Like, totally not grassroots...

http://m.wimp.com/combustionengine/

Very cool. I'd love for them to convert my engine for experimental purposes.

I was going to post this a while back. I love this idea, just think if if it was coupled with some direct injection fuel system. They have another video of how you could use it in the multi-displacement mode and shut off the gas and open the valves so there's no compression resistance and run different cylinder modes to cut fuel at low speeds and idle. Pretty niffty stuff. http://www.youtube.com/watch?v=N3fSfBQSK0w

Electronic control over valve timing was my one great idea in high school. Too bad I didn't have the knowledge, skill, or means to pursue it in 1988...

At least I have one example of a time my brain had an idea that turned out to be good

ransom wrote: Electronic control over valve timing was my one great idea in high school. Too bad I didn't have the knowledge, skill, or means to pursue it in 1988... At least I have one example of a time my brain had an idea that turned out to be good

Hey I had that idea in high school too, although it was in 1997. Solenoids get too hot and are not consistent enough, or at least that was the problem then.

I cannot wait to see this setup in one of his supercars. I seriously cannot.

ransom wrote: Electronic control over valve timing was my one great idea in high school. Too bad I didn't have the knowledge, skill, or means to pursue it in 1988... At least I have one example of a time my brain had an idea that turned out to be good

Everyone tried it. Mostly starting near when Tuna "first thought of it" as an idea. We had some cars running around with it.

note the past tense for now.

This is very similar- just using compressed gas instead of electricity. But I think the scale of benefits are pretty speculative. Maybe over a crappy engine. There are a quite a few questions that need to be worked out, but if it works- it would be pretty interesting. (and it would make his company lots and lots and lots of money)

If you want to drive yourself crazy- how would you run them? Having that kind of degree of freedom is not exactly easy. Oh, and if you are not a fan of electronic throttle, this takes that to a totally different level.

In reply to alfadriver:

Last question first; I figure the starting point is actuation based on a good, basic cam profile first. Then simulating a good low-end torque cam at low RPMs transitioning lift, duration, and timing to a good mid and then high-RPM profile. Then there are probably other economy, cruise, emissions considerations which could be added in... While optimizing from scratch for all these things is daunting, coming up with maps from known cams for a few conditions and being able to switch or transition smoothly between them seems like it would provide a huge advantage over being stuck with one profile.

Not that I'm suggesting that any of that is trivial, mind you.

IIRC I'd imagined hydraulic actuators, but I was at least as clueless then as I am now about how dumb that may be. It certainly adds the mass of the hydraulic fluid to the list of things being accelerated and decelerated...

I guess that's an implicit note that I do want to drive myself crazy (and have trouble not attempting to answer rhetorical questions). And at this point it's no longer a new idea, and there's no point outside the fun of thinking about it.

I am a fan of electronic throttle, I think. All the issues I may have now are implementation choices, not issues with the technology (and honestly, the issues I have with my WRX I have no way of knowing whether it's throttle mapping, tuning, or the nature of turbos; But I've spent enough time driving electronically controlled R/C cars to know the response isn't a fundamental issue with the technology itself).

I thought that BMW already had an engine with (electronic) solenoid controlled valves in production?

Also, can any of the proposed solenoid controlled valvetrains alter lift as well as duration? Would that even matter if you had precise enough control of timing and duration?

yes.. I think some of the smaller (europe only) engines were not only running solinoid controlled valves.. but did away with the throttlebody by varying the intake valves

Not electronically controlled valves. but electronically controlled constantly variable lift with a traditional cam, making a throttle unnecessary. There are a few engines like that in production:

http://www.autozine.org/technical_school/engine/vvt_5.html

Quite interesting and amazing. That probably really is the wave of the future. Don't F1 engines run something similar?

I too had thought of some sort of solenoid drive for valves. This idea looks pretty cool, I'm normally a hater of new tech...The compressed air idea for a boost of power is really something!

A buddy and I wanted to run a bunch of starter solenoids to drive our valves. We drank countless beers in the garage talking about the benefits

(insert typical beatnik voice here) "Dude, imagine.... you could run one grind for slow gas mileage driving, then stand on the skinny pedal and BAM! Your cam timing changes. It'd be limitless man"

My buddy also came up with the "switcharoonie" where the car has two cams, but only runs on one. You pull a lever in the cockpit to pull the chain to the more aggressive cam when you want more power. (actually the idea that led to the solenoid idea).

Damn, those were good times.

ransom wrote: In reply to alfadriver: Last question first; I figure the starting point is actuation based on a good, basic cam profile first. Then simulating a good low-end torque cam at low RPMs transitioning lift, duration, and timing to a good mid and then high-RPM profile. Then there are probably other economy, cruise, emissions considerations which could be added in... While optimizing from scratch for all these things is daunting, coming up with maps from known cams for a few conditions and being able to switch or transition smoothly between them seems like it would provide a huge advantage over being stuck with one profile. Not that I'm suggesting that any of that is trivial, mind you.

Well, to poke a hole right away- don't think of a cam profile. That's not needed anymore. We are now talking valve events, so you can mess with it from one event to the next.

Then you have to program a bunch of other actuators- spark being a fun one- since valve timing has a big influence on MBT and borderline spark.

Knowing the complications of just dual independant VCT, when the cam lobes don't change, the control system for so many degrees of freedom is rather complex.

That, and one still needs some kind of vacuum, so there will still be a throttle somewhere.

So- throttle (mostly open), valve timing, spark, fuel (air charge calculation), EGR (which is new tech version, to make combustion more effiecent), still add vapor recovery, turbo (wastegate, bypass valve, and scroll control valves), forgot direct injection timing and pressure, all that put together with some kind of demand system (torque, power, whatever).

It was easy when you had just fuel, air, and spark- with limited or no movement of all that.

Not that it can't be done- it's just really complex and interesting.

jstein77 wrote: Quite interesting and amazing. That probably really is the wave of the future. Don't F1 engines run something similar?

Sort of- they still use camshafts, but the springs are not metal, but compressed air. That's one of the key reasons they can spin to 20k rpm.