Close

Construct the trial piece

A project log for 3D Print Babbage's Difference Engine No.2

A fully functional replica of Charles Babbage's Difference Engine No.2 with 3D printed parts

mechredpandaMechRedPanda 11/27/2023 at 05:330 Comments

Welcome back, everyone, to the fourth episode of the Difference Engine Replication Project with Mechanical Panda! In this segment, I'll share the journey of creating a prototype, aiming to validate the principles behind the difference engine using 3D printing technology.

The main objective behind crafting this prototype was twofold: to test the strength and tolerances of 3D-printed parts and to validate the rationality of the design blueprint. Think of it as a preliminary research project for the grand-scale difference engine. Although small in scale, every structural aspect and layout of this machine mirrors the full-sized difference engine.

A trial piece constructed by the London Science Museum before they constructing the full scale replica.

One of the first hurdles in replicating the difference engine was sourcing the design files. You might ask, why not simply replicate Babbage's blueprints? Well, Babbage left behind a couple of "trapdoors" in his design. Firstly, his drawings lacked any annotations or explanatory notes. His intricate designs remained cryptic for over a century, further complicated by his unique symbol system designed for the machine's control sequence. Additionally, Babbage unwittingly designed a significant bug in the machine, mirroring the blunder of reversing the rotation direction of an entire set of figure wheels.

Thankfully, the Science Museum in London guided me through these pitfalls. Their comprehensive 400-page analysis deciphered Babbage's original drawings and provided a fix for this bug. Unfortunately, their restored blueprints are not publicly available, and the difference engine's original design materials were based on brass and cast iron, vastly differing from my 3D-printed plastic replication. These differences in material strength and tolerances necessitated a complete redraw of all components.

The centerpiece, the figure wheel, determined the entire machine's size. After several trials, I settled on a 60mm size for the figure wheel, allowing the maximum dimensions of the test engine to be 240mm, perfectly suiting my printer's capacity. The choice of a 20mm stainless steel tube as the driving shaft for the figure wheel prompted a recalibration of all other parts’ sizes and positions.

Despite encountering conflicts in parts' motion paths during assembly due to oversights in the blueprint, redesigning through 3D printing offered advantages. The original design required multiple independent parts and subsequent welding, introducing assembly errors, which 3D printing's integrated manufacturing process effectively mitigated. Every part of this prototype, excluding the driving shaft and nuts, was meticulously crafted using a FDM 3D printer, with a total printing duration of around three days.

After nearly two months of dedication, I finally completed this youthful rendition of the difference engine. However, limited by its eight figure wheels, this "Youth Edition" can only store two four-digit numbers, capable of computing only linear functions. Nevertheless, this prototype's construction validates the feasibility of replicating a difference engine through 3D-printed parts, meeting the mechanical strength and precision required for its operation, albeit with alterations to the original design.

Yet, this prototype merely marks a small milestone in the replication of the difference engine. Due to its limited number of figure wheels and lack of a control system, I am already designing a more comprehensive model. If you’re intrigued by the journey of this replication project or keen on following the progress of my ongoing endeavors, don’t forget to subscribe!

Until next time, this is Mechanical Panda bidding farewell for now!

Discussions