Slashdot Mirror
Camless Internal Combustion and the Digital Age (hackaday.com)
szczys writes: The internal combustion engine is amazing, and it continues to evolve. Carburetors gave way to fuel injection, and a computer now monitors all kinds of sensors to ensure these engines operate at peak efficiency. But there is one thing that has remained largely unchanged: the cam shaft. This is a device responsible for mechanically timing the operation of the cylinders. It's possible to build an engine that uses digitally controlled actuators instead of a camshaft to decide when each cylinder should fire. These exist as prototypes — we have the technology, so why aren't we building with it? The answer is that change is hard, and as with the carburetor it could take an outside force (in that case mandatory efficiency benchmarks) to get automobile manufacturers to wager a bet on new technology.
The fundamental parts of the engine are all mechanical. They work without a battery.
Resilience to electrical failure is important.
Camshaft hooked up to my Raspberry Pi. Ruby on Rails controlled Iot webserver platform with home automation built in. Insteon X10 platform protocols provide robust social media sharing.
rumor is the hydraulics used a ton of power. The thing was much less efficient than a traditional cam driven engine. Sure, the valve timing and lift was perfect, but it was otherwise a nightmare.
Ever break a timing belt on an interference engine? Very bad.
One of our competitors trademarked the term "hypothesis". From now on, we will call them "boneheaded ideas".
http://jalopnik.com/what-its-like-to-ride-in-a-car-with-the-camless-engine-1529865968
I gather the price is still prohibitive...
It's possible to build an engine that uses digitally controlled actuators instead of a camshaft to decide when each cylinder should fire.
Camshafts don't control when cylinders should fire, that's an already replaced component called the distributor. Camshafts control the timing of inlet and outlet valves, and there are already formula one and other engines using electronically actuated pneumatic valve lifters.
The problem is that cam shafts are very reliable, and a single fault in valve timing, in an interference engine especially, results in catastrophic engine damage, so the software and hardware has a very high bar to meet for it to replace mechanical cams.
Also firstpost.
-puddingpimp
I was under the impression that getting enough mechanical force to actuate the valves quickly by any other means is a hard problem.
Being able to vary the valve timing under computer control could have some advantages, but are they outweighed by the need to use high-power, sophisticated electrical actuators?
We've already got hybrids. Driving a car with electrically activated valves might be like owning a top-of-the line VCR just before DVRs.
I highly doubt it'll give any benefits - and it replaces a proven principle developed over 150 years. Through variable Camshaft Gears modern engines already control the in- and outlet timing for maximum efficiency today. And if you ever changed your own parts on the valves you'd know that it takes tremendous force to depress the valve into the "open" position on larger engines. It probably just isn't worth it to replace the (nearly free) mechanical force a camshaft uses to press them (each valve moving "up" helps depressing the others through its spring-force) with an electromagnet of the same force. It uses large amounts of energy, you can no longer "roll-start" a petrol engine with a drained (or damaged) battery, you have to include power electronics to control the magnets all for (probably) no measurable gain in efficiency whatsoever?
These exist as prototypes â" we have the technology, so why aren't we building with it?
Probably because a camshaft will last for 200K+ miles, and this new technology will not.
Inefficient, noisy, polluting, maintenance intensive, expansive, complex, why are we still trying? There's no alternative?
I'm far and away not an engine guy, but I always thought the reason was limited lifespan of solenoids.
A cam is just a spinning part. A solenoid would have to electrically activate perfectly every time, thousands of times a minutes, for 15-odd years of usage. To match that kind of usage, you're talking some serious solenoid. Probably they do exist but they're not exactly standard hardware, as far as I know.
And even the article suggests you tinker with models where the pistons can't crash into the valves. When you're tinkering, maybe that's okay. When you're designing engines it's not really okay.
Like all things, it's not that it can't be done. It's that the investment to make it work, work right, work first time every time, and prototype it to oblivion so you know that, probably far outweighs what you'll get back in any kind of efficiency saving on a non-trivial engine. Even rotary engines are comparatively rare compared to other types.
Things are most certainly heading all-electronic. But if you're going that way, almost certainly your investment is better of in electric drive, rather than huge investment into a critical piece of technology that - if it goes wrong - means a new car, for the sake of an slight efficiency increase.
TL;DR - Because its a stupid idea and you clearly haven't put more than a fuzzy seconds thought into it. Theres not really any point in doing so.
The first question is: WTF is a digital cam, other than some retarded idea you came up with because OMG DIGITAL DUDE!!@$#!@$. Don't use words you don't understand and don't make any sense at all.
The first statement is: The instant you said the cam shaft controls when the cylinder should fire ... you made it completely clear that you don't know shit about engines since you can't even get basic terminology right. You've lost all credibility already by anyone with a clue.
Now lets get to the meat:
Any changes to the cam shaft are done in order to change the power band of the engine, what RPM range where it produces the desired output power and efficiency. Any change to it just changes the ideal power band, so unless your running across a wide power range, really going all over the place all the time, then this is pointless.
There is ultimately little reason to come up with some electronic gizmo to do this because an engine can simply be built to fit the task at hand. For most engines, the RPM range is very small and constant. Even cars have a relatively small RPM range when in motion thats limited to less than 1000 RPMs difference across the entire power cycle when the transmission is taken into account.
So awesome, you can gain %2 efficiency at the end of the power band in automobiles ... and it'll break 1,000,000 times more often because the existing design is a single solid chunk of metal that sits in the second hottest most stressful part of the engine ... and you want to replace with a bunch of moving bits and magnets.
You'll get more efficiency out of your engine by just using the proper oil than you're going to get out of a highly dynamic cam shaft.
When you exclude cars ... then almost every engine remaining runs at a single given RPM ALL the time, meaning there is absolutely 0 value to a dynamic cam configuration and the net is a negative value due to increased complexity and decreased reliability.
Now go back to making a digital hammer and a digital screwdriver ...
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
if today's engines ran so efficiently, why the fuck is my 3 year old subcompact getting worse mileage (30 mpg ave commuting if lucky) than a 10+ year old escort in the 90s (40+ mpg similar driving and weather, and much more neglected in terms of maintenance)
One of the big limiting issues in this field (BTDT) is energy consumption by the actuators and associated circuit components. Valve are heavy relative to the accelerations needed by the motion profiles. This results in ferocious energy use and dissipation.
If this power consumption is more than the engine power/efficiency gains from tinkering with profiles, the answer is an easy No.
My only relevant direct experience was for an R&D engine to test different cam profiles without having to grind sets of camshafts. It used plant electrical power, can't remember exactly how much but the equivalent horsepower was in the teens.
Bent, folded, spindled, and mutilated.
Modern app appers drive apps that use APP engines, not LUDDITE camshafts!
Apps!
Seriously. If the technology is mature (regularly survives a 500 mile race) while providing tangible benefits (more horsepower meaning a faster car with better fuel economy which means fewer pit stops) customers will demand it.
Otherwise, it's of no perceived value to customers and might be seen as just another piece of electronic junk that is being foisted upon them (like anti-lock brakes for those of us who remember people who couldn't see their value).
Mimetics Inc. Twitter
Automakers embrace whatever gets people to chose their car over the competitors car, is cost effective to manufacture, and is reliable enough to get past the warranty period but not so reliable the car last too long past the warranty.
There were fuel injected cars long before it was practically mandatory. I believe there were Corvette's from the 1950's that had early fuel injection. I test drove a 1982 Fiat that had it. Fuel injection was slow to be adopted due to cost benefit ratios along with performance. Early fuel injection wasn't always better than a well tuned carb, in fact it was very popular among enthusiast to remove fuel injection systems from 1990's Mustangs and put carburetors back on.
Cars still use camshafts because camless systems haven't been developed to a point that mass production and implementation makes sense to a manufacturer. Not just a custom built lab system, not just a small run on rinky-dink 3D printers, but something that can be churned out by the 1,000's.
I was hoping new Slashdot ownership would steer us away from these pro-regulation propaganda pieces.
The preceding post was not a Slashvertisement.
What it really means is that the efficiency improvements are so poor it is not worth the additional cost. Unless some government agency, for good reasons or for bad, forces the manufacturers to change it is not worth it.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Here is an example of a cam-less engine that has been in development for over 10 years.
http://www.dukeengines.com/
46137
Just a reminder, when your fuel injector goes tits up, you just replace the fuel injector. When your timing belt/chain breaks ("digital" or mechanical), you replace your engine.
Why bother?
More parts to break.
"If any question why we died, Tell them because our fathers lied."
I thought electronically controlled pneumatic valves were old tech (as in almost 20 years old)? Im surprised they haven't evolved into modern cars yet.
Konigsegg has one about to be incorporated into their newest car.
Ethanol in the fuel is a major culprit; it's killed the volumetric efficiency of engines (MPG). The emissions controls also have a non-trivial cost, as back pressure through a catalytic converter and resonator is much larger than just a muffler, and compression has to be controlled to limit NOx generation. Additionally, the mixture is not running for ideal power or maximum energy efficiency, but is now leaned out enough to reduce CHx emissions. Not that these are a bad thing (other then ethanol, if you were dumb enough to miss that it was a payout to ADM and Monsanto; their stock prices are beautiful) but they have a very real cost. Yes, the VW fixes will most certainly increase CO2 emissions by reducing the efficiency of the engine in order to meet emissions criteria. It's a nasty tradeoff.
Will sooner than you think look just like a horse and wagon.
http://www.anl.gov/articles/argonne-achates-power-and-delphi-automotive-investigate-new-approach-engines
From the TFA:
As the pistons reverse course and slide to opposite ends of the cylinder, ports machined into the cylinder allow exhaust gases to escape while fresh air is taken in, then the pistons move together again to compress and ignite in a two-stroke cycle. The design eliminates cylinder heads — which are a major cause of heat loss and inefficiency in conventional engines — and allows the engine to run with diesel-like efficiency and power, while maintaining gasoline's emissions benefits.
An analysis by Achates Power indicates the new engine will yield fuel efficiency gains of more than 50 percent compared with a downsized, turbo-charged, direct-injection gasoline engine, while reducing the overall cost of the powertrain system.
Electronic valves have been discussed in many car forums for a while. The main issue is always been reliability. I believe some racecars or prototype engines have used electronic valves in the past. Thing is, actuators fail.. Often.. Especially at the speed and precision these have to operate at. In a modern V8 with 4 valves per cylinder, you' will have 24 independant actuators! ONE WILL FAIL before 20,000 miles. And popping the valve cover off an engine tucked into a tight compartment is not going to make changing these easy..
So as to the OP's question "Do we have the technology to do this?" Yes.. And we have for a while.. Its far harder technically to make a variable CAM. BUT.. A CAM with a strong timing chain is FAAAR more reliable than 24 independent actuators.
They have a retrofitted mini-van prototype they've been using to stress test it.
My Ducati has cams for opening and closing the valves.
Will I need a larger battery when they go all electric valves?
A bullet may have your name on it, but artillery is addressed to " Whom It May concern"
I have a 1991 Mazda RX-7 with a rotary engine in my garage, no cams there!
Just because it is possible doesn't make it a good idea.
Many engines are interference engines, where the valves sweep through the same space of the pistons during different parts of the cycle. This is largely due to the need to have high compression ratios (IC engine efficiency is strongly driven by compression ratio). Timing is crucial - if the timing is off, valves crash into the pistons and your engine tears itself to pieces. Timing belts, chains, or even gears are used to prevent this. It is particularly important to get timing belts serviced regularly, because they are a cheap part that can cause destruction of the engine if they fail. However, it is only a single point of failure.
The digital solution introduces at least two and possibly more points of failure per piston Loose or frayed wire? *BOOM* - destroyed engine. Mechanical failure of a single lifter? *BOOM* - destroyed engine.
Left MS Windows for Linux Mint and never looked back!
Vote for Bernie in 2016!
The forces involved in a valvetrain, coupled with the speed of operation needed are going to call for some chunky power electronics driving the valve actuator solenoids. Given that the typical failure mode for power MOSFETs is to go short circuit, this would drive the affected valve wide open and hold it there. The open valve then gets bent over and possibly driven through the top of the piston.
I can't ever see this kind of high power (multiple kW), high speed electronic switching system being as reliable as a timing belt that typically runs 100k miles before recommended replacement.
Remember "News for Nerds, Stuff that Matters"? Help make it a reality again! http://soylentnews.org
No, wait... We're already talking about cars... Hmm, where does one go to for an analogy when the subject is already cars?
Lucas was working on this 36 years ago.
The real "Libtards" are the Libertarians!
Of course, if the cylinder firing is regulated digitally, then a software bug (or remote hack) has the potential to *destroy* your engine. Handy!
http://www.motorauthority.com/news/1101737_video-shows-inner-workings-of-koenigseggs-camless
I think this is a bad idea. People have already shown on TV how easy it is to take control of an auto's computer network, a mechanically actuated system is much more secure and reliable.
Given that the typical failure mode for power MOSFETs is to go short circuit, this would drive the affected valve wide open and hold it there.
There are ways to design around that. Safety critical power electronics has been done for decades.
It has one camshaft for operating the intake valves and one for the exhaust valves. What it does have is separate cam LOBES for the opening and closing followers. A single solenoid for each phase could open both intake/exhaust valves, and another pair could handle the closing. With the twin, though, that's four solenoids/head, which would need to be cooled, and the vertical cylinder is already a packaging hassle.
I have both an air cooled 2-valve and a liquid cooled 4-valve, and, while it might be nice to get away from the periodic checking/adjusting, I wouldn't trade the simplicity or reliability for more electronics.
I understand the use-case of replacing the carburetor with fuel-injection, and computer-controlled fuel injection allows you to optimize the performance of the engine for a variety of cases that are only measurable with electronic (i.e. computer) sensors. But what's the use-case for replacing the camshaft. It's got a monkey-stupid job, lifting and dropping poppet valves. It performs that job perfectly because it's dead-simple. It's directly connected to the crankshaft, so it operates for as long as the engine is turning. It's got a dead-easy task, of lifting and dropping lightweight valves, so it contributes virtually no parasitic drag to the engine, and because of the simplicity of the design, (it's a rod with bumps on it) it lasts FOREVER. Whoever heard of replacing a camshaft? The reason why nobody's moved to electronic valve control is that the camshaft is in the Dieter Rams design hall of fame. It's great design: It's not part of the problem, so don't change it.
The engine is just a constant-load, constant temperature (once it warms up) generator to top up the battery. Any adaptation to power demand should be handled by the electric drivetrain.
Nullius in verba
Can you put this into simple terms that I can understand? Perhaps maybe use cars as a simple analogy.
Lots of discussion about what happens with a mistimed valve in an interference engine. It's a valid concern, but as has already been pointed out, happens a lot in regular camshaft engines also. The most trivial example is a broken timing belt.
Someone mentioned that electric valve actuators could be less forceful than a mechanical valve train, so that the piston slamming against an open valve would be bad but not catastrophic. But valve springs tend to be pretty strong, so it'd take lots of force to open a valve, so maybe the impact would be catastrophic after all.
To me the advantage of digitally actuated valves is the elimination of not only the camshaft but also the springs. As Ducati discovered awhile back, if your valve train pushes the valve open and then pushes the valve closed without springs, you can maintain much higher RPMs without danger of valve "float", where the spring isn't able to push the valve closed in time. This same type of arrangement should be possible in digital systems also, where the valve is electronically opened and also electronically closed. Then if the system goes dead (electoronic version of a timing belt snap) the valves are under no pressure and the impact force with the piston is much lessened.
A mistiming might still be difficult, but there should be a way to design the system to survive it.
I don't see the primary benefit to be one of efficiency. I see this as a way to get wickedly high RPMs. Think of a motorcycle engine that redlines at 3000 RPM or more.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
Go buy the solenoids with the required lifetime specs. What do you figure, 10,000 operating hours per year, 100000 rotations per hour, a valve opening every other rotation?
That's 10M actuations per year. An engine is good for about 10 years, 100M actuations. 16V per engine (or so), so you need on average 1 failure or less per 16V per 100M actuations.
Now go price that out and I think you'll find the reason why we don't do it this way to be obvious.
Or just ask anyone who has had a fuel injector replaced and remember that an engine usually as 4x as many valves as fuel injectors.
http://lkml.org/lkml/2005/8/20/95
why are we still burning fossil fuels for transportation anyway?
I hope your link gets the attention of the moderators because that was really cool.
Opposed Piston Opposed Cylinder engines have two pairs of pistons facing one another, each in a cylinder on opposite sides of the crankshaft. There is no cylinder head, just a ring of ports for intake and exhaust in the cylinder walls near where the pistons bottom out. With a slight timing offset, the exhaust ports will open before the intake ports. It is a fascinating design, simple and elegant, with very few moving parts and a high power density. The engine is completely balanced, and all of the linear forces cancel, leaving little load on the bearings, just torque. There are other interesting concepts out there, but this one is actually being mass produced today.
Electric cars are certainly attractive, but the reality is that hydrocarbon fuels are going nowhere. The energy density and flexibility are simply too great, and we have an immense amount of infrastructure and equipment that make use of them. The fastest way to a greener world isn't through electric cars, but rather synthetic carbon-neutral fuels, which can be efficiently produced using heat from nuclear reactors. Nuclear Ammonia is particularly interesting, because the feedstocks are readily available from air and water. Other replacement fuels can also be synthesized, but extracting carbon from air or water will add to the cost.
a really cool TRIANGULAR opposing cylinder layout engine.
Read that as "camelless". Oh good we can have cars without camels?
I'm a good cook. I'm a fantastic eater. - Steven Brust
just what i want, another piece of plastic with a wire trace-mask governing a critical part, which some disgruntled programming skittler dictating the screw up. as of now i go Throttle body injection, that shit was the poison to all if corroborated is the cure.
It's not about being more reliable.
It's about being more flexible. Maybe faster.
Bound to be a way to make the circuit fail closed.
deleting the extra space after periods so i can stay relevant, yeah.
https://www.youtube.com/watch?v=f4p-55a3WV8
Koenigsegg has had at least one car driving around with air/hydraulicly operated valves for a few years now. It's got tens of thousands of miles on the setup.
Now.. someone just needs to get the system to market.
You would have to be crazy to be sane in this world. -Nero
I designed some spacecraft. We were limited on what we could use for a microprocessor because we needed something with history. We almost used an 8051 variant. Same goes for engines, lets say you invent an awesome engine, it even makes you toast in the morning amongst other great things like saving you fuel. You take it to one of the manufacturers, and they love it, because it saves their customers fuel and it give them more incentive to buy their products. Then they tell you, we have these things called warranty's that we offer on all of our vehicles can you tell us the MBTF? Most of the parts are new so you go back to your lab and run it for 5-10 years.
I try to read slash-dot every day... Never posted before. :)
If I understand your post? I would say electromagnets coupled to the valves would be the answer.
It's possible to digitally controlled electromagnets instead of a camshaft.
P.S. let me know what you think?
Koenigsegg have a Saab car running with camless technology as a study model for their own hyper cars. There's a few videos on the t00b.
When you come home, you open your mechanical door by pushing. Why not have sensors and motors? Will it improve the experience significantly. I think that is the crux. Unless the technology significantly improves something, like electronic fuel injection, electronic fuel pumps etc., did, there is no point.
My Aurora : http://www.youtube.com/watch?v=o91ZsGwJYyg
FB : https://www.facebook.com/TanveersPhotography
Notwithstanding the OP observation concerning the rate of development of the internal combustion engine (and at the risk of starting a flame war, which is not the intent) why not move away from *all* the sins of reciprocating engines and go with something like a Wankel rotary? Yes, I appreciate that the biggest single failing of the rotary is often rotor wear, but the rate of development of materials science has been so incredible over the last 10-15 years that it is pretty much certain that a clean-sheet design started today would be a quantum leap forward from anything we have seen thus far. I've never personally driven, let alone owned, a rotary engine, but maybe starting with a problem statement of "How do we get rid of mechanically operated valves?" is setting our sights a bit low?
While the pros for a technology change like this would be interesting especially for motor tuning and efficiency, I assume the cons are pretty big ... for one, the rise and lowering of the valves isn't a binary on/off setting, but requires specific speeds (otherwise you may mess up the valve springs etc.). Also, with the current cam shaft technology, destroying a motor is more or less dependent (at least as valves and cam shaft go) on failure of the timing belt or chain (which is why there are specific recommendations in which intervals they should be replaced). With more sophisticated actuators, any failure in controlling them correctly could lead to bent valves or worse ...
Stop using it. It makes no sense. Use electrical motors that are cleaner, leaner, easier to repair. Optimizing it is like climbing a garbage dump, when you're at the top things still stink.
> I can't ever see this kind of high power (multiple kW), high speed electronic switching system being as reliable
In modern electric locomotives, IGBT converts up to 6000kW of power, from 25kV, 50/60Hz AC to ~1500V DC to 3-phase AC, actively regulating or even interrupting 1000+ Amperes of current flowing to the traction motors, up to 2000 times per seconds! Reliability is very important for the railways, so as no to block the tracks with a fallen engine.
The IGBT tech is also found in the Tesla. In the recent decade or so it has become a kind of sorcerers' stone for the electrical engineers. Cure the ill, turn lead into gold or AC into DC or vica-versa, the answer is invariably to use IGBT magic, except for some insanely large kVA projects where GTO / Thyristor still rules.
The answer is that there is no benefit in the majority of cases.
Just because you can digitize something doesn't you must.
The cam shaft is extremely simple and extremely unlikely to fail. It just works. It only has two failure modes, one internal one external.
Internal: structural failure, which for those up to speed on their material properties, is extremely unlikely. Any structural failure root cause is more likely to be the result of a bad batch of material than anything else.
External: rotation failure. This isn't a failure of the shaft, but a failure of another component of the engine that is failing to drive it's rotation.
It is because of this simple reliability that there has been little reason to reengineer or replace the camshaft.
By contrast digital timers have much more complexity and points of failure. Now I do believe that reliability is ultimately an engineering problem, and one that will be solved in time. And there may come a point when the pros of digital cam replacement carry a net benefit. But that time is not now. The primary scenarios I can envision (currently) where the digital replacement would be preferable would be things like extremely vibration sensitivity, or where physical space requirements are too tight to allow a cam shaft, or possibly where the reduced need for lubrication (though if you have a need to avoid lubrication you probably want to avoid internal combustion and its myriad moving parts entirely).
The guy who said the election was rigged won the presidency with the second-most votes.
Here's a rotary engine with no valves: Liquid Piston
The video from Koenigseeg has been on youtube since July of 2013:
https://www.youtube.com/watch?v=f4p-55a3WV8
They claim that they can support engines that operate at up to 20,000 RPM.
Take a look at the FIAT MultAir engines - https://en.wikipedia.org/wiki/...
Changes in momentum are what cost energy. Having a bunch of pieces of metal constantly reversing direction inside the cylinders seems pretty 1920s.
--- wad
Suppose the autos still equipped with 5 speed manuals or automatics got finer control over valve lift and dwell at all RPMs and conditions? There'd be significant improvements, no doubt about it. BUT suppose an engine and CVT were carefully designed to work together for maximum efficiency. I seriously doubt that there's that much to be gained over computer-controlled mechanical/hydraulic variable valve timing and lift. Notice that most Subaru models are now available with CVTs and many manufacturers are touting 8 and 9 speed transmissions.
I can imagine having both a camshaft and individual actuators where the cam opens the valve the minimum amount of time and duration and the electric or hydraulic actuator supercedes it much of the time. Then if the actuators failed there'd still me a limp-home mode. What about interference? Don't do it. There are plenty of efficient engine designs that don't have it. It's already bad news when mixed with timing belts.
I think I'll wait for the digital crankshaft.
Camless engines (electroniccontrled injection) are aready used since ~ 10 years in ships. It is true the engines are huge, diesel and work at low rpm.
Or just use a relatively small air control valve and solenoid. Pneumatics then open the valve. And a hefty spring to close it. If you're being clever, I bet the vented bypass air from the pneumatics could be injected into the cylinder during intake along with the fuel.