Hacker News new | past | comments | ask | show | jobs | submit login
Fully digital valves open up engine possibilities (newatlas.com)
68 points by rmason on Aug 12, 2018 | hide | past | favorite | 56 comments



I thought koenigsegg was already doing this with their freevalve thing

Also I saw an interesting point in an article about the koenigsegg thing that would pertain to this too; environmental laws would be much harder to practically enforce with this as you could run virtually any engine anemic enough to get though the emissions tests (and set it as the default mode to get around the Volkswagen nonsense) and then offer high/sport/cruise modes that runs the engine at full capacity otherwise


For those of you interested in seeing some more details on Koenigsegg's work with their freevalve engine:

https://www.youtube.com/watch?v=Bch5B23_pu0 (2013)

https://www.youtube.com/watch?v=S3cFfM3r510 (2016)


> environmental laws would be much harder to practically enforce ... and then offer high/sport/cruise modes that runs the engine at full capacity otherwise

Only if your company is ready to willfully break the law.

The vehicle ought to, at the time of testing, meet or exceed all relevant standards, and not have a user accessible option to circumvent the standards.


Well there’s a car hacking then for a future Defcon tuning panel then... because if I buy it, I want to be able to hack and modify it in any way I want. As long as I’m not violating emissions, then it should be within the law

Thankfully, automakers have quite and uphill battle to actually make something secure and inaccessible, so I feel confident a solution would be found :)


This does happen now in car culture where people will tune it “chip[1]” their car to increase performance. In practice, it’s difficult enough (you can’t just do it through the car’s infotainment system) that your average driver won’t do it and it’s probably not to big of an empact on the environment[2] and people’s health[3].

[1]”Chipping” used to enclose physically replacing a car’s ECU but now it usually just means the control software has been chnaged digitally. [2]Unfortunately, some awful people will and do “roll coal” to purposefully increase emissions and smoke. It’s illegal in many places but difficult to enforce.

[3]Your car still has to pass emissions and sometimes people do bring the ECU back to stock firmware before annual inspections just to pass. Difficult to enforce but tuning usually sacrifices fuel economy for performance and its not simple to do. Most people who just want a grocery-getter won’t be trying it.


Wouldnt the test just need to apply to all modes? If its a matter of automatically tuning through a series of conditions then its reasonable that those could be documented and tested by the facility / standards body.


Efficiency testing of appliances (eg. Dishwashers) only tests the default mode.

That's why all dishwashers have a mode which takes 4 hours and does a lousy job. It's the mode that only uses 3 drips of water and a sip of electricity.


Do you have a link to any resource which talks about what the best way to use it? Everything easily searched seems to be how to get your dishwasher to run at its most efficient.

I've never had a problem with our one with it's 2.5hr eco mode, when running overnight (which is 90% of our loads). If we need a quick load during the day, we'll use the 40min cycle. Either way the wash quality seems the same.


I'm sure it varies widely by machine. Ours when on eco mode doesn't really dry the dishes (drying uses heat which uses electricity!), and only does one rinse to save water so dishes still have a bit of grime on them, especially if lots of the plates have thick gravy or anything on.

I would put eco mode as 'passable', but you have to remember to scrape off the plates well beforehand, rince especially gungy stuff (like porridge), and to probably do a bit of hand drying of items afterwards.

One the other hand, the 'intensive' mode only uses 50% more electricity, takes a quarter of the time, leaves the plates clean and dry every time, and uses a hotter wash temperature which I suspect is better at killing pathogens, and doesn't need any pre-rinsing or plate scraping beforehand. It's a no-brainer in my mind.


Sure, but emissions testing after you've got the car would be utterly useless.


The solution to this is to mandate that the car have built in emissions modelling.

The car itself must keep track of how much Nox it has produced and how many miles it's driven. The ratio must stay below some legal limit (or you pay an extra pollution tax to go over it)


How so? Roadside emissions testing is very much a thing, as are annual emissions checks.


Not in all states. I haven't had to do it in 20 years.


> environmental laws would be much harder to practically enforce with this as you could run virtually any engine anemic enough to get though the emissions tests

lol

Changing a cam (or cams) in most engines isn't all that difficult. Even without a camshaft you'll still be heavily limited by the shape and size of the combustion chamber and the geometry of the valves. You simply won't be neutering the engine that much.


Changing a camshaft in a classic car is huge pain that requires specialist equipment for setting the timing etc. and getting it wrong can destroy the engine. It's more than possible in a home setting but most home mechanics I know wouldn't touch them. I'm sure it's a lot more complicated on a modern car.


> Changing a camshaft in a classic car is huge pain that requires specialist equipment for setting the timing etc. and getting it wrong can destroy the engine.

That's overstating the case quite a bit. There are plenty of engines where you don't need special tools.

> I'm sure it's a lot more complicated on a modern car.

Wait. You were talking about older engines? Then, no. You're absolutely wrong. Swapping out cams on most older engines is a piece of cake. Pretty much anything without variable valve timing will be super duper easy. Pushrod motors are especially easy. There are some exceptions, of course, like the DeLorean's V6 (the cams slide out of the head so you generally won't have room to pull them out without pulling the head).


After market engine modding would be interesting too.


The idea has been around for a long time. Most of the big auto companies have played around with it. There are at least four startups in this space. The big problem is a good actuator, not compute power. Here's an actuator demo from 2014.[1] It can be made to work, but so far, it's not cost-effective. Maybe this time.

[1] https://www.youtube.com/watch?v=ZYEjAv0hFug


When I did my mechanical engineering degree a long time ago this would have been the holy grail for engines. Let's hope reality doesn't offer drawbacks to this system like it happens to so many other engine designs.


It seems in tech just when the incumbents seem like they will lose to the upstart they achieve a breakthrough and the inevitable is delayed for a few more years.

Combine this breakthrough with what Michigan State is doing in gas powered engines and battery powered cars dominance is delayed for at least another decade.

http://www.wkar.org/post/reworking-michigan-msu-develops-lig...


The inherently self destructive nature of ICE's make it likely that ICE vehicles will always be more complex, less reliable/ higher maintenance than comparable all electric vehicles. The simplicity advantage of electric will soon take a quantum leap when driveshaft-free 4 motor all wheel drive vehicles hit the market [1].

On the other hand as long as batteries are heavy, ICE's will have a place in aircraft, and I would think the OP's technology could be huge for helicopters.

1 http://www.electric-vehiclenews.com/2017/06/hondas-all-elect...


> The simplicity advantage of electric will soon take a quantum leap when driveshaft-free 4 motor all wheel drive vehicles hit the market

You can't just eliminate the driveshaft because the wheel must be on a suspension. Remove the driveshaft and you have to put the motor on the wheel, increasing unsprung weight, exposing the motor and making high voltage parts subject to damage.

4 motors is more complex than 2 and doesn't buy you much except in niche packaging applications (like buses). As far as torque vectoring, you can achieve the same effect with differential braking (the duty cycle of torque vectoring is very short, so you don't lose any efficiency). A 2-motor Tesla is probably already the optimal passenger car configuration.

Agreed on everything else though.


4 electric motors on wheels is extremely simpler than 2 motors, mechanical differentials and couplings.

I actually helped a mechanical engineer to create a simulator for car dynamics in software.

Today's cars are over engineered mechanically in order to support and resist the forces and torques from the motor to the wheels.

When you put the motors in wheel you basically reduce the required rigidity and the weight of the car substantially.

You point out some disadvantage of the approach. The biggest disadvantage of all is that is not a proven technology, like 100 years old driveshaft. That means the company that commercializes it will eat all the risk, like what happens when people exposes the motor-wheels to wet surfaces and some part of it is damaged. The lawsuits could bankrupt even the biggest company.


I agree with you. I should have clarified - 4 motors in the arrangement envisioned by Honda (see the link above) is more complex than 2. It doesn't eliminate any of the driveshafts.

I don't think we'll see in-wheel motors any time soon for the reasons you and I mentioned. (Also, in high performance applications (anything but econoboxes) won't you need to strengthen the suspension/chassis in some places even as you get to lighten it in others, since the motor is now torquing the rest of your suspension?)


" 4 electric motors on wheels is extremely simpler than 2 motors, mechanical differentials and couplings."

Why not 2 on one axle? Do you really need all wheel drive?


Is anyone making hub motors for cars? The technology is probably workable now.


Also, while it's simpler, it's not really much cheaper (which is the big driver of a lot of automotive decisions). Motors and power electronics are far more expensive than the mechanical components which you are removing (and this is not easy to change, though there is more headroom for economies of scale on the EV motors ATM).


> On the other hand as long as batteries are heavy, ICE's will have a place in aircraft, and I would think the OP's technology could be huge for helicopters.

The future of ICEs for aircraft is turbines, not piston engines with digital valves. Most commercial aircraft are already turbines; the only thing holding them back for light aircraft is cost.


>when driveshaft-free 4 motor all wheel drive vehicles

Two words: unsprung weight.

CV shafts are gong nowhere for the foreseeable future.


Powerful electric motors are getting much much lighter.

Especially with liquid cooling, you could get a 100hp peak (20 hp continuous) motor inside 30 lbs if that was a design goal.

Motors for cars will get lighter as price competition pushes for designs involving less mass of raw materials.


Well this quote from the original article struck me as odd:

> Nobody's ever done variable valve timing on the exhausts, so the benefits are less well understood.

BMW's been doing this for twenty years. Honda's been at it for over thirty years. On. The. Exhaust. Valves.


toyota 1998 Dual VVT


I wouldn't put either free valves or the "wave disk" engine on the list of technologies likely to affect mass produced ICE efficiency in the next decade or two. HCCI/SCCI gas engines with VVT, variable displacement, and cylinder deactivation are very close to full scale production though, and promise very substantial improvements. Mild hybrid (electric regenerative braking and assist with a small battery/large capacitor) gives another boost, though at a substantial complexity cost (not nearly as bad as a full hybrid though).


I would think that could help EVs, as it would be cheaper (in weight and dollars) to add a gas powered range extender to EVs with smaller batteries. I'd rather have an 80 mile EV with a range extender than a very expensive heavy 400 mile EV.


That’s probably one of the least interesting applications, really. We’re already quite good at optimizing engines for a narrow RPM/power range.


I was thinking more of optimizing for weight. Also, if you optimize for narrow RPM, then you are only using the gas engine to power a generator, so you have some loss there. I'd rather have the gas engine power i.e., the front wheels, while the EV powers the rear wheels (or vice versa).


I wish cars had had this 50 years ago. Camshafts are a terrible way to control an IC engine. But it's too late now.

Ironically, the same advances in power electronics, cheap high-speed computers, and electromechanical actuation that made individual valve control possible also enabled efficient electric drive motors that made valves in cars a non-problem.


You don't necessarily need such valves to make engines easier to start. I remember my father-in-law's Lister diesel generator, from over forty years ago, that would start automatically when you turned on a light or appliance. It had a tiny starter motor and also a starting handle in case the battery was flat. There was a switch or valve that you would turn manually, or if the battery was not flat automatically, that made it possible to turn the motor. It was pretty much impossible to turn the engine over manually without doing that; with it, it was easy.

I'm not a mechanical engineer and I have no idea how it worked beyond knowing that it reduced the degree of compression.


Its called a decompression valve.


The temporary 2 stroke idea doesn't sound very well cooked. Fine control over the valves is great, but the exhaust and other aspects of the engine need to be tuned for that use case.


Also the "2 strokes are the future" remark seems a bit off too. The future of enduro bikes and chainsaws maybe.


Yeah, this idea is just in time for the all-electric revolution.


The "all-electric' revolution you've got in mind is likely much further away than you expect. There will be a market for internal combustion engines for many more decades before all-electric vehicles can replace the markets in areas that are more remote, or require long hauling vehicles etc.


Yes, but like the hard drive market, as soon as a replacement technology comes over the horizon, all investment in improvements to the existing tech pretty much stops.


agree. but you can barely get a 2 stroke passed the existing emission laws, and they're not generally in use in motor racing outside of enduro. ¯\(°_o)/¯


Nitpicky, but compression release for easy (easier..) starting is already a thing. It talks about completely opening the valves so the starter only overcomes friction. Presumably they're talking about having the cylinders that aren't yet fired spinning over freely? Otherwise it's a very efficient way to spin and engine that can't possibly start without "compressing gasses".


I guess it'd lower the starting torque which might let them use a smaller starter motor? You'd spin the engine for a couple of revolutions, then use the stored kinetic energy to kick start the engine.


Exhaust valves are open during the exhaust stroke anyway. Inlet valves are open during the intake stroke anyway.

The strongest plausible interpretation of what¹ they're saying is to open valves that are partway through the compression stroke, and open / close any other valves as necessary at the best possible time to promote rapid starting.

1. https://news.ycombinator.com/newsguidelines.html


that really makes the most sense. at that point i imagine it would be a more finely controlled compression release. i shouldn't read too much into the wording of what is essentially a marketing statement i suppose :)


They have replaced a pure mechanical linkage with a motor in each valve.

A typical 4 cylinder engine will now need sixteen extra motors to run it.

While I like the idea of controlling the timing of each valve interdependently for the other, I am not for the additional electronics in the engine.


Its even worse, they replaced 1 camshaft with 4 camshafts, because electric linear actuators are not there yet (and probably never will be).


This means you can actually have a push-button anti-lag system. Push once for normal, nice keep the home owners association happy mode. Push twice for Group B hell on Earth mode.


Can private citizens obtain these?


This article is pretty much wall-to-wall snake oil. The bit at the end is just completely false. Some people, who hadn't really bothered to do the math at the time, believed fifty years ago that spring-closed valves would float at high speeds and desmodromic valves were the solution. But that isn't true: metallurgy marches on, and nobody has been seriously impeded in making an engine turn at any desired speed with valve springs. Motorcycle engines turn at stupefying speeds without experiencing valve float. In practice it's actually the desmodromic designs that are troublesome, because of all that moving mass.

Anyway, there are lots of other dumb claims in here, like nobody has ever done variable timing of exhaust valves, which is obviously not true.


>Motorcycle engines turn at stupefying speeds without experiencing valve float. In practice it's actually the desmodromic designs that are troublesome, because of all that moving mass.

Sure, go tell Ducati.


Tell them what? That literally every other major manufacturer makes a motor with higher max engine speeds, higher reliability, and lower maintenance costs? Or that in the 60-year history of Grand Prix motorcycle racing they've only won once?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: