I wanted to have a gas cylinder that would hold up to about 10 litres of gas (oxygen, mainly) at 20 atmospheres of pressure. Or, if that's impossible, 7 atmospheres would suffice.

The setting is pre-modern, so late medieval tech at best. I'd like the lowest tech solutions possible, but anything from the 16th century would be fine.


These don't need to be high quality or particularly safe gas cylinders (early boilers were plenty dangerous), but they need to basically work. They don't need to be precise or scientific in the exact pressures and measurements, either, but could work in the longstanding tradition of playing it by ear, where journeymen learn traditions from previous masters. They would also need to be able to exchange gas between canisters, using smaller ones to fill bigger ones.

Generating Pressure and Gas

I actually have the details worked out for how they can generate pressurized gasses. The question I have is how they contain it at those pressures. I can give additional details if necessary, but you can likely guess, if you've seen my other questions of late.


If they could get suitably pure copper, it has more than enough tensile strength for the task... but I don't know enough about the development history and purification of copper. Bronze is stronger, so it may be better. Since copper and bronze can be cast much more easily, they may be much easier to form into canisters.

Early boilers were made out of many small, copper, riveted plates, since copper withstands heat better than bronze. But the heat combined with the riveted construction made it limited in the pressures it could withstand (something like 500 psi was commonly handled, all the same). When they developed larger plates, the boilers got a lot safer.

This suggests several advantages for our canisters. Since they only need to contain a few litres of gas/liquid, instead of huge amounts of a boiler, they can be made possibly out of a single mould hammered into shape, instead of many small plates. Since it isn't a boiler and doesn't need to handle high temperatures, we can use bronze instead of copper. And since the heat from boilers is said to have weakened the copper considerably, we could assume our bronze canister can handle much greater pressures.

Notably, thickness makes an exponential difference to strength, so presumably what they lack in pure copper and flawless bronze alloys, they could make up for with thicker, heavier canisters? Of course, some merchants will use thinner, cheaper, less safe canisters, as is good for a story/setting.

Valves and Pressure Regulators

I had trouble finding information on the historical development of these, aside from very broad strokes. I don't need highly tuned pressure regulation, but there will need to be valves that allow for gross adjustments of pressure. It is also necessary that it's possible to refill one canister with another.

So, within the constraints of the setting, would it be possible to produce suitable valves?

  • $\begingroup$ canisters are easy, they could make cannons after all, it is the valves that are tricky. What do the valves need to do. are looking for emergency relief valves or just a gate valve. the romans could make bronze stock cock valves. we need to know what they are trying to do with the valve. $\endgroup$
    – John
    Dec 28, 2020 at 15:44
  • $\begingroup$ may be helpful valvias.com/history.php $\endgroup$
    – John
    Dec 28, 2020 at 15:45
  • $\begingroup$ What date or technology watershed cut off constitutes "pre-modern"? $\endgroup$
    – Slarty
    Dec 28, 2020 at 17:42
  • $\begingroup$ @John EPR in this case is probably just a thin plate over a round hole, if they find it necessary. A gate valve may be enough, but some kind of rudimentary pressure regulator may be necessary. I mostly described what I needed them to do, above. I was thinking they might need to express 90 psi of pressure in a jet of air, for about 1 second. They'd also need to be able to express low psi, so that they don't use up the canister too quickly... so some simple form of regulation is necessary. The other needs I think are listed in the question? Let me know if you need anything else. $\endgroup$
    – Johnny
    Dec 29, 2020 at 6:48
  • $\begingroup$ @Slarty The Modern period is usually considered to start in the 17th century. In this case, I was hoping to use pre-17th century tech. $\endgroup$
    – Johnny
    Dec 29, 2020 at 6:49

3 Answers 3


Is an invention from 1679 sufficiently Pre-Modern for you?

Denin Papin's Steam Digester, invented 1679, could safely handle 6 Bar(for boiling temp of 165C), and supposedly would rupture at about 15-20Bar, if the pressure release valve malfunctioned.
Supposedly, the action of this pressure relief valve led to the first concepts of steam engines! When near the release pressure, it would open, release some pressure, and close again, quite energetically.

Material: Cast iron body, steel & brass fittings. picture of a very small version from here, i think usage here is within the licensing. https://www.scienceandsociety.co.uk/results.asp?image=10182360&wwwflag=2&imagepos=1
enter image description here

  • $\begingroup$ It is pretty close, and pretty close to what I wanted. I don't guess there's any reason it couldn't be made with cast bronze instead of iron. Being able to hold up to 15 bar is also quite suitable. Thank you once again, PcMan. Interestingly, it seems to keep the valve closed with a simple weight on a lever? $\endgroup$
    – Johnny
    Jan 1, 2021 at 7:28
  • 1
    $\begingroup$ @Johnny conical hole, conical needle in the hole. Adjustable weight on lever-arm, pressing the needle down with known pressure. Primitive as heck, but it works. Inventor was cheapskate and used cylindrical hole and near-cylindrical needle, and it acted as a piston! $\endgroup$
    – PcMan
    Jan 1, 2021 at 13:07
  • $\begingroup$ Seems a good application of the KISS principle. $\endgroup$
    – Johnny
    Jan 1, 2021 at 14:06

I think the answer to your question is as simple as could.the civilization manufacture a cannon. If yes, the same metalurgy will easily contain 49atm being the liquification pressure of oxygen and hence give you todays storage densities.

the weight will of course be more than simply containing at atmospheric pressure

Cannons were never particularly safe though

This answer is for Oxygen, other gases have different liquification pressures

  • $\begingroup$ Thanks. Cannon have been around for quite a while, so that should be possible for them to make the equivalent of a small bronze cannon. That does leave the question of valves, of course. $\endgroup$
    – Johnny
    Dec 29, 2020 at 6:50

I think that the best example I would look at would be the early fire engines (steam engines) used to drive pumps in the Cornish tin mines. The issue that they had (and issue that you will also find) is that they were not very efficient due to leakage. However, these did not come about until the end of the 17th century so are probably too late for your needs.

Although cannons did contain pressure, there are two main issues with them in this context. How do get the gas in and how to keep it in there once it is in. The point of cannons is that the gas is entered into the cannon in powdered form (i.e. as gunpowder that is then burned) and then you really don't want to keep it in there once it has turned into gas.

One last thing to think about is that the ancient Greeks did come up with a type of steam engine, the Aeolipile (written about by Hero of Alexandria). It is therefore realistic that this could have been developed, but it was just not.

Valve wise, there is not much evidence of anything that I know of that would have been able to reliably hold gas under pressure, esp if you are keen on there not being leakage. So theoretically, you can have a cylinder that could contain gas at a certain pressure, but there is not really any way to get that gas in or out without it leaking all over the place.

  • $\begingroup$ Welcome Galactic. You make good points. My idea for containing it was to have a check-valve, using a copper ball and a rubber o-ring. You can then press on the ball, via lever, in order to open the valve. With the ancient steam engines, the Egyptians had one too, which allowed a statue to be raised up by the heat of the sun. As far as I'm aware, the materials were too limiting to make practical use of these until the late medieval era, where we begin to see boilers. $\endgroup$
    – Johnny
    Jan 1, 2021 at 7:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .