We have started implementation of our own 3D printer, the SlantBox. This 3D printer has been customized for the rigors of production.

The SlantBox is the most rugged and reliable 3D printer in the world. It is meant to be deployed by the thousands inside of future 3D printing farms.

The key capabilities of the printer are is modularity and intelligence. Each printer can be assembled quickly and repaired equally quickly. But most important is the intelligence on-board. Each printer has a full computer on-board which networks with our cloud software so that printers can coordinate production jobs. The printers also have on-board cameras that will soon be used in conjunction with machine learning to monitor and improve print reliability over time.

Regarding reliability. With just the basic design of the SlantBox we have improved print reliability from a traditional 50-60% to over 90%. And we have made each machine self-tending so there is no need for external workers or robots to remove parts from the build platforms. This truly makes the printer viable for production.

The SlantBox and the infrastructure around it is going to fundamentally shift the capabilities of 3D printer farms to produce at scales competitive with processes like injection molding.

If you look around for a little while at new robots coming out, whether they be consumer telepresence robots like Ohmni, stem robots like the LittleBots, or a high performance robot arm like Dexter, they are all 3D printed. This is a trend that has really started gaining ground over the last few years. Robots are heavily utilizing 3D printing to make their plastic components for production. But why? Why are robotics start-ups manufacturing with 3D printers as opposed to injection molding.
​
Well the first reason is start-up cost. Robots are generally pretty complex machines. They often have dozens, if not hundreds, of individual parts. If a company were to create molds for all of the plastic parts on the robot it could cost hundreds of thousands of dollars. 

​Lets take an example. Look at the Dexter, open-source, high precision robot arm. They have approximately 50 parts on the robot. Assuming the average cost of a mold for each part was $3000 (though it likely would be higher as the parts are very complex) Then Haddington Dynamics, the company that produces  Dexter, would have to produce $150,000 just to get the molds. Then there is the cost of the parts themselves. There is also the time lost in perfecting the molds and the money lost from mistakes in mold design and production. It becomes very expensive very fast before they are even producing a product to sell.