Worth noting the design of the internal combustion engine hasn't changed much in 50 years.
The thing that has changed is the control systems.
What used to be a primitive mechanical way of mixing fuel and air (the carburettor), is now an electronic fuel injection system, with the fuel air ratio very carefully matched to reduce pollution (fun fact: modern cars release so little carbon monoxide, you won't kill yourself by starting one in a garage (but don't try it just incase your car is faulty)). Catalytic converters use any tiny fuel air imbalance to reduce carbon monoxide and soot, and on the other side nitrous oxides, by slightly increasing and decreasing fuel air ratios.
There's also been advancements in cylinder head technology (i.e., VTEC, VVT, etc), which I guess also falls under control systems, but worth mentioning as these technologies are very cool. Honda's iVTEC has it down to a damn science with how to optimize valve lift & duration across the entire RPM spectrum.
The orbital family of ICE engine of 1972 has seen a lot of ongoing innovation and crossover variations with other rotary engines (like the Wankel) in past decades - such engine variants have had a major place in UAV (and other) drones.
Another fun fact: the body's suffocation response isn't caused by lack of oxygen. It's caused by high CO2 levels in the blood [0]. If your car floods the air with CO2, you will respond like someone is holding your head under water.
Carbon monoxide kills you by binding more strongly to the hemoglobin than the O2. You don't notice, because the CO2 removal mechanism continues to work fine. You'll usually die before realising anything is wrong.
It’s because of the good enough precision at low cost that we use cams, it’s extremely difficult to get precise movement and the forces and speeds required when an engine is operating at >5000 rpm. It’s not impossible but the trade offs are rarely worth.
The exact valve timing isn’t really a big factor in emissions as much as temperature control and exact AFR control. I mean valves need precise timing to avoid coming into contact with pistons but if you’re already at that level of precision then you gaining more precision won’t really reduce your emissions.
Of course being able to change the timing of your valves helps with both efficiency and emissions, but VVT does that pretty well.
> Presence of oil is critical here as it creates conditions for hydrodynamic lubrication.
You can hear this effect in some vehicles at initial startup time for a few seconds. I know of certain Ford engines where it actually causes issues over time. The model years with auto start/stop have the worst of the cam rattle disease.
I thought it pretty well established that auto on/off is bad for the engine, as is intermittently turning off some cylinders as some do. Is that wrong?
Note that that sentence is talking about the crankshaft bearings and their hydrodynamic lubrication, which is, well, elsewhere and separate from any cam rattle issues (including the cam phaser oil starvation that you might be referring to).
It's being pushed up by the crankshaft which keeps moving because of the inertia of the moving engine parts, especially the flywheel.
On multi-cylinder engines, other cylinders add torque at different phases of the rotation.
I still remember the first time I tore down a pushrod V8. I decided it had to be close to the pinnacle of elegant design. Nothing wasted, everything had a purpose, and it all came together in a perfect mechanical symphony. We have since made engines which are significantly more efficient and powerful, but all of that at the cost of elegance, slapping on overhead cams and machinery just to adjust the valve timing, etc. Fantastic technology in it's own right for certain, but feels tacked on, like an expensive optimization.
Reminds me that I want to get something for my kids to work on which will maybe show them some of that same elegance. I don't currently have any V8s in the garage to go tear down :)
The thing that's missing here that really drastically changes the story is all the emissions control hardware that would exist on such an engine.
This is a circa 1990s engine in the US market i think? Dual Overhead Cam didn't really become popular in the US market until then i think. 70s-80s for single overhead cam to become established.
The diagrams are beautiful and informative as always from this author.
The bearing surfaces in an engine (ex: crankshaft main bearings) have very tight tolerances, usually in the 15-25 thousandths of an inch. The engines oil pump fills those tiny gaps with pressurized oil which allow the metal surfaces to spin thousands of times per minute without damage.
This is also why if you have any issue with oil pressure (ex: oil pump failure, cracked oil line) or oil starvation (ex: driving a regular car on a race track, cornering forces slosh oil away from the oil pickup in the sump) issues, you'll damage your engine nearly immediately.
It's 0.0015, that's 1.5 to 2.5 thousandths, or 15-25 "tenths" as they're called.
That's not a particularly tiny gap in the machinist world, it's large so that you can pump viscous oil in it and deal with a wide variety of temperature changes.
25 thousandths would be sloppy, a nominal clearance hole for a 1/4x20 bolt is about that much.
Who is this person. A beautifully written and illustrated explanation of a fascinating machine. A website filled with other explanations of other things, also wonderfully written and clearly explained.
An instagram filled with beautiful landscape photographs, an "X" page consisting only of links back to this blog, and a Patreon with hardly any more information that that.
I love this. Fantastic content. Zero ego. And if there was any AI use, it's invisible. Certainly there is none in the writing.
There is no AI use here. I'm somewhat offended by the suggestion even though I have no relation to the author. Bartosz has been shredding for a long time now. His javascript is also unminified and perfectly readable, making it a great reference for study despite not being advertised on his pages. The animations are all hand-made. The guy is cooking. To suggest that any of it is AI is an utter blasphemy. AI is for noobs, grifters, and low-effort content farmers and Bartosz isn't any of those things.
I agree completely, and if you re-read what I wrote I didn't exactly suggest that it was used. But if he's reading this I'd hope he'd nevertheless accept my apology for even the hint of a suggestion.
Happy to find someone else who hates that infernal technology as much as I :)
Why does this show a 4 cylinder engine firing as two pairs? My ignorance is on display, surely it's more subtle and there are 4 distinct phased firings not two?
There are four distinct firings. Look for the little yellow poofs. The pistons move in pairs for balance (vibrations). Each piston is doing one of the four strokes (suck, squeeze, bang, blow is the crude way to remember) at a given time. None of them are doing the same thing at a given time.
Wonderful but it irritates me that so many descriptions of internal combustion engines refer to "explosions" of the fuel. You don't want that. It causes knocking and pinging and engine damage. You want a controlled burn that generates heat smoothly.
Even more confusing to anyone who doesn't know the lingo, detonation in the context of an internal combustion engine means something specific. It is a synonym for pinging and knocking, and happens when unburnt fuel/air mixture explodes after the spark plug has already fired. It can damage the engine, but typically takes some time. Preignition is when the mixture ignites before the spark plug fires, and is typically much more damaging, often destroying the engine in just a few revolutions. It can pound through the boundary layer between the mixture and the face of the aluminum piston and melt it, or break something else in the engine like a rod.
It's been a minute, but at one point GM had some pretty interesting videos up on YT where they talked about preignition testing on Cadillac Northstar V8s and how quickly it would grenade the engine. Fascinating stuff.
Not exactly. You do want a deflagration and not a detonation, but "explosion" is more loosely defined and, depending on who you're talking to, a self-sustaining subsonic flame front and a sharp pressure spike are a perfectly valid explosion.
explosion/detonation causes engine knocking or pre-ignition which is both very bad. a properly working combustion engine is driven by controlled burning.
I don't think it's any slower than the dust explosions that occasionally happen in factories or grain silos. Although those do appear more violent due to the scale & damage.
And why not call it "Four cylinder four stroke ICE" or "Inline four cylinder four stroke ICE" or "Inline four cylinder dual overhead cam four stroke ICE"? Of course there are other forms of ICE, the description does not need to detail exactly which subtype is described.
If you like this kind of stuff go and look up videos on the Rolls Royce Crecy engine from WWII. Absolutely insane engineering that died due the dawn of jet propulsion.
I'm showing this page to my team and investors every couple of weeks. Visual, animated explanations are MUCH better than textual content for deeply grokking something. This is what we're trying to build for large software systems. I love the animations on this site so much, thank you for building them.
This is a stupid question but I'm a stupid EE/SWE who knows very little about physical objects.
In the all these animations of the pistons I see linear motion translated into rotary motion using the crank shaft - but how do you design the pison/crank to always turn clockwise or counter clockwise (based on how you view it, obviously)? Is it possible for the crank shaft to lock up if it's perfectly oriented at 0 degrees?
The starter motor turns the engine in a defined direction, this acts as a turning force directly on the crankshaft so it doesn't matter where the crankshaft is. The pistons only start firing after the crankshaft is already moving.
It is actually possible for an engine to turn the wrong way, this occurs on motorbikes with kick starts. When you don't kick start it correctly (or if the ignition timing is way out), a piston can fire prematurely before top dead centre and force the crankshaft against the direction that the kick lever turns it, this is known as kick back and is about as fun as it sounds when the engine's force goes through the kick lever.
> but how do you design the pison/crank to always turn clockwise or counter clockwise (based on how you view it, obviously)?
You can design the starter motor to ensure the engine always starts up moving in the right direction, and after that it's "just" a matter of timing (e.g., spark plugs controllled electronically in more modern cars, mechanically in older ones).
> Is it possible for the crank shaft to lock up if it's perfectly oriented at 0 degrees?
Sibling explanations are correct re: direction of rotation. As to your second question, the flywheel keeps things spinning so it doesn't lock up and has the momentum to compress the air/fuel mixture. BTW 0 degrees is called top dead center or TDC and is a useful point for calibrating timing and ignition.
You meant - awful knocking combustion in the first, main animation?
I never catches any real bug is those great posts, but this one, especially as first animation on the page - weird.
You might be misreading the animation. It's a direct injection engine, the thing that happens during the compression stroke is fuel injection. Ignition happens a few degrees before TDC, which is realistic.
The carbon footprint of producing an EV is higher than an ICE. Nobody disputes that.
But the lower carbon footprint of operating the an EV makes the EV have a lower footprint over the life of the vehicle. This is true even if your electricity comes purely from coal, as a coal power plant gets an economy of scale that an ICE doesn't achieve. If your power comes from a renewable source, then the trade-off happens even sooner.
Worth noting the design of the internal combustion engine hasn't changed much in 50 years.
The thing that has changed is the control systems.
What used to be a primitive mechanical way of mixing fuel and air (the carburettor), is now an electronic fuel injection system, with the fuel air ratio very carefully matched to reduce pollution (fun fact: modern cars release so little carbon monoxide, you won't kill yourself by starting one in a garage (but don't try it just incase your car is faulty)). Catalytic converters use any tiny fuel air imbalance to reduce carbon monoxide and soot, and on the other side nitrous oxides, by slightly increasing and decreasing fuel air ratios.
There's also been advancements in cylinder head technology (i.e., VTEC, VVT, etc), which I guess also falls under control systems, but worth mentioning as these technologies are very cool. Honda's iVTEC has it down to a damn science with how to optimize valve lift & duration across the entire RPM spectrum.
The orbital family of ICE engine of 1972 has seen a lot of ongoing innovation and crossover variations with other rotary engines (like the Wankel) in past decades - such engine variants have had a major place in UAV (and other) drones.
* https://en.wikipedia.org/wiki/Sarich_orbital_engine
* https://www.orbitaluav.com/
* https://www.aieuk.com/225acs-wankel-rotary-engine/
> fun fact: modern cars release so little carbon monoxide, you won't kill yourself by starting one in a garage
Modern cars still release as much CO2 as older cars… which is still incompatible with human respiration.
Another fun fact: the body's suffocation response isn't caused by lack of oxygen. It's caused by high CO2 levels in the blood [0]. If your car floods the air with CO2, you will respond like someone is holding your head under water.
Carbon monoxide kills you by binding more strongly to the hemoglobin than the O2. You don't notice, because the CO2 removal mechanism continues to work fine. You'll usually die before realising anything is wrong.
What I am surprised by is that cams are still used to operate the valves.
Given to how precise these need to be these days, I would've guessed they would've switched to electronically actuated valves.
It’s because of the good enough precision at low cost that we use cams, it’s extremely difficult to get precise movement and the forces and speeds required when an engine is operating at >5000 rpm. It’s not impossible but the trade offs are rarely worth.
The exact valve timing isn’t really a big factor in emissions as much as temperature control and exact AFR control. I mean valves need precise timing to avoid coming into contact with pistons but if you’re already at that level of precision then you gaining more precision won’t really reduce your emissions.
Of course being able to change the timing of your valves helps with both efficiency and emissions, but VVT does that pretty well.
Fiat (and other's I'm sure) have the [MultiAir system](https://www.caranddriver.com/features/a16580674/fiats-multia...) which doesn't use cams to actuate valves
And here's an example of someone that retrofitted a Miata to use a similar [air-actuated valve system](https://youtu.be/E9KJ_f7REGw)
1 reply →
I was going to point to Koenigsegg’s Freevalve system and, in the process of looking for a link, learned that it was cancelled.
> What I am surprised by is that cams are still used to operate the valves.
Kinda, it's all continuous variable valve lift now:
https://en.wikipedia.org/wiki/Variable_valve_lift
And there are also other parts for reducing the pollution like DPF, EGR and Adblue systems.
Unfortunately they also reduce the reliability.
> Presence of oil is critical here as it creates conditions for hydrodynamic lubrication.
You can hear this effect in some vehicles at initial startup time for a few seconds. I know of certain Ford engines where it actually causes issues over time. The model years with auto start/stop have the worst of the cam rattle disease.
Auto start/stop isn't off for enough time for the oil to drain from the galleries and especially not out of the bearing journals.
It's the first few seconds after an engine has been off for hours (or worse, for potentially years) that are the problem.
I thought it pretty well established that auto on/off is bad for the engine, as is intermittently turning off some cylinders as some do. Is that wrong?
7 replies →
Note that that sentence is talking about the crankshaft bearings and their hydrodynamic lubrication, which is, well, elsewhere and separate from any cam rattle issues (including the cam phaser oil starvation that you might be referring to).
Lifters also often drain while sitting and valve lash is greater at start until they get slapped a few times
That's the timing chain tensioners losing oil pressure.
It's unclear to me, why would the compression stroke (the 2nd out of 4) happens? There's nothing pushing the piston upwards.
It's being pushed up by the crankshaft which keeps moving because of the inertia of the moving engine parts, especially the flywheel. On multi-cylinder engines, other cylinders add torque at different phases of the rotation.
I still remember the first time I tore down a pushrod V8. I decided it had to be close to the pinnacle of elegant design. Nothing wasted, everything had a purpose, and it all came together in a perfect mechanical symphony. We have since made engines which are significantly more efficient and powerful, but all of that at the cost of elegance, slapping on overhead cams and machinery just to adjust the valve timing, etc. Fantastic technology in it's own right for certain, but feels tacked on, like an expensive optimization.
Reminds me that I want to get something for my kids to work on which will maybe show them some of that same elegance. I don't currently have any V8s in the garage to go tear down :)
The thing that's missing here that really drastically changes the story is all the emissions control hardware that would exist on such an engine.
This is a circa 1990s engine in the US market i think? Dual Overhead Cam didn't really become popular in the US market until then i think. 70s-80s for single overhead cam to become established.
The diagrams are beautiful and informative as always from this author.
"in real running engines the rotating crankshaft should float completely on a very thin surface of oil" - I found this to be a great insight.
The bearing surfaces in an engine (ex: crankshaft main bearings) have very tight tolerances, usually in the 15-25 thousandths of an inch. The engines oil pump fills those tiny gaps with pressurized oil which allow the metal surfaces to spin thousands of times per minute without damage.
This is also why if you have any issue with oil pressure (ex: oil pump failure, cracked oil line) or oil starvation (ex: driving a regular car on a race track, cornering forces slosh oil away from the oil pickup in the sump) issues, you'll damage your engine nearly immediately.
It's 0.0015, that's 1.5 to 2.5 thousandths, or 15-25 "tenths" as they're called.
That's not a particularly tiny gap in the machinist world, it's large so that you can pump viscous oil in it and deal with a wide variety of temperature changes.
25 thousandths would be sloppy, a nominal clearance hole for a 1/4x20 bolt is about that much.
3 replies →
That's the point of all uses of oil, other than rust prevention.
Some engines are also substantially cooled by oil, but those are either older designs (think “air-cooled” Porche) or industrial prime-movers.
9 replies →
I hope Ciechanowski posts some more in the near future. He has some of the best "how does this work" articles I've come across.
Pro tip: Show a message if WebGL is disabled instead of a blank space.
There is a very good engine simulation software:
https://github.com/Engine-Simulator/engine-sim-community-edi...
Who is this person. A beautifully written and illustrated explanation of a fascinating machine. A website filled with other explanations of other things, also wonderfully written and clearly explained.
An instagram filled with beautiful landscape photographs, an "X" page consisting only of links back to this blog, and a Patreon with hardly any more information that that.
I love this. Fantastic content. Zero ego. And if there was any AI use, it's invisible. Certainly there is none in the writing.
There is no AI use here. I'm somewhat offended by the suggestion even though I have no relation to the author. Bartosz has been shredding for a long time now. His javascript is also unminified and perfectly readable, making it a great reference for study despite not being advertised on his pages. The animations are all hand-made. The guy is cooking. To suggest that any of it is AI is an utter blasphemy. AI is for noobs, grifters, and low-effort content farmers and Bartosz isn't any of those things.
I agree completely, and if you re-read what I wrote I didn't exactly suggest that it was used. But if he's reading this I'd hope he'd nevertheless accept my apology for even the hint of a suggestion.
Happy to find someone else who hates that infernal technology as much as I :)
Why does this show a 4 cylinder engine firing as two pairs? My ignorance is on display, surely it's more subtle and there are 4 distinct phased firings not two?
There are four distinct firings. Look for the little yellow poofs. The pistons move in pairs for balance (vibrations). Each piston is doing one of the four strokes (suck, squeeze, bang, blow is the crude way to remember) at a given time. None of them are doing the same thing at a given time.
Thank you. TIL!
When I was a kid, pre-internet, my dad took me to the local library and checked out some books that explained how an engine works.
These animations are so much better than what I had!
When I was a kid we had a visible V8 model that let you see quite a bit!
[2021] Originally 2333 points and 392 comments:
https://news.ycombinator.com/item?id=26991300
Always good to revisit his older work, though I admit I did get excited that it was a new post!
Very interesting technology. Would be exciting to see a hardware startup build a product around this.
Hmm not sure about that. It seems like a lot of complicated effort to produce locomotion. I’ll stick with my reliable horse, thanks.
Wonderful but it irritates me that so many descriptions of internal combustion engines refer to "explosions" of the fuel. You don't want that. It causes knocking and pinging and engine damage. You want a controlled burn that generates heat smoothly.
Even more confusing to anyone who doesn't know the lingo, detonation in the context of an internal combustion engine means something specific. It is a synonym for pinging and knocking, and happens when unburnt fuel/air mixture explodes after the spark plug has already fired. It can damage the engine, but typically takes some time. Preignition is when the mixture ignites before the spark plug fires, and is typically much more damaging, often destroying the engine in just a few revolutions. It can pound through the boundary layer between the mixture and the face of the aluminum piston and melt it, or break something else in the engine like a rod.
It's been a minute, but at one point GM had some pretty interesting videos up on YT where they talked about preignition testing on Cadillac Northstar V8s and how quickly it would grenade the engine. Fascinating stuff.
Not exactly. You do want a deflagration and not a detonation, but "explosion" is more loosely defined and, depending on who you're talking to, a self-sustaining subsonic flame front and a sharp pressure spike are a perfectly valid explosion.
explosion/detonation causes engine knocking or pre-ignition which is both very bad. a properly working combustion engine is driven by controlled burning.
3 replies →
There are videos on YouTube of what the combustion actually looks like in slow motion. It's fast, but far less violent than an "explosion".
I don't think it's any slower than the dust explosions that occasionally happen in factories or grain silos. Although those do appear more violent due to the scale & damage.
You don’t want detonation, but you do want deflagration.
This should probably be called “Four Strike Engine”. There are other types of ICE that work differently.
Four stroke.
And why not call it "Four cylinder four stroke ICE" or "Inline four cylinder four stroke ICE" or "Inline four cylinder dual overhead cam four stroke ICE"? Of course there are other forms of ICE, the description does not need to detail exactly which subtype is described.
You mean stroke?
If you like this kind of stuff go and look up videos on the Rolls Royce Crecy engine from WWII. Absolutely insane engineering that died due the dawn of jet propulsion.
I'm showing this page to my team and investors every couple of weeks. Visual, animated explanations are MUCH better than textual content for deeply grokking something. This is what we're trying to build for large software systems. I love the animations on this site so much, thank you for building them.
This is a stupid question but I'm a stupid EE/SWE who knows very little about physical objects.
In the all these animations of the pistons I see linear motion translated into rotary motion using the crank shaft - but how do you design the pison/crank to always turn clockwise or counter clockwise (based on how you view it, obviously)? Is it possible for the crank shaft to lock up if it's perfectly oriented at 0 degrees?
The starter motor turns the engine in a defined direction, this acts as a turning force directly on the crankshaft so it doesn't matter where the crankshaft is. The pistons only start firing after the crankshaft is already moving.
It is actually possible for an engine to turn the wrong way, this occurs on motorbikes with kick starts. When you don't kick start it correctly (or if the ignition timing is way out), a piston can fire prematurely before top dead centre and force the crankshaft against the direction that the kick lever turns it, this is known as kick back and is about as fun as it sounds when the engine's force goes through the kick lever.
> but how do you design the pison/crank to always turn clockwise or counter clockwise (based on how you view it, obviously)?
You can design the starter motor to ensure the engine always starts up moving in the right direction, and after that it's "just" a matter of timing (e.g., spark plugs controllled electronically in more modern cars, mechanically in older ones).
> Is it possible for the crank shaft to lock up if it's perfectly oriented at 0 degrees?
That's what the starter motor is for!
Sibling explanations are correct re: direction of rotation. As to your second question, the flywheel keeps things spinning so it doesn't lock up and has the momentum to compress the air/fuel mixture. BTW 0 degrees is called top dead center or TDC and is a useful point for calibrating timing and ignition.
Some engines, nowadays mostly large marine diesels, are reversible. Change the valve timing, and start it in the opposite direction, and off it goes.
This is why the starter motor spins the shaft in a known direction.
While they look perfectly round, piston skirts are actually slightly oval.
...something which has been the case for at least 80 years:
https://www.web.imperialclub.info/Repair/Lit/Master/003/inde... )
This is an excellent explanation!
Thank you!
Excellent animations.
You meant - awful knocking combustion in the first, main animation? I never catches any real bug is those great posts, but this one, especially as first animation on the page - weird.
You might be misreading the animation. It's a direct injection engine, the thing that happens during the compression stroke is fuel injection. Ignition happens a few degrees before TDC, which is realistic.
1 reply →
One of the rare situations where someone wants a bit of retard?
1 reply →
[flagged]
Probably better for the environment too.
The carbon footprint of producing an EV is higher than an ICE. Nobody disputes that.
But the lower carbon footprint of operating the an EV makes the EV have a lower footprint over the life of the vehicle. This is true even if your electricity comes purely from coal, as a coal power plant gets an economy of scale that an ICE doesn't achieve. If your power comes from a renewable source, then the trade-off happens even sooner.
4 replies →
No