You may have heard of 3D printing and rapid prototyping. Have you seen videos of concrete printers? We aren’t yet building apartments this way, or even backyard sheds, but the idea has been picked up and its development is now funded by very large companies.

I was thinking about this while spooning up a dish of Ben & Jerry’s 30th anniversary celebratory ice cream flavor, Cake Batter. I wondered what machinery had achieved the perfect swirling of ice cream flavors and gooey frosting. I have toured their factory in Waterbury, Vermont, and I even stood nearby while it was being built, but I’ve never seen up close the twirling, swirling nozzles. I wondered about their capabilities.

We had a 9-pin dot matrix printer when I was a kid. Eventually we got a 24-pin model that printed nicer-looking fonts, followed by a laser printer and a color bubble jet. I disassembled each one to see how it worked, or I read about it if the mechanism was too small. I found that if you ran a page enough times through any printer, the ink would build up on the paper until eventually the printer jammed. 3D printing is easy to imagine but not so easy to do.

Now if Ben & Jerry’s ice cream swirls could be created in a pint of ice cream by moving a rod having individually-controlled swirl ports, you might be able to print messages inside the pint, to be successively revealed by spooning away layers from the top. This has the obvious difficulty of the effect of the rod moving through the ice cream, swirling into oblivion whatever was deposited before.

It would be better to squirt the ice creams and swirls simultaneously into the pint to form one layer at a time. Having on its pint-wide print head a grid of nozzles arranged horizontally, the machine would print while moving the head vertically through the length of the pint. If it had the ability to control the nozzles to squirt different ingredients, the machine would be a 3D ice cream printer and my dream come true.

All of the forgoing occurred to me between bites of Cake Batter ice cream. These thoughts occupied a small part of my mind through the day until I began to think back on older printing technologies. Most interestingly, all of the laser printers I had seen contained a heating element called a fuser which baked the toner that had been deposited on the paper. What if you could automatically deposit cake batter and bake it by layers? You could extrude a fully-baked cake with a different birthday wish in every slice.

One mechanism appeared to me: tiny, retractable, non-stick heating elements would extend into the printed batter at the rate of print head advance, effectively staying put in the batter until it has finished baking and then retracting into the print head to be deployed again. Another idea involved inflating tiny balloons with hot liquid to bake the batter and then deflating them, mimicking the natural consistency of cake with the voids left behind.

I imagined that the resolution of such printing would be directly limited by the composition of the cake. Air bubbles are transparent holes that transmit color from the solid parts at their boundaries. Higher resolution would depend on smaller bubbles. Up to this point I was thinking primarily in the visual field. Bakeries and consumers already have access to frosting printers, which are simply Canon bubble jet printers with edible sheets of “paper” and edible inks, but this is 2D printing and not what I wanted to think about. Continuing on the 3D track, I thought separately about flavor printing and structure printing.

Flavor printing can be imagined very simply in one dimension by equipping a sausage mill with computer-controlled nozzles, each one adding a different ingredient to the filler and being turned on and off to create a pattern of flavors that the eater would experience in a sequence of bites. It is easy to expand the concept mechanically into two and three dimensions, printing out multi-flavored cakes or gelatin molds. I don’t know what consumer need this would fill. Still, I’d like to see it done.

Structure printing (creating, say, a frosted yellow cake in the form of the Statue of Liberty) brings up an important consideration: the physical properties of the food. Where are the compression strain diagrams of yellow cake? What is the tensile strength of frosting with and without coconut? How best to anneal the chocolate rebar?

If you know any mechanical engineering grad students who secretly wanted to major in home economics, or vice versa, please send them a link to this post. I would love to consult on any project that stems from these ideas.