The Worst Machine

The two-stroke engine.

>>7647915
Most human brains.

>>7647915
ur mums semen chute

>>7647915
It would be the one with the most moving parts desu

the space shuttle

>>7647915
just about any consumer grade printer

hell, printers in general

>>7647915
Do machines like that come with emergency shut off switches behind a breakable glass box?

Like if someone's hand gets sucked into it or whatever? I always liked the way they looked, seems like whoever actually got to shut off a machine with one of those would feel like a hero

but the other guy's hand is still fucked for sure i'm just saying

>>7647918
2 stroke engines may be inefficient but they are simple as fuck and have a huge output within a narrow rev range.
Unsurpassed for things like chainsaws.
Not much to repair on them unless it's a race bike where you strip it down every season, sleeve the thing, new piston, rings etc, brand new agian. Takes a few hours with an off the shelf kit.

The Wankel rotary is a piece of shit. Nice idea but the apex seals are never going to be reliable and there is no way around the problem.

I'm in maintenance & diagnostics and I could name many machines that are really bad but they are one off, purpose built things. Log de-barkers, things like that.
Someone can always improve on something, eliminate a problem or build something better unless it's an engine concept, most of which never amount to anything.

>>7647960
It's probably just a red button somewhere along with that laser curtain bullshit that trips whenever you make a measurement.

>>7647960
If there are people working in the vicinity then yes there will be emergency stops everywhere.
Automated plant like this have no humans around to mess things up.

>>7647960
LOTO

Lock-Out Tag-Out

look up /diy/ archives of stories of people working around heavy machinery who didn't lock-out tag-out

production code I write at work ;^)

File: 2stroke_diesel.gif (29KB, 124x239px) Image search: [Google]

29KB, 124x239px

>>7647918
Large two-stroke Diesel engines (truck, locomotive, marine) are some of the most efficient, most reliable machines in history, you mong.

Combustion engines

AN/GRC-211
fuck that fat, narsty bitch

File: $T2eC16RHJGkE9no8hm5lBP9eehnESQ--60_35.jpg (27KB, 284x300px) Image search: [Google]

27KB, 284x300px

>>7647960
almost every production machine has one of these or something exactly like it

>>7648022
Where's the spark to start the explosion in the cylinder? is it the green things on the corners?

>>7648069
>spark
>diesel

silly anon

>>7648069
>>7648073
although, to answer your question, diesel don't use a spark to start, they just need pressure in the chamber

this can lead to a phenomena called "runaway diesels" where if a seal fails it will literally eat it's own oil and go nuts on you
there are some really funny - or scary - .gifs of this, i wish i had some to show you
i'm sure you could youtube runaway diesels

>>7647952
3d printers are way way less reliable. Some large volume 3d printer operators even sacrifice small animals to appease the gods of 3d printing.

>>7648022
Detroit power!, awesome engines and really interesting how they work

>>7647915

This thread reminds me of my factory days. We were a growing mom-and-pop shop, operating on anything from 90-years-old-to today equipment. The bulk of the equipment had husks from the 30s, 40s, 50s, 60s with regular maintenance around the business ends and work centers.

We (I don't work there anymore but I still identify with the place) never had more than a week's worth of key parts, owing to successful and rapid account closes on the sales side. Running out of any one specific part would put that finished item (or any associated item) dead in the water until a fix was in. I spun the plates for a while, understanding the product line from a bottom-up (pleb) perspective that managers never stuck around long enough to internalize. In time, I left too.

Manufacturing theory quickly butts up against bottlenecks, which I experienced in detail (to OP's point about efficiency). Probably the most "finicky" machine in this joint was a smaller, extremely loud die-press that would knock out various essential parts on "progressive dies", meaning the work is formed in one spot on the die, then advanced to another spot on the same die for more molding, cutting etc. This machine itself was quite efficient once it was up, but it had to do parts for everything else so it was never fast enough, and there was always die trouble (one newer operator had a rabbit's foot, so I unscientifically liked when he ran. He didn't last though).

>>7647915
A first generation Landrover Discovery.

I challenge anyone to beat this.

>>7648608
Tunnel boring machines

>>7647915
The Jumo turbojets in the Me262

EM drive

File: 6035087765_26fe1d6a17_b.jpg (184KB, 958x728px) Image search: [Google]

184KB, 958x728px

>>7648117
>one newer operator had a rabbit's foot, so I unscientifically liked when he ran. He didn't last though
Do you mean like a good luck charm? And what does science have to do with him running?
I'm so confused.

File: dash-pre-order-03-02-2010.jpg (36KB, 600x228px) Image search: [Google]

36KB, 600x228px

>>7647915
Interesting question.

So many undefined metrics.

There is that pointless box machine that's on function is to turn itself off. But arguable it has decent entertainment value.

There are many older machines people have bad mouthed for constant maintenance and repairs, but economically speaking the ability to fix them on site is actually a real asset. Old mil spec was respected for a reason despite how dated it is.

I got a Sony Dash a few weeks ago for my birthday. I kind of respect it for it's vision and forethought, but the execution failed so hard I question if it was a hundred dollar gag gift.

Despite all the thing running through my head at this time I say the Sony Dash is the winner.

>>7648698

Some operators were just plain better than others, and better at diagnosing their own equipment and preventing failures/downtime. I was usually a little too "close" to the problem so when setting up a 24-hour plan, I'd put people where they made the most sense, and this particular guy was most impactful when he ran the above finicky press.

A VP once conveyed to me his own impression of skill sets: "Some people are mechanically inclined, and some just aren't. Take me, for instance. I don't "see" things in artworks the way other people do. Some people can look at a painting of flowers and see history, politics, and they can explain why and it makes perfect sense when you hear it. I just don't have /that/ knack, personally. There's a similar thing at work when interviewing press operators - some people can "see" the machine in full color, where others can't."

This was a bright guy, rather odd thing for him to have said (it makes him sound like an autist but he was anything but), but I remember it because I like art history and I can read images in the manner he described.

>>7648807
This is amplified where your primary area of training isn't electromechanical systems. I work in a medical pathology laboratory and most of the staff have an incredibly basic understanding of technology.

For example, one of the sample handling systems has a pneumatic element supplied by a compressor rated for intermittent operation. One of the regulators was damaged and started leaking, resulting in the compressor working continuously. I warned them that this was less than ideal but was told to shut up and focus on my area of work. An hour later, it failed exactly as I predicted (overheated, cooked off its oil and seized, the thermal fuse was held on by a thermal pad that had long since rotted, causing it to detach). It was astonishing that they continued to ignore the problem even as it secreted a thin white smoke from the cowl and made a sound "like a dryer with a brick in it".

>>7647918
two-stroke engines have amazing power to weight ratios.
>>7647978
I swear the US military made wankels werk for their drones?

I personally don't like helicopters. They are noisy, have dangerous giant flying blades, are hard to fly, are slow and expensive. I look forward to the day when the multirotor puts them to sleep once and for all.

>>7648938
>I swear the US military made wankels werk for their drones?
Mazda made wankels werk for some of their cars.

>>7648954
Exactly, the seal problem isn't insurmountable it's just the piston cartel keeping it suppressed as usual.

>>7647960
i doubt that kind of places are places where a human can go when its running

but look up industrial lathe accidents on google, that's some gore for you right there to get some sort of understanding what might happen with such machinery

>>7648962
idk I always thought the rx8 was kinda cool but you do have to swap the engine every 80k km or so I've heard.

Human body

File: newcdiag.jpg (28KB, 381x446px) Image search: [Google]

28KB, 381x446px

Newcomen engines were horrifically inefficient, getting down to less than a percent of useful work done for the energy input.

>>7647915
That's really a non-sensical question. For any given "worst" machine, there is some way to make it even worse.

You're basically asking what the largest integer is.

>>7649018
>Human body

There are lots of animals way worse than humans.

>>7649230

all boards are 18+

File: Xayqs3IS.jpg (321KB, 1971x1971px) Image search: [Google]

321KB, 1971x1971px

>>7649235
You're a CIA nigger

Go away

File: 1412350524084.webm (2MB, 640x360px) Image search: [Google]

2MB, 640x360px

>>7649239
o rly?

File: cianiggers.png (398KB, 1079x1497px) Image search: [Google]

398KB, 1079x1497px

>>7649357
God says
>surely die; his blood shall be upon him.

A Dell laptop

>>7647960
There are probably a million sensors that would either cut off power if they were tripped or would at least sound an alarm of some sort. Also, pressure sensors would detect a hand stuck in the production line, say if a robotic arm is applying over X amount of pressure, it knows something is wrong and stops

>>7647936
Dude the brain only uses like 100w, seems pretty efficient to me.
Also more often than not it can last 80 yrs without repair before giving out.

Perpetual motion machine
/Win

CNC Routers.
Anything below the super expensive tier of router is honestly just a gigantic paperweight.

>>7650028
I'm interested m9, what components are the most prone to shit?

>>7649534
human brain typically uses around 12w

>>7650246
That's because people typically only use 10% of their brain though.

>>7648081
>this can lead to a phenomena called "runaway diesels" where if a seal fails it will literally eat it's own oil and go nuts on you
not true. that gif clearly shows the piston pulling air

find the hose that goes to that, and suck on it

diesel engine stops

had a mercedes 240d that did this. worth it for 40+ mpg

>>7647915
I suppose it really comes down to how you define inefficiency. A Rube Goldberg machine in very inefficient from a mechanical standpoint, but when you consider that its actual purpose is to entertain, it is spending almost all of its energy doing exactly what it is supposed to.
I think the right answer was >>7649947
Absolutely none of a perpetual motion machine's energy is spent producing perpetual motion.

>>7650318
Bait?

File: image.png (46KB, 625x626px) Image search: [Google]

46KB, 625x626px

>>7650370
I kind of assumed so, but just in case it isn't, I'm going to reply.
>>7650318
The whole "you only use 10% of your brain" thing is a common misconception brought about by the fact that metabolic brain scans usually show around ten percent of the human brain lighting up at any given time. These scans are calibrated against baseline metabolic activity and do not indicate an efficiency of 10%. The human brain is quite efficient indeed, you can rest assured.

File: 1447147842506.png (26KB, 625x626px) Image search: [Google]

26KB, 625x626px

>>7650379
I could not help but notice your png was not optimized anon.
I have optimized your png.
Your png is now optimized.

>>7650439
how do you optimize pngs and why do you do it?

File: 1447154524117.png (5KB, 625x626px) Image search: [Google]

5KB, 625x626px

>>7650439

That's not optimizing.
This is optimizing

>>7650443
I use a script I made that ties together a number of programs. I like putting things into a more optimal state, and figure reducing file size once carries for both its future storage and every subsequent time it's posted, reposted, and saved.

It also provides test images with attributes I might not have thought of or tested before, which allows better evaluation of how the script presently works, and over time possibly is useful for improving heuristics (ie for better results or more efficient operation).

>>7650455
pngquant etc aren't lossless.

>>7650462
so you're a programmer testing your script basically?
what language is the program in?

I'm just really bored, sorry for hijacking the thread. I'm learning java.

>>7650463

>muh lossiness

They look exactly the same, even without the flowery subpixel stuff.
You didn't need more than 4 colors indexed.

>>7650470
It's just written in batch. Which minimizes portability, but it's only really intended for personal use, and a few of the core programs don't have (good) mac or linux binaries available anyway.

I'm slowly programming a few other programs to go along with it though, I'd like to stop using imagemagick to compare decoded outputs. Gives an opportunity to work with huffman / arithmetic coding for jpegs, and deflate / all of png's scanline filters, which is interesting and will be much faster when it's done. Writing that part in C, which I'm not very proficient in yet.

>>7650475
"Need" is relative. If I wanted to work with the goal of near transparent lossy compression, there would be much better ways to go about it.

>>7649032
But it was necessary for more advanced engines to appear, and it cleared out all those nasty forests England used to be plagued by.

>>7650475
>They look exactly the same,
You might need glasses.
The third one is clearly lower quality.

I need glasses and I can tell the difference, this guy is blind.

>>7650476
nice good for you, keep doing it :)

>>7647915
Is that a fucking laser guided pretzel twister?

What a world.

>>7649018
if you knew anything about how it works you wouldn't be spouting such excrement, silly goose

>>7650685
Sounds like someone has never had it suffer a cascade of failures.

It's riddled with "blind spots", inefficiencies, and poorly constructed aspects that just happen to work well enough. Well enough is not ideal. Sometimes it isn't even functional.

>>7650703
In the context of OP's question, the human body works much better than most designed machines.
And since you're stretching "machine" to include life, you'd have to compare us to all other life forms.
I really doubt we're the worst.
Most animals live far shorter lives than we do, for instance.

>>7650719
Agreed. Life in general tends to be more efficient than most designed machines. The only exception I can think of is life's chosen method of information storage. You would not believe how much of our DNA contains useless instructions on how to be a fish that has just been commented out.
>>7650439
Thank you for optimizing my png, anon.
I appreciate every one of those twenty KB that you saved me, and I really and truly mean that.

>>7647915
Pretty sure you mean my printer.

File: failmachine.jpg (6KB, 250x159px) Image search: [Google]

6KB, 250x159px

>>7647915
very few of you know the horrors and the agony of having to see one of these fail during the middle of a major project. these machines always jammed. always bunched. always overheated.

It's what you get when you work with paper in general.

>these are used not just for grading scantron tests
>muh respect for the digital age

>>7647942
this is a good answer

while designing that piece of shit the scientists never stopped to think if it was a good idea. a huge waste of money.

>>7647988
they got fired, right? RIGHT?

>>7650741
>implying there is no use for noncoding DNA
>implying random mutations wouldn't fuck your shit if literally every piece of DNA would code for proteins

File: 1423538051254.jpg (39KB, 283x269px) Image search: [Google]

39KB, 283x269px

>>7648069
>>7648069
I smell a troll and people are actually responding...

The turboencabulator.

The government

>>7652335
>scientist
>>7652335
It wasn't so much the engineers (look up the difference to a scientist) but the politicians who messed it up. The original design was a two-stage rocket, both stages were reusable. Then the budget was cut to the point where they had to strap solid fuel boosters to the tank to get anywhere. Look up which spending increased while NASA was hobbled.

I'm sure the engineers knew boosters weren't a good idea, but if your job depends on delivering, you often keep going.

>>7652543
A story about the boosters being designed after a horse ass. It's a very indirect link, but the boosters are about two Roman war horses large for a reason.
http://www.snopes.com/history/american/gauge.asp

>>7652741
>http://www.snopes.com/history/american/gauge.asp
Why would the track's gauge (separation of the two rails) have much to do with the loading gauge which is determined by the clearance offered by tunnels, bridges or other obstacles along the line? You can put a very wide load on a narrow carriage as long as the load will clear obstacles and if its centre of gravity stays between the wheels, e.g. in curves.

>>7647988
>friend of mine working robot maintenance in auto factory
>machine is locked out and safed
>dickering with fiddly robot arm bits
>has to put hand between the jaws of an arc welder in order to hold a tool at the right angle.
>stops everything to check for the 4th time that everything is locked and safed
>puts hand between jaws
>jaws clamp down
>has to calmly explain to panicking assistant exactly how to power on the machine and release the clamp without initiating the normal startup routine, which would instantly rip friend's arm out of socket/completely off
>this takes approximately 30 minutes.
>after months of investigation, it is never discovered how the machine managed to snap the welder shut at that exact moment, without power or any kind of control signal. It had been off for several days while they waited for parts.

>>7650342
>not true
Nigga you dumb.
Diesel engine runaway is real.
You did describe the best way to stop it.
If you can get to it before it grenades.
Risk you're willing to take?

>>7653474
http://youtube.com/watch?v=3NRaqgab0_w

>>7647915
If array.isnotsorted()
Fuel.burn()
Piece.break()
Bogobogosort()
Slap that on an arduino

>>7653483
fukken sick cunt

>>7647915
Probably yuor benis :D

>>7649357
Is that old man Scaruffi in the middle?

Italian cars.

File: M16-A-nothing.jpg (56KB, 750x167px) Image search: [Google]

56KB, 750x167px

>>7647915
>What's the worst machine? Worst as in, the most inefficient, unreliable, or needs to be repaired the most often.
/k/ here. This is required posting for me.