The CandyFab did 3-D printing in sugar by blowing hot air onto the sugar to melt it, a process that requires delicate temperature control to avoid caramelizing the sugar all the way to carbon foam.
But carbon foam is in some ways a more useful material than sugar. It has a much higher strength-to-weight ratio, it's vastly more heat-resistant, it's less dense, and often it can be prepared as a conductor or an insulator depending on processing temperature.
In addition to hot air, you can also convert sugar to carbon foam by dripping or squirting concentrated sulfuric acid on it, every high-school chemistry teacher's favorite scary chemical reaction. In a powder-bed 3-D printing process like that used by the CandyFab, this permits you to selectively deposit carbon foam in a sugar powder bed. This could be useful for a couple of different reasons: first, you might have fillers that are sensitive to the heat needed to melt sugar, but not to sulfuric acid, perhaps styrofoam beads or something similar; and, second, you might be able to inject the sulfuric acid more precisely or more quickly than you can inject the heat. In Needle binder injection printing I outlined a variant of this process that would work well with sulfuric acid as the "binder" being injected deep within the powder bed.
A third possibility is injecting a susceptor, then "baking" the whole powder bed with microwaves or with a dielectric heater; for example, vegetable oil should be a sufficient susceptor to dehydrate sugar in a domestic microwave oven, but other candidates include silicon carbide, graphite, and magnetite. These should work with other kinds of powder-bed processes that require post-heating as well, like those described in 3-D printing by flux deposition, enabling more rapid heating of large powder-bed-embedded objects than can be achieved by heat conduction alone.
Possible fillers for this process are highly varied, and they can be selectively deposited in the powder bed, as fluxing agents are in the process described in 3-D printing by flux deposition. The simplest is powdered, sieved coke, which will simply produce a denser carbon foam and costs US$0.70/kg, according to Likely-feasible non-flux-deposition powder-bed 3-D printing processes. The cheapest is silica sand for construction, US$0.012/kg; carbon foam ought to stick well enough to that. An alternative to sugar that similarly foams up by dehydration, and sticks fabulously to silica, is dried sodium silicate (waterglass), US$1.10/kg. Copper or brass (US$4/kg) could form conductive traces; steel wool or glass fiber (US$6/kg) could provide tensile strength.
Sugar, being water-soluble, can also be used as a binder simply by squirting a bit of water onto it; but it will virtually never dry out by itself at room temperature --- you'd have to bake it.
(A quick stovetop experiment shows that granulated table sugar, when heated to dehydration, is able to bind construction sand together at about 25% sugar, but not at about 10% sugar. Presumably this depends not only on the quantities of sugar and sand but also their grain size distributions. Even at 25% sugar, the mass is quite crumbly; it disintegrates with a touch. 50% sugar is quite a bit more solid.)