Living Throttleless

[NiHaoMike]

What about use piezoelectrics for the water injection? (The saying for the "Amy Aldrich" piezoelectric water injection idea is "Hard water or not, Amy Aldrich just keeps on squirtin'.")

[ZeroFuel]

Originally Posted by jesse.rizzo

Just the other day I came up with an idea for a throttleless gas engine. It would have to have electronically controlled valves. The ecu would automatically cut fuel and hold open the valves for a split second really fast. So for example it would provide fuel for one revolution, and then cut it for 3 revolutions, and the engine would be producing 25% power. But it would be fast enough that the driver would never notice. The only question is if the slight amount of engine braking would be too much.

Anyway, congrats on the fairly out there mod. Interested to see how it works out.

Check out new Fiat valve train

Fiat Multiair Technology Increases Power, Lowers Emissions – Automotive News & Car Rumors at Automobile Magazine

[stevey_frac]

Kudos for playing with this.

An idea for when you get her back together, setup a water injection pulse width controller to keep the exhaust temps at a certain temp? Like, it could vary how much water to inject to keep the exhaust at say, 700 degrees or lower. This would conserve water, and help prevent thermal runaway.

It wouldn't prevent the failure you observed, but it would be an improvement i think.

Regarding the NOx production. I would suspect that it's not as bad as you think. The water your injecting should keep the peak cylinder temps down.

Really cool project.

Subscribe'd

[Shawn D.]

Originally Posted by Frank Lee

Incidentally, years ago I read about BMW's experiments with no conventional throttle. They manipulated the intake valves to provide the throttling.

It's not an experiment -- it's been in production for years: Valvetronic - Wikipedia, the free encyclopedia

Originally Posted by theunchosen

... heat syncs...

Heat sinks. Even if that works (I doubt it will because it's not just temperature but detonation that melted that hole), pulling more heat out is reducing your thermal efficiency, opposite of your intent.

[theunchosen]

Alright I'm supposed to be mowing some hay. . .but its raining so I'm taking a break and slacking off.

Mike, I have no experience with piezoelectrics. The wiki informed me its something akin to a memory metal with electricity rather than heat. So I guess it replaces the pump? Or am I missing something on piezoelectric injectors? What failed on the unit was not the pump it was the actual nozzle. I am uninformed about them so its hard for me to speculate whether or not they would solve my problem.

Zero I checked the images. I haven't the foggiest on how to actually make that work for me. I'm sure it does, and I don't doubt the claims but I'm definitely not talented enough by myself to undertake such a large change that requires incredibly small tolerances.

Stevey, Yeah I had thought of piggying it onto an EGT sensor, but after sunday. . .saturday? I think I need the injectors to be AHEAD of the engine. I know it sounds stupid but I think I need to tie the injectors to the MAP sensor. If i tie in at the EGT its going to be probably 3-4 revolutions behind the air flow. Maybe more(probably more as the pump can't possibly change rate that fast) and after my very fortunate one piston catastrophe I'm tempted to run the the pump as close to locking up the engine as I can get(with excess water). The MAP is still completely intact and feeds to a DMM in the cabin so I could reroute it to the pump and that would keep it electronically about 4 inches of intake ahead of the actual engine. Probably not enough time for the pump to actually stay in step with the engine but it won't be semi-static like it is now and it would respond more quickly to runaways. Either way would definitely be an improvement on the high end of the RPM band where I'm currently not able to get to(probably not quick enough to adjust the flow rate and it gets ahead of me). Thanks for the input, Hadn't considered it just yet ^_^.

Shawn, Since its brought up again I'll go ahead and say I'm not a fan of extremely complicated engineering. I'm not a fan of the Nissan GTR, at all. Its complicated. I'm well aware and I'm tweaking something that was pretty simple and just got very much more ridiculous and is less reliable(obviously ^_^). On the other note the piston mess sitting out in my driveway was caused exclusively by heat. Det occurs most of the time around stoich and rich cylinders. Mine weren't even remotely close to that. Det occurs when the mixture in the cylinder has a very rich spot and it inundates the O2 and keeps it from igniting(gas is less flammable than O2 and has a higher octane rating). Eventually the pressure from the ignition event creates enormous temperature and the pocket of air ignites late(as far as we are concerned simultaneously but in absolute terms its picoseconds or whatever later than it should be causing the fuel to actually detonate(hence detonation) rather than burn). If the piston had cracked or splintered or anything or if I had been running it for a while and detonation had caused something in the engine to get too hot then it could have caused the pre-ignition. But thats definitely not the case. Det usually only kills high output engines and its a stretch to call my D series high output. Its nowhere near aggressive enough in any dimension to cause serious det problems and with far less fuel it puts it further and further from possibility.

You get two paragraphs ^_^. I do want to draw heat out of the motor and yes it is counter productive. My radiator does it and I'm trying to make it do it faster. I know carnot would call me an idiot, but carnot's engine doesn't melt and mine does. Surprisingly the crank case is a good bit cooler than the rest of the block, because Drag teams were worried about pulling near vacuum on the back of their pistons because it would overheat very quickly because there wasn't enough mass to ride out the temperature surges. Turned out the crank case stayed cool enough that it didn't matter. Even if its only 15 degrees cooler and I can cool 7 degrees its going to be worth it. Its also not all about my TE, its about my fuel efficiency(fuel strictly as gasoline). My engine is now about 1/4 as thermally efficient as it was, but it uses about 1/4 as much gasoline and still produces just and i mean just enough power to do what I need it to. I'll still have a rubbish Cd so I'm not looking at 100 mpg but maybe in the neighborhood of 50-60 up from 35.

[NiHaoMike]

Originally Posted by theunchosen

Mike, I have no experience with piezoelectrics. The wiki informed me its something akin to a memory metal with electricity rather than heat. So I guess it replaces the pump? Or am I missing something on piezoelectric injectors? What failed on the unit was not the pump it was the actual nozzle. I am uninformed about them so its hard for me to speculate whether or not they would solve my problem.

It would operate similarly to ultrasonic humidifiers.

[turbodan]

Detonation occurs any time you have enough heat and pressure to cause a pocket of fuel and air to spontaneously "detonate". It occurs after the spark has ignited the mixture but before the flame front reaches and ignites the whole combustion chamber. Excessively lean AFR's or too much ignition advance cause detonation. Either one generates excessive heat, some of which is blown out through the exhaust and some of which is retained in the combustion chamber.

Theres that and theres preignition. Preignition occurs when you have a hot spot that ignites the fuel and air before the spark is fired. This generates extreme and unsustainable amounts of heat.

The problem with this is that theres nothing regulating cylinder filling other than the volumetric efficiency of your engine at any given RPM. The realistic limit for lean running under any kind of load is about 16:1. It wont run much at all leaner than 20:1. Its going to run wide open all the time, and for it to run efficiently its going to require reasonable AFR's. You know what happens when it leans out under load. Heat sinks and oil spray nozzles wont keep that from happening. All you can do is give it enough fuel to keep from melting itself down.

[theunchosen]

Perfect Nihaomike, I'll look into it. I'll need some that have a reasonably high output.

dan, You're right det is caused by a pocket that fails to ignite with the rest of the cylinder but in my experience most of the time when det is really hammering it out is when the cylinder is too rich. It simply just doesn't have enough air in the pocket to overcome the octane rating and the pressure. It can happen under any circumstances, if the pressure gets great enough fast enough it can trap the fuel in a high pressure state until it detonates when the temperature continues to increase.

Pre-ignition is what toasted me. It toasted me from the amount of heat being output during lean burn without being regulated.

I disagree on the AFR. Honda, Toyota, Ford, GM and others have all built mule engines that can do better than 35:1. Not only that they can do it without suffering diminishing returns. The problem with typical engines that go lean. . .is they go lean with very very little fuel to start with. I'm trying to create the amount of fuel you get at say 2000 TP 24% with the amount of air you'd get at 2000 TP WO. The problem and the reason you can't traditionally go much higher than 20-25 is. . .they kept the same amount of air and dropped it to a marginal amount of fuel. The small engines use less than a drop of fuel per injection. If you lean it out at 2000 rpm. . .your banking on pure dumb chance that the fuel is going to be near the spark plug. The fuel ignites just fine in lean conditions if the probability is high enough that there is fuel near the spark plug. With a normal cylinder full I haven't really had problems, not to mention the fact that the O2 air mixture has a lower octane rating than gasoline in those volumes and with the additional heat stored from each combustion event.

[Shawn D.]

Originally Posted by theunchosen

I disagree on the AFR. Honda, Toyota, Ford, GM and others have all built mule engines that can do better than 35:1. Not only that they can do it without suffering diminishing returns.

There's a tremendous difference between building unique/one-off/prototype/mule engines for research and attempting to replicate the results on an engine with stock architecture. Unless you are completely redesigning and fabricating new pistons, chambers, valve location/angles, injector location, spark plug location, etc., you cannot make such a comparison. Adding heat sinks to pistons doesn't count as an architecture change.

BTW, why is this in "Success Stories" when the main result so far is an engine meltdown?

[turbodan]

At some point overly rich AFR's can make an engine more susceptible to detonation but thats only because an extremely rich mixture burns slowly enough that it allows more time for detonation to take place. The cooling effect of relatively rich AFR's during cylinder filling and the cooler burn during combustion makes motors much less susceptible to detonation than they are running leaner AFR's. Its not like you have saturated rich pockets in the combuston chamber. Generally the amount of turbulence that occurs during the cylinder filling process makes for a pretty well mixed fuel and air charge.

I can say for sure that if I were to lean out my turbo car past 14:1 the results would be quite expensive.

If you want to utilize direct injection and very fancy combustion chamber designs, you can get away with running some traditionally ridiculous AFR's. Even then, you're still going to need a throttle to make the engine suitable for use in an automotive application.