not be as sweet as ordinary sugar).  And because there is still some water left in caramel, substitute the caramel for some of the nonfat milk (which is mostly water)?  You can see the dilemma.  Any help would be appreciated.

Link to post
Share on other sites

Hi there Jim!


There is indeed a recipe for salted caramel ice cream on the blog http://icecreamscience.com/homemade-salted-caramel-ice-cream-recipe/

I haven't had time to install a 'next page' button on my recipes page so you can't see all of the recipes yet. I'll be updating the blog over the next couple of months.


I'd recommend using the recipe and quantities from my vanilla bean ice cream recipe but follow the instructions on the salted caramel recipe for the caramel part.


I hope that helps. Let me know if you have any questions.


All the best,



Link to post
Share on other sites


I just finished churning the salted caramel ice cream.  It has a wonderful flavor.  I did have to make a last-minute adjustment:  I was so conscious of getting the caramel dark enough that I went just a second or two beyond the correct point.  After I had heated the mix and then chilled it, I did a taste test, and the flavor was definitely too far on the burnt side (though not by a huge amount).  So I heated a small amount of cream and dissolved some brown sugar in it, cooled that, then added it to the caramel mix.  I know that I was throwing off the fat content and the proportions of the recipe a bit, but I thought it was better to take that chance rather than have ice cream that people would not like.  After churning, it appears that the added sweetness of the brown sugar and the slight diluting of the caramel taste with the added cream worked.  I'll be more restrained in caramelizing the sugar next time.


I should add that there is a wonderful ice cream shop in Cambridge, Massachusetts (Toscanini's), that sells "burnt caramel" ice cream (delicious with an apple dessert), and their flavor is so far on the burnt side that they would probably have found my final product quite bland, but I know my audience for the ice cream I made.


The heating/evaporation process took a long time.  I must find a 9" pot in which to do the heating (mine is 8").



Link to post
Share on other sites
5 hours ago, Jim D. said:

The heating/evaporation process took a long time.  I must find a 9" pot in which to do the heating (mine is 8").


There is a solution.


Note that howsoever you speed up the evaporation process you still need to cook long enough for pasteurization*.  This time seems to vary from country to country.  For the US see time/temperature table in Modernist Cuisine vol 1.


*Does not apply if you're not producing ice cream commercially and don't care for your family much.

Link to post
Share on other sites


About the KitchenAid attachment:  I read the linked thread but ended up somewhat confused.  First, what KitchenAid mixer models are actually compatible?  I have a tilt-head mixer.  Williams-Sonoma specifies "Compatible with KitchenAid Artisan, Metallic Series and Design Series tilt-head mixers," but the KitchenAid site states "Fits all Tilt-Head Stand Mixers models, or as a standalone unit."  I suppose I could order directly from KA and (perhaps) settle the question.  Second concern:  In that other thread, the overall assessment of the attachment (I'm speaking of using it for making Ruben Porto's ice cream) was not all that positive.  JoNorvelleWalker said the temp was quite a few degrees off from where it was set and that it dropped a lot when the lid was removed.  I'm assuming the lid can stay on while doing the evaporation procedure, but does the evaporation happen with the lid off?  It would seem necessary to remove the lid from time to time to check the mixture.  And how does one know when enough evaporation has taken place without removing the whole attachment and weighing it (as the Ruben process describes when using a pot on the stove)?  I'm sure the temp drop quite a lot if this happened very often.  Forgive me if I am missing something obvious.  Any help would be appreciated.

Link to post
Share on other sites

There are two models of the Precise Heat Mixing Bowl -- one for bowl lift mixers and one for tilt head mixers.  I have the bowl lift model.  William Sonoma is where I bought mine.  WS has a good return policy as I recall.


Measuring the percent reduction by weight is something I cannot do with my current setup.  Not that I wouldn't like to.  However once you get the variables figured out, cooking in the PHMB is reproducible.  The variables are temperature, time, and mixing speed.  (Neglecting ambient temperature, relative humidity, and I suppose atmospheric pressure.)


Ruben's method fixes the temperature, so the variables are then time and mixing speed.  Once the proper reduction is achieved the recipe result can be repeated over and over without thinking about it.


Here is what I'm doing:  I heat the ingredients quickly on the stove (in a pot, not the PHMB) to 160 deg F (this takes about 10 minutes).  I pour the mixture into the preheated PHMB set to 198 deg F, and mount the PHMB on the mixer, set at speed 4 with the wire whisk, and cook for one hour -- completely hands free and unattended.  The mix stays at 160 deg F.  This is all with the lid off, of course, if that was not clear.


After the cooking is completed I homogenize and chill the mix in an ice bath as normal.



Edit:  it may be possible to skip the stovetop step and do all the cooking in the PHMB but I haven't tried this yet.  I'm sure it will take longer to come to 160 deg F.

Edited by JoNorvelleWalker (log)
Link to post
Share on other sites

Thanks for the helpful reply.  Are you using Ruben's original recipe that called for heating for one hour rather than the newer version?  The amount of the ingredients varies in the two versions, and he is now replacing the older recipes on his website.  So you assume the evaporation is what it should be after an hour in the PHMB?  Is there any practical way of weighing the PHMB + mix to be certain?  For instance, how heavy and how easily removable is the PHMB?  Would the temp plummet during the weighing so as to make this impractical?

Link to post
Share on other sites

I'm using a lot higher butterfat than what Ruben's recipes call for, but, yes, I started off with his original one hour method with no added milk solids.


The PHMB is quite light, lighter by far than the Le Creuset I was using previously.  It is as easy to remove as a normal KitchenAid bowl.  I wouldn't say the temperature would plummet but it might go down a little.  Unfortunately my scale goes only up to 300 grams.  (Hangs head in shame.)


In my case the mixing time is based on the quality of the resulting ice cream.  Longer times made the ice cream slightly softer than I would like.  I might even go shorter, keeping in mind again pasteurization.

Link to post
Share on other sites

Hi Jim!


Good to hear the salted caramel turned out well after some slight tinkering. I've heard of Toscanini's burnt caramel flavour but just can't imagine it being all that nice. I'll have to give it a go the next time I'm over the other side of the pond.


Have you tried the new 25 minute heating recipes on the blog Jim? Jo and Jim, have you looked into using a magnetic stirring hot plate to heat and stir your mixes? I think we already spoke about this Jo if my memory serves me well. There are some used hot plates on e-bay by IKA that are good. 


Merry Christmas to all.



Link to post
Share on other sites

Have you tried the new 25 minute heating recipes on the blog Jim? 


Yes, that is the recipe I use.  Remember, you suggested I use the caramel ingredients but the 25-minute vanilla technique?  The problem is that it took over an hour to reach the target weight for the caramel.  All of the other recipes have taken about 45 minutes, except when I used a (non-stick) wok for heating--it was much wider and cut the time dramatically.  But with the wok I could not use my probe thermometer from Thermapen that I bought specially for this purpose--the sloping sides of the pot didn't allow the probe to be beneath the surface of the mix.  I am now looking for a 9-inch pot (stupidly I used my 9-inch enamel pot for making no-knead bread, and the high heat has ruined it for anything else).

Link to post
Share on other sites
6 minutes ago, Ruben Porto said:

Hi Jim!


Good to hear the salted caramel turned out well after some slight tinkering. I've heard of Toscanini's burnt caramel flavour but just can't imagine it being all that nice. I'll have to give it a go the next time I'm over the other side of the pond.


Have you tried the new 25 minute heating recipes on the blog Jim? Jo and Jim, have you looked into using a magnetic stirring hot plate to heat and stir your mixes? I think we already spoke about this Jo if my memory serves me well. There are some used hot plates on e-bay by IKA that are good. 


Merry Christmas to all.




I was the one who suggested using a magnetic stirring hot plate, Ruben, but I went with the PHMB.  As it turns out the price would have been about the same...after factoring in the cost of the new commercial KitchenAid.  Though the orange power cord is so pretty, I must say.


Do you find that your new recipes with the shorter cooking times give better quality than the original recipes, or is it a matter of convenience?


Link to post
Share on other sites
29 minutes ago, Jim D. said:

Yes, that is the recipe I use.  Remember, you suggested I use the caramel ingredients but the 25-minute vanilla technique?


Ah, I see. I wasn't sure whether you ended up going with the 25 minute method for the salted caramel. Interesting that the recipe took an hour to make. How long does the reduction process take in your 9" wok? I would recommend that you keep your mix at around 72°C for at least 25 minutes, even if you reach the desired reduction weight before then. The aim of the heating process is two-fold: 1. is to concentrate the mix, thereby increasing the non-fat-milk-solids, primarily the protein, which significantly contributes to smooth texture, and 2. to promote reversible protein unfolding, which, again, significantly contributes to smooth texture. If you don't heat for at least 25 minutes, you are unlikely to get the same rate of reversible protein unfolding and I'm certain that the texture won't be as smooth and creamy.

Link to post
Share on other sites
40 minutes ago, JoNorvelleWalker said:

I was the one who suggested using a magnetic stirring hot plate, Ruben


Oops. I knew I had you to thank for the stirring hot plate introduction Jo. :)


The new 25 minute heating method does produce excellent texture but it is not quite as smooth as that produced by the longer heating time. I've introduced it purely to make my recipes more accessible to the home cook. I've also had to increase the fat content in the new method, which gives a slightly heavier ice cream and masks the flavour just a tad. Interesting to see that you use a higher fat content than I do. Let me know if you would like me to send through a copy of the spreadsheet I use to calculate my mixes.

Link to post
Share on other sites
4 hours ago, Ruben Porto said:


How long does the reduction process take in your 9" wok?

I didn't keep close track of the time, but it was in the 25-30 minute range.  I had to hold a Thermapen in the mix the whole time, and that was not an acceptable arrangement (at least for me).  The wok is about 10" in diameter.  I think that a 9" pot will be a solution for me.  As attractive as the idea of the KitchenAid unattended attachment sounds (and I haven't totally abandoned the idea), I don't think I would feel comfortable without weighing the mix from time to time.


Happy Boxing Day, Ruben

Link to post
Share on other sites
  • Similar Content

    • By Kasia
      This year, mischievous nature tried to upset my daughter's birthday plans. Spending your birthday in bed with a thermometer isn't an excellent idea ¬– even for an adult. For a teenager it is a drama comparable to cancelled holidays. My daughter told me that you are thirteen only once. And she was right. Literally and figuratively.

      I wanted to sugar the pill for her on this day and cheer her up for a bit, so I prepared a caramel cake with bananas – banoffee in the form of a small birthday cake. My sweet magic and the dinner from her favourite restaurant worked, and in the end her birthday was quite nice.

      Ingredients (17cm cake tin):
      150g of biscuits
      75g of butter
      200ml of 30% sweet cream
      250g of mascarpone cheese
      2 tablespoons of caster sugar
      2 bananas
      300g of fudge
      1 teaspoon of dark cocoa

      Break the biscuits into very small pieces or blend them. Melt the butter and mix it up with the biscuits until you have dough like wet sand. Put it into a cake tin and form the base. It is worth rolling it flat with a glass. Leave it in the fridge for one hour. Spread the biscuit layer with fudge and arrange the sliced bananas on top. Whisk the chilled sweet cream with the caster sugar. Add the mascarpone cheese and mix it in. Put the mixture onto the bananas and make it even. Sprinkle with the dark cocoa and decorate as you like. Leave it in the fridge for a few hours (best for the whole night).

      Enjoy your meal!

    • By Kasia
      Sometimes we have days – may there be as few as possible – when nothing works out. I can even burn the water for tea. I have two ways of dealing with such days. The first is to sit in a corner and wait it out – maybe it will sort itself out. I can only do this when I'm alone. When I have a hungry family I have to look for another way. My second way is to use only well-known recipes and stick to them irregardless of how well I know them. Any experiments in this situation will end in failure.

      Last weekend was just difficult. My husband helped me prepare dinner, but the dessert was my problem alone. Following the rules, I used a recipe for napoleon that is so simple there is no way you could fail. I recommend it to anyone struggling with creative impotence or who likes glamourous results after not too much effort in the kitchen.
      Ingredients (for 9 napoleons)
      1 pack of chilled French pastry
      500ml of milk
      6 tablespoons of sugar
      1 packet of powdered blancmange
      50g of butter
      2 egg yolks
      1 teaspoon of vanilla essence
      1 tablespoons of potato flour
      2 tablespoons of flour
      caster sugar

      Heat the oven up to 180C. Cover a baking tray with some baking paper.
      Cut the French pastry in half. Bake one half for 20 minutes. Remove it from the tray. Cut the second part into 9 squares. A cake prepared in this way is easier to divide into portions. Put them on the paper and bake for 20 minutes.
      Now prepare the crème. Boil 400ml of the milk with the sugar, vanilla essence and butter. Mix the rest of the milk with the powdered blancmange, flour and potato flour and egg yolks. When the milk has boiled, take it off the heat and add it to the mixture, stirring constantly. Put it on the heat and boil, stirring until the mixture is coagulated. Take the pot off the heat. Put the warm mixture on the whole part of the French pasty and then cover it with the sliced part of the pastry. Cover the dessert with aluminium foil and leave in the fridge for a few hours. Cut and sprinkle with the caster sugar before serving.

    • By Kasia
      One of our friends said recently that he doesn't cook for himself. He eats what his wife prepares: sometimes it is something healthy and other times something yummy. It was a joke, of course, because his wife cooks really well, but this sentence is now in our friendly canon of jokes.

      Inspired by our talk about groats, flakes and healthy food, I prepared a dessert which combines excellent taste and healthy ingredients. The original recipe comes from the Lidl cookery book. I would like to share with you my version of this dish. I recommend Crème brûlée with millet groats to everybody who counts calories. It is mild, not too sweet, wonderfully creamy inside and with an incredible crunchy crust on top. That's why we love crème brûlée, don't we? I prepared a cranberry-orange preserve to offset the sweetness of the dessert. The whole dessert looked beautiful and tasted perfect.
      Ingredients (for 4 people)
      crème brûlée
      100g of dry millet groats
      350ml of almond milk
      1 teaspoon of vanilla essence
      2-3 tablespoons of brown sugar (3 additional tablespoons for the sugar crust)
      juice and skin from one orange
      150g of fresh cranberries
      juice and peel from one orange
      4 tablespoons of brown sugar

      Put the millet groats in a sifter, clean them with cold water and then douse them with hot water. Put the groats, almond milk, sugar and vanilla essence into a saucepan with a heavy bottom. Boil it with the lid on without stirring for 15-18 minutes until the liquid has evaporated. Leave to cool down. Add the orange juice and peel, mix it in and blend until the mixture is perfectly smooth. Put the dessert into small bowls and leave in the fridge for one hour. Wash the cranberries. Add the orange juice and peel and the sugar and boil for 10-15 minutes. Try it and add some sugar if you think the dessert is too sour. Take out the bowls from the fridge. Sprinkle them with the sugar and burn it with a small kitchen burner to make a crunchy caramel crust. Decorate the dessert with a small teaspoon of the cranberry preserve. Serve the rest of the preserve separately in small dishes.

    • By Kasia
      One of my friends from Ukraine told me about her traditional Christmas dishes. Except for stuffed cabbage with potatoes (which I have made already) I was surprised about cranberry kissel. I searched the Internet and I saw that in many Polish homes Christmas Eve supper ends with cranberry kissel. In my home we always drink compote with dried fruit, but maybe this year we will try a new dish on our Christmas menu.

      I wonder why cranberries are on the Christmas table. I didn't find any particular information about it (except the fact it is tradition). I think that a few years ago cranberries were treated as a natural cure which aids digestion, and this could be quite useful after a hefty Christmas meal!

      At my Ukrainian friends' home Christmas kissel is runny like a drink, but you can prepare it like a dessert with a more dense texture. I made the drink version, but you should choose which is better for you.

      500g of cranberries
      a piece of cinnamon and a couple of cloves
      6-8 tablespoons of sugar
      2-3 tablespoons of potato flour

      Wash the cranberries and put them with the cinnamon and cloves in a pan. Pour in 500ml of water and boil until the fruit is soft. Remove the cinnamon and cloves and blend the rest. Add the sugar and mix it until it has dissolved. Sieve the cranberry mousse to make a smooth texture. Mix the potato flour with a bit of cold water. Boil the cranberry mousse and add the mixed potato flour, stirring constantly so it is not lumpy. Boil for a while. Pour the kissel into some glasses.

      Enjoy your meal!

    • By Kasia
      The worst thing about my cranberry-apple cake is the way it looks. It didn't look impressive, but it was so yummy it disappeared from the baking pan before it had completely cooled down. My children said that it was a colourful apple pie, and it really was something like that. Apples with cinnamon are the basis of apple pie – one of my favourite cakes. However, the sour cranberries make it more fresh and interesting. The crumble topping was, for my son, the most important part of the cake. I had to drive him away, because otherwise the cake would have been deprived of its crunchy top.

      Ingredients (18×26cm cake tin ):
      200g of flour
      150g of butter
      3 eggs
      1 packet of powdered vanilla blancmange
      1 teaspoon of vanilla essence
      200g of sugar
      1 teaspoon of baking powder
      pinch of salt
      250g of fresh cranberries
      1 apple
      3 tablespoons of brown sugar
      1 teaspoon of cinnamon
      crumble topping
      5 tablespoons of brown sugar
      100g of butter
      150g of flour
      First make the crumble topping. Put the cool butter, flour and sugar in a bowl. Knead them until you have small lumps. Leave it in the fridge.
      Heat the oven up to 180C. Cover a cake tin with some baking paper.
      Mix the flour with the baking powder and salt. Cream the butter with the sugar. Add egg after egg to the butter, stirring constantly. Add the flour, vanilla essence and powdered vanilla blancmange. Mix it together until you have a smooth dough. Put the dough into the cake tin. Wash the apple, remove the apple core and cube it. Mix the cranberries, apple, sugar and cinnamon in a bowl. Put the fruit on top of the dough. Cover the fruit with the crumble topping. Bake for 50 minutes.

      Enjoy your meal!

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Create New...