I love pneumatic logic! I’ve gone to sleep many nights thinking about pneumatic logic gates, and they are still used today on pneumatic equipment powered by wind, like a lot of pipe organs in Europe made in the 1800’s where there was no electricity to turn on and off the pipes.
Re:Basics of Pneumatic Logic
by Animats on Monday June 15, 2020 @09:17PM
(#23534972)
Yes, some pipe organs had enormous switching systems. They didn't do all that much; they just switched the keyboard to pipe rank connections. Physically, they were huge, like multiple upright pianos. Some needed compressed air for power; others were entirely electrical.
They had no direct function in producing sound. All this was separate from the actuators for the pipes.
Re:Basics of Pneumatic Logic
by fringleot on Tuesday June 16, 2020 @02:21AM
(#23536429)
Still handy in ATEX rated zones
EDIT: a little more context..
ATEX rated zones are where there may be or always is an explosive atmosphere, this makes computer controlled systems an interesting challenge.
IMHO control systems are all distributed systems and from a creator centric perspective there is something exceptionally satisfying when part of the distributed logic and state management is in pneumatic valves.
IANAL or expert but it is also an ingenious way to get around IEC 61508 complexity. When I worked on control systems priority one for safety concerns was making logic physical (not software at all) as it is very very hard to prove software running on an OS is in any way safe - while engineering diagrams with know failure modes etc was deemed preferable (by our team at least).
Re:Basics of Pneumatic Logic
by arethuza on Tuesday June 16, 2020 @06:21AM
(#23537511)
Many year ago, I worked on a project providing training simulations for the Reactor Shutdown Safety Sequencing Equipment (RSSE) of a UK AGR power station - the project manager had been involved in the construction of the plant and explained that the equipment involved actually used mechanical sequencing essentially based on clockwork as it was designed using technology that was really trusted when the AGR plants were first designed in the 1960s - which basically meant 1940s level technology.
Edit: Pretty sure I have the words in RSSSE correct though perhaps not in the right order.
Re:Basics of Pneumatic Logic
by jessaustin on Tuesday June 16, 2020 @12:37AM
(#23535986)
If you've ever been to the dentist, you've probably been pretty close to pneumatic logic. Every delivery unit I've seen has used it. (Power a handpiece only if it's not in its holster, spray water only when the water switch is on and the handpiece is turning, etc.) Fixing them is a pain. You have to have just the right rubber sheets, gaskets, o-rings, etc. Also they're just a block of metal with a bunch of hoses attached. It's impossible figure out what an individual valve does just by looking at it.
Re:Basics of Pneumatic Logic
by VBprogrammer on Tuesday June 16, 2020 @04:56AM
(#23537164)
I believe very similar control circuitry is used in aircraft landing gear hydraulics. For example to open the landing gear doors before extending the legs.
Re:Basics of Pneumatic Logic
by quickthrowman on Monday June 15, 2020 @10:24PM
(#23535374)
There are still pneumatic controls operating HVAC equipment all over the place. A lot of it is being replaced with electronic temperature controls as the service life of the existing HVAC equipment comes to an end and it gets replaced. Facilities are now looking to interface HVAC temperature controls in with wider building automation systems.
The only context I see pneumatic controls in is a system to be demolished and replaced, so it’s neat to see how they work and how they’re functionally equivalent to electronic logic gates.
Re:Basics of Pneumatic Logic
by meditative on Monday June 15, 2020 @11:20PM
(#23535641)
There's no remote monitoring, they're more complex and expensive. You need either multiple compressors, or run long stretches of hose/pipe.
They're super cool and rewarding to work with, everything feels like steam punk, but there's no real practical reason to keep them there. The electronic stuff is _very_ reliable when installed correctly and quality parts are used, much like the pneumatic systems.
The only place that _might_ make sense to leave them in a little longer is inside a hazardous environment, which requires detailed certification for installed electrical equipment.
Re:Basics of Pneumatic Logic
by ganzuul on Tuesday June 16, 2020 @05:15AM
(#23537244)
You should test that shielding though, and AFAIK the dangerous voltage enters equipment through cables. Unless everything is fiber optics you would need some chonky optocouplers on all signal paths.
Pneumatic air is usually supplied with plastic tube and hose. There are no unexpected electrical circuits coupling to unrelated equipment.
Re:Basics of Pneumatic Logic
by ganzuul on Tuesday June 16, 2020 @09:31AM
(#23538923)
I think the problem is one of bootstrapping. With PLCs controlling even sewage pumps society will have a lot of more important things to do before we come around to fixing elevators, automatic doors, or even chip fabs. Then there are herds of forgotten VAXen and AS400s that have been chugging along for decades running who knows what mission critical code, all gone without as much as specs remaining.
The less basic infrastructure which is wiped out the faster we can rebuild. Right now it looks like we will be falling back to whatever cam-actuated automatic lathes which have not been sold below scrap value that are left. And I hear those things are a bitch to set up.
Re:Basics of Pneumatic Logic
by seanmcdirmid on Tuesday June 16, 2020 @12:22AM
(#23535918)
I don’t know. Pneumatics involve a lot more friction than a digital system would. That regular maintenance would probably involve replacing things as they were worn down.
Re:Basics of Pneumatic Logic
by twirlip on Monday June 15, 2020 @11:07PM
(#23535561)
Oh, I've only read about pneumatic logic in Ted Chiang's "Exhalation" and Neal Stephenson's Cryptonomicon. How neat to read about actual gates and their functioning.
Re:Basics of Pneumatic Logic
by rypskar on Tuesday June 16, 2020 @03:54AM
(#23536880)
This brings back memories. The first "programming" I was exposed to at school was using pneumatic. I remember one task was to construct a system to act like the doors on a buss, where one button is used to both open and close the doors. Working with real components make the logic so much more intuitive.
Re:Basics of Pneumatic Logic
by ganzuul on Tuesday June 16, 2020 @05:19AM
(#23537260)
Lots of maintenance guys are also exposed to boolean logic through pneumatics. Expectations are not super high but surely some students discover a fondness for logic this way.
Edit: Reminds me, Festo makes some really unexpected and creative stuff but ostensibly they produce pneumatics. I would really like to know what kind of culture is behind this creative bent of theirs and if it is related to logic.
Re:Basics of Pneumatic Logic
by rypskar on Tuesday June 16, 2020 @06:41AM
(#23537608)
It is incredible what some used to do, I once used a day or two trying to document an old logical system that was using 20-30 relays and contactors by tracing the wires. Did end with giving up and had it rewritten in PLS-code. The old system was created a long time before I started working there by someone who where good at that kind of work and never learned to code in PLS.
I don't know how it is done now, but at least 20 years ago more advanced systems where expensive and creating something using logical gates was what we where used to, so used that for simple systems. In some ways creating something using logical gates and get it to work is like doing something simple with command line tools instead of writing lots of code to do it
Re:Basics of Pneumatic Logic
by Koshkin on Tuesday June 16, 2020 @08:34AM
(#23538368)
Apparently the idea of "jet logic" as a useful substitute for electronics in some situations (hence dubbed "pneumonics"), gained some popularity in Russia back in the '60s. Here is a moment in a fascinating educational video where they explain the workings of a jet NOR/OR gate:
Re:Basics of Pneumatic Logic
by amelius on Tuesday June 16, 2020 @06:06AM
(#23537440)
Downsides:
Mechanical wear, leaks, difficulty to put a pressure probe at various points in the circuit, slow speed, size, complicated 3d geometry, need for compressed air, audible noise, ...
Here’s an example of such apparatus: https://www.woodofhuddersfield.com/uploads/1/1/2/3/11238427/...
They had no direct function in producing sound. All this was separate from the actuators for the pipes.
It replaced mechanical track action, keyboard coupling, register switching and even had memory to ‘remember’ a register presets or register sets.
This document has some designs and more details: http://www.gutenberg.org/files/21204/21204-h/21204-h.htm , chapter III.
https://hackaday.io/project/45555/gallery#5233f682e2114006e2...
https://en.wikipedia.org/wiki/Fluidics
Also worth checking out Bubble Logic (PDF warning: it's an MIT PhD thesis):
http://cba.mit.edu/docs/theses/08.09.Prakash.pdf
EDIT: a little more context.. ATEX rated zones are where there may be or always is an explosive atmosphere, this makes computer controlled systems an interesting challenge.
IMHO control systems are all distributed systems and from a creator centric perspective there is something exceptionally satisfying when part of the distributed logic and state management is in pneumatic valves.
IANAL or expert but it is also an ingenious way to get around IEC 61508 complexity. When I worked on control systems priority one for safety concerns was making logic physical (not software at all) as it is very very hard to prove software running on an OS is in any way safe - while engineering diagrams with know failure modes etc was deemed preferable (by our team at least).
again: not a lawyer or expert :)
Edit: Pretty sure I have the words in RSSSE correct though perhaps not in the right order.
The only context I see pneumatic controls in is a system to be demolished and replaced, so it’s neat to see how they work and how they’re functionally equivalent to electronic logic gates.
They're super cool and rewarding to work with, everything feels like steam punk, but there's no real practical reason to keep them there. The electronic stuff is _very_ reliable when installed correctly and quality parts are used, much like the pneumatic systems.
The only place that _might_ make sense to leave them in a little longer is inside a hazardous environment, which requires detailed certification for installed electrical equipment.
Pneumatic air is usually supplied with plastic tube and hose. There are no unexpected electrical circuits coupling to unrelated equipment.
The less basic infrastructure which is wiped out the faster we can rebuild. Right now it looks like we will be falling back to whatever cam-actuated automatic lathes which have not been sold below scrap value that are left. And I hear those things are a bitch to set up.
They are also less sensitive to heat than electrolytic capacitors.
(Disclosure: I'm on the team that built this site)
Edit: Reminds me, Festo makes some really unexpected and creative stuff but ostensibly they produce pneumatics. I would really like to know what kind of culture is behind this creative bent of theirs and if it is related to logic.
I don't know how it is done now, but at least 20 years ago more advanced systems where expensive and creating something using logical gates was what we where used to, so used that for simple systems. In some ways creating something using logical gates and get it to work is like doing something simple with command line tools instead of writing lots of code to do it
https://youtu.be/gzGRLdB0BpE?list=PLgqu8dCNRlmX_d0Oz3T6gRTyw...
https://escapepod.org/2009/04/10/ep194-exhalation/
https://traffic.libsyn.com/escapepod/EP194_Exhalation.mp3
16:22 - debugging self own pneumatic brain
How this kind of logic works with reverse airflow? (Would it be like time going back when Inhalation?)
Mechanical wear, leaks, difficulty to put a pressure probe at various points in the circuit, slow speed, size, complicated 3d geometry, need for compressed air, audible noise, ...
I've commented on HN before that Ted Stevens attended engineering school in the heyday of pneumatic machines.