The Beam Engine

Really cool site! Did you create the animations with videos of assemblies in a CAD package? Very noice!!!

This reminds me a lot of the articles at: https://ciechanow.ski/archives/

Not just in terms of presentation, but also quality.

Amazing!

Excellent work, glinscott. Thank you for sharing! I am about to send this to all my pals who are parents to schoolkids (um, and also pals who are still, like me, basically a schoolkid :p XD).

Building things up from first principles, and ideally also tactile / visceral examples, is my absolute favourite genre of explanations. Alas, a rarefied genre, because it is so much harder to do well (without doing disservice to the learner's intellect and ability) than expert talking to (presumed) expert. So thanks for contributing to it.

People like George Polya [0], Richard Feynman [1], Grace Hopper [2] are exemplars of this tradition.

[0] https://www.youtube.com/watch?v=h0gbw-Ur_do

[1] https://www.youtube.com/watch?v=EKWGGDXe5MA

[2] https://www.youtube.com/watch?v=ZR0ujwlvbkQ

(And I too try in my own meagre way through my blog and source code, albeit it's all plain text because visual explainers, animated ones that too are ridiculously difficult. So I have some idea of how much of a labour of love your piece is. <Claps hands, Whistles, Doffs hat>.)

Despite all the details, you still left out a lot. Safety valve, pressure gauge(s), regulation of the water pump, water level gauge for the boiler, lubrication (you only have two of the glass/metal can reservoir types for the main axle), and then maybe also the valve gear (the control mechanism for the valves which takes into account where in the cycle the machine is and maybe also how fast it is going, how fast it should be going, and in which direction). And a way to drain water from the bottom of the cylinder + maybe a way to make the steam as dry as possible.

And each of those could be expanded to much more than the page you already made.

Things are almost always a lot more complicated than they seem. It seems so simple -- "water + heat makes steam, steam pushes piston, back and forth motion makes rotary motion" -- but it is really anything but.

> This push on every wall is pressure, and we will measure it in atmospheres, multiples of the ordinary pressure of the air around us.

No, don't do that. Because the pressure of the atmosphere isn't constant. It varies with weather and how high you are.

Instead use Pascal (Pa) or Hectopascal (hPA). I really pays back to use SI units for physical things. Suddenly you have formulas without weird constants, like 1 Pa = 1 N/m^2

It's a sad thing that new articles still propose units that are (since many, many years) outdated: the metric system is from 1793, the SI is from 1960.

BACK then people didn't knew better and used the varying atmosphere as base. But today ?!?!?

This looks like a labour of love - fantastic

I don't have much to contribute to the conversation but I wanted to congratulate you on the work and thank you for sharing.

This is impressive work!

Thank you! I have a few ideas for the next article. Locomotives are up there on the list, but they are the final challenge of steam engineering, there is so much to cover.

Also considering the origins of precision - the intertwined evolution of machining tools and measuring systems that allowed these machines to be built.

RE: 3 year old - yes, mechanical systems are wonderful for children (and adults). There is something incredibly satisfying about watching them in motion doing work.

High performance "modern" steam engines are fascinating, they're not necessarily simpler than their internal combustion successors, just different. Monotube boilers create superheated steam that pushes 1200°F, which is right around the same temperature as the exhaust gasses in truck's diesel engine pulling a large hill at maximum power. Hot enough to melt aluminum save for the presence of oil jets to cool the piston and a jacket of cooling water surrounding the cylinder. However, since there's no combustion happening in the cylinder of a steam engine, you don't have any cooling water surrounding it--that would be totally counterproductive. So steam engines operate in a thermal regime that is far more extreme than internal combustion engines. Consequently, bores are often cut slightly tapered to account for the temperature gradient across the stroke, in a double acting engine the cylinder bore has a slight hourglass shape, otherwise the thermal expansion would cause the piston to bind at the ends.

For a fun rabbit hole to dive down, check out Doble steam car technology or the Besler airplane engine.

One of my treasured toys - probably from a few years older than your 3 year old though - was a rocking cylinder steam engine my father gave me. It was a present from his father as part of a steam powered boat they built together probably in the late 1940s.

Something similar to this: https://en.wikipedia.org/wiki/Oscillating_cylinder_steam_eng...

And you can still get them reasonably inexpensively: https://www.aliexpress.com/item/1005007964198565.html

I bought one for a friend's 5 year old a while back and became BEST UNCLE EVER!

The run well on compressed air (like an electric car tyre inflator or small air compressor), or for a bit more high temperature danger they run great attached to the steam wand of an espresso machine. When I was a kid dad built me a steam boiler out of a steal aerosol can and some copper tubing with a folded brass sheetmetal alcohol burner. That's not a 5 year old friendly thing though, from memory I'd have been maybe 10 or 12 when he started letting my light alcohol on fire in that garage or backyard. (Modern parenting may disapprove...)

Similar to balls to the walls (or similar). Coming from the analog aviation controls I believe.

tbh, any time i've said it i've been referring to testicles, but thanks for the helpful excuse in future.

Definitely a fun note :). I think I heard it first from a Jay Leno youtube video where he was showing his amazing collection of steam powered machinery.

This is from the article, but governors also played a role in kicking off modern control theory! """ In 1868, James Clerk Maxwell studied when these oscillations grow or die away in his paper “On Governors”[https://www.damtp.cam.ac.uk/user/gold/pdfs/teaching/old_lite...]. This became one of the beginnings of modern control theory. """

I suppose this is also the less colorful true origin of "balls to the wall" than the phrase itself suggests?

> When the engine slows, the balls fall and open the cock again.

The choice of wording makes me suspect that they intended both entendres. ;)

and i suppose “fuck” is actually an acronym for “fornicating under command of king,” right?

While fun and satisfying, this is extremely unlikely to be true.

The first attested use is from 1945, on the nose of a P-47 Thunderbolt flown by Captain Milton Thompson of the 509th Fighter Squadron. And his art was exactly what you'd expect: a charging bull with the words "balls out".

There is also a steam engineers term called "high-balling" which referred the little ball in the steam pressure gauge used in the early days. So high-balling was travelling at high speed.

What about the phrase "testicles out"?

That was a reaction engine, something like a steam rocket confined to rotate in a circle, and not a positive-displacement engine: https://en.wikipedia.org/wiki/Aeolipile

They couldn't use it as anything else than a toy. The reason they couldn't because they didn't have the metallurgy and the machining to create pistons, valves, etc.

IMHO they definitely had the processes to basically add a PTO to one, but it was probably dismissed as impractical.

> The reason they couldn't because they didn't have the metallurgy and the machining to create pistons, valves, etc.

It was a turbine. You don’t use pistons in a turbine. They lacked knowledge of turbine blades, the metallurgy to make pressure vessels, and the tools to make parts with the tolerances needed to spin at high RPM and withstand high temperatures.

100% - his articles were a major inspiration, and first line in the credits: - Thanks to Bartosz Ciechanowski, whose exceptionally well written and designed articles inspired this one.

Glad you enjoyed the units :).

You might enjoy How to Make Evertything on YT.

For this reason, I always point out that toolmaking was an essential and often forgotten part of Manchester’s Industrial Revolution.

One of the most conceptually simple ways to make a vacuum is a Sprengel pump. You have, from top to bottom, a pipe full of mercury, a restriction so it comes out in drops to the next stage, a chamber where the vacuum is developed, and an exit pipe. Mercury drops fall through the chamber, trapping air between them as they fall down the exit pipe. The energy comes from lifting the mercury back to the top.

A way that's easy to execute, but doesn't produce a perfect vacuum, is to fill a chamber with steam (actual high temperature steam, not the clouds coming off a kitchen pot), seal it and cool it. This is only a near-vacuum because it's filled with low pressure water vapor, but it's very easy to do.

There’s also the amazing Crossness, which does have occasional steaming days. And if you’re in Manchester the Museum of Science and Industry has a great collection of small and large engines which they run

Blondihacks is amazing. Beware though, you may be buying a lathe and a milling machine soon after you start watching :).

It's a great hobby. Manual machining is totally different from writing code, but you get into a similar state of flow.

See also Wreck2Restored, a channel that documents the ongoing restoration of a 2-foot gauge steam locomotive. Crossheads and sliding valves featured!

https://www.youtube.com/@Wreck2Restored

I'm only ~5 minutes into that first video and every single thing she has said so far rings so stupid true that its almost painful hearing it said so bluntly. Making things takes a lot of time, requires lots of fundamental skills, is expensive to do yourself, generally isn't going to be better than something other people are selling, and should be fun the entire time you're doing it.

This should be the comment that people read on this post. Watch these videos.

On the topic of steam-related machinery YouTube channels, Keith Rucker gets my recommendation: https://www.youtube.com/channel/UCyjwQ6oz4cqqtEcWGboSU3g

Wow, that Blondiehacks video opened my eyes to a whole unknown world. Thanks!

Thanks!

And https://modelengineeringwebsite.com/Classic_ME_beam_engine.h... is almost exactly the CAD model behind this beam engine. Designing that on paper would have been quite a challenge.

Yes, great details! Not sure if you've read it, but "Richard L. Hills, Power from Steam: A History of the Stationary Steam Engine (1989)." is a truly wonderful overview, that really digs deep into the business. It has many letters directly from Watt, including the problems they had getting these huge engines built (before any decent roads!).

why are you badly summarizing the post?

It wasn't a vacuum engine, as per post the water just cooled the cylinder. And it wasn't half of the coal, it was a third.

When I clicked on the link, I first thought it was going to be about, well, steam engines. Then I realized this is Hacker News, so it's probably about the Erlang VM. Turns out my first thought was actually the correct one!

Glad to not be the only person to expect Erlang instead of steam.

Yes, I also got briefly very confused about the top comment.

Awesome! It's pretty amazing all the concepts that need to come together for even these very old engines. I found I built a better understanding of what was going on as I built out the animations as well.

Thanks!

RE: units - Good catch - there is almost certainly a better way to phrase this one. I always tried to give physical examples to make things more concrete. But who knows, maybe the bathtub will take off as a unit :).

Those damn imperial measurements.

Oh my. Yes, you are totally right. This was tricky because I needed to merge two different animations (the governor moves independently from the rotation of the wheel), and I seem to have messed that up.

Fix shall be deployed shortly!

It was a mix of techniques, this article has been in progress for quite a while :). The first part was modeling all the relationships for an accurate beam engine in onshape (great cad tool). Then, I built a custom exporter to cleanly export it to three.js. This ensured physical realism for the hero of the article. All of these are linked from the article.

For the figures, I had prototyped some by hand, and it was slow going. Fable was the first LLM I found that was actually able to understand and model mechanical linkages correctly enough to be useful. I built a style guide, and iterated many times on each figure and animation until we arrived at something useful. There are also quite a few backing tests to ensure physical correctness of the animations (as far as reasonably possible :).

My uni had an original 1800s style beam engine (from a whiskey distillery) as a feature in the Engineering faculty building foyer. It did move, but driven by electricity, and only a couple of times a year AFAIR.

https://forum.irishwhiskeysociety.com/viewtopic.php-p=12194....