melting pistons and lean fuel mixture 3

[mustben]

i am designing a gasoline vaporizer that will run a stream of air thru liquid gasoline then inject the the mixture into the intake manifold of a 6 cylinder 98 nissan maxima. this will be in addition the gasoline injected via the EFIs. i hope to be able to run a lean mixture perhaps up to a air fuel ratio of 20:1. i have been told to be careful about melting the pistons because the lean operating mixture can reach very high temperatures. i am a retired chemical engineer without any automotive experience. all feedback will be gtreatly appreciated. mustben

 

[TDIMeister]

I suspect that designing the fuel delivery system will be the least of your worries. An AFR of 20:1 equates to a Lambda of about 1.4, and that's coming close to the lean ignitability limit of gasoline. No amount of vaporization or fancy pre-treatment will help your cause.

The part and cause of melting pistons is poorly understood by most. The adiabatic flame temperature is a property of the fuel. It peaks slightly rich of stoichiometric at around lambda 1.1 for most hydrocarbon fuels and falls off on either side of this peak. The laminar flame velocity also falls off on either side of a peak near this region of AFR. The bit that causes melted pistons is that lean mixtures extend the combustion duration due to the slow flame speed, and therefore the mean cycle temperature rises. Peak flame temperatures for most hydrocarbons could be in the region of 2500 K and doesn't cause any damage to pistons, but it takes far lower mean cycle temperatures to damage them, usually first from lubrication failure before outright melting.

The way in which Diesel- and lean-burn gasoline engines designed operation run lean without incident is by stratification (spatially heterogeneous distribution of AFR) of the charge. That is, the global AFR in the combustion chamber is appreciably greater than stoichiometric, but there are that has AFR near stoichiometric for robust combustion to initiate. Aside from designs like the Honda CVCC of the late 70s and 80s, the most common way to achieve charge stratification is by direct injection of fuel into the combustion chamber with late timing close to the spark plug firing. This may require special spray geometry or air motion to ensure an ignitable mixture is present near the spark plug at the time of ignition.

When you do get to running mixtures that lean in a homogeneous, port mixture preparation, ignitability will be a major problem, usually further requiring a higher-energy ignition system.

 

[TDIMeister]

"The adiabatic flame temperature is a property of the fuel. It peaks slightly rich of stoichiometric at around lambda 1.1 for most hydrocarbon fuels and falls off on either side of this peak."

Correction: a mixture rich of stoichiometric will have a lambda less than 1. This is the inverse of the relative fuel-air ratio phi used by some.

 

[dicer]

Simply put a super lean mixture, will actually be cool because of the excess of air. It was done in the 1950's by some airlines running reciprocating engines.

 

[rotorimps]

Instead of air have you thought about ultra sonic atomization?

 

[romke]

what exactly are you trying to achieve? more power, lower fuel consumption?

anyway, with a mixture as lean as you describe you may well be outside the range that would normally be the operating range of a three way catalyst (assuming there is one) and then problably the EFI system would try to compensate for the changed air/fuel ratio and try to get it again within the range the threeway catalyst is supposed to operate.

so, even if you succeed in injecting the extra air/fuel you would somehow seem to override the standard engine electronics.

 

[mustben]

by using veporized gasoline i am trying to operate as close to the otto cycle as possible. a combustion engine that runs on fuel in its vapor form uses much less fuel , produces less polluttants because more fuel is burned and pefroms better. when running lean i plan to use o2 sensor simulators instead of the o2 sensors before the cat converter, this will send a signal to the ecu not to increase the amount of fuel via the efis, my goal is to use as much vapor gasoline. I am looking into modifying the ecu to accomodate these modifications. My concern is that i have been told by a nissan technician that i run the risk of melting the pistons by running a lean mixture, at this stage i just want to know from an automotive engineer if this is true at all? when using hydrogen or natural gas, combustion engines do not seem to have this problem. Does anybody have any empirical input? thanks .mustben

 

[Lou Scannon]

"Lean" depends on your point of reference. Typically gasoline automobile engines run rich of stoichimetric at maximum power to promote survival of vulnerable components such as pistons and exhaust valves. If you run the same power with a leaner recipe these components may fail prematurely.
At part throttle it's a different ballgame. Automobile gasoline engines run approximately stoichiometric as that is what the 3-way catalyst requires as feedgas to maintain low emissions. If you run a leaner recipe here emissions will skyrocket.

That's it in a nutshell.

 

[romke]

another problem when running lean in an uncontrolled way is preignition and detonation. when in certain areas of the combustion chamber conditions are so unfavorable that not-wanted ignition occurs a shockwave of locally pre-ignited fuel may act as a kind of blowtorch on specific parts like the piston top.

For more in depht info on pre-ignition and detonation, see:

http://www.streetrodstuff.com/Articles/Engine/Detonation/index.php

"Burned pistons" are mentioned on page 6

 

[TDIMeister]

"by using veporized gasoline i am trying to operate as close to the otto cycle as possible. a combustion engine that runs on fuel in its vapor form uses much less fuel , produces less polluttants because more fuel is burned and pefroms better.

I'm sorry to say it mustben, but the reasoning is flawed from the outset, specifically the statement that, "a combustion engine that runs on fuel in its vapor form uses much less fuel." A combustion engine does indeed run best when the fuel is ultimately burned as completely as possible. The ideal Otto efficiency is approached (but never reached) when heat addition approaches a constant-volume process, that is, in a practically instantaneous amount of time at firing TDC.

However, in a port-fuel injected engine, barring extremely bad design, the fuel is already completely evaporated by the time of spark ignition. I can't think of ANY modern, emissions-controlled implementation that has the mixture preparation so bad as to still have liquid fuel present during the beginning of combustion. Having so would be manifested by large quantities of soot emissions, which we know is not the case for a normally-operating SI engine.

The flawed thesis is the same case that's made by purveyors of "vapour carburetors" and their likenesses. Somehow, the fact is lost that lean combustion cannot be extended without limit, but, these people argue, is dependent only on how good the degree of fuel vapourisation is.

Granted, some improvement in fuel efficiency could be achieved by a reduction of pumping losses by some degree of dethrottling in a fully-vapourised, lean fuel-air charge. Some improvement in indicated efficiency could be achieved through a higher cycle-averaged specific heat ratio.

However, there will be efficiency losses due to the longer combustion duration of homogeneous-lean combustion that would actually cause you to deviate AWAY from a constant-volume heat addition that is defined by the Otto process. As noted above, detonation is a very real threat made more likely by lean mixtures.

 

[mustben]

how do engines that use hydrogen or natural gas differ? are they built or operated differently? with hydrogen the AFR is 34. mustben

 

[Lou Scannon]

"how do engines that use hydrogen or natural gas differ? are they built or operated differently?"
Well, methane and hydrogen are pretty much the two endpoints on the spectrum when it comes to spark-ignited fuels, at least in terms of ignitabilty and flame speed. That drives vastly different combustion recipe designs.

 

[patprimmer]

TDIMeister is correct.

You are trying to solve a problem that does not exist.

The inefficiencies in modern engines comes mostly from heat losses, pumping and friction losses to a lesser degree and incomplete combustion to a very small degree. Most of the incomplete combustion is deliberate to keep the catalytic converter hot so as to minimise emmissions.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[TDIMeister]

Hydrogen: two things -- first is that the laminar flame speed is very much faster than all other hydrocarbon fuels; not just by some percent, by an order of magnitude. Second is that it has a much wider ignitability range. Gasoline will start to misfire on mixtures around lambda 1.3-1.4 (19-20:1). Hydrogen has extremely wide ignitability range, and with a stoichiometric AFR of 34.2:1 in air. Therefore the tiniest traces of hydrogen can ignite; part of what makes it so dangerous. Backfiring into the intake is a very real possibility; all it takes is for a hot spot in the presence of hydrogen for it to ignite.

Methane in natural gas also has a wider flammability range (lambda 2.1 lean limit) and higher stoichiometric AFR (17.2:1) than gasoline, and the laminar flame speed is slightly higher. It also enjoys a RON of 140.

 

[BrianPetersen]

Should also note that spark-ignition engines that are designed (or have been converted) for using gaseous fuels (propane, natural gas) don't have spectacularly high thermal efficiency compared to their liquid-fuelled brethren. For that reason alone, I'm skeptical that injecting some yet-to-be-determined portion of the gasoline as a vapor will accomplish much.

History is full of inventions of vapor carburetors and the like. Plenty of those so-called 100 mpg carburetors were some form of vapor carburetors. If an engine doesn't have a spectacular efficiency improvement by running on propane (gaseous), there is no reason to suspect it will have a spectacular improvement by running on vaporized gasoline as opposed to liquid gasoline.

Plenty of modern engines rely on the cooling effect of the fuel as it evaporates to keep temperature down and control detonation. If you remove that by adding heat to vaporize the fuel, you may find the engine more prone to detonation, and may have to do something (e.g. retard the ignition timing) that is not in the interest of higher efficiency, in order to save the engine.

There are technologies out there that claim to be able to fire leaner mixtures (plasma-jet plugs and similar systems that give much, much higher ignition energy than standard) but the demand has not been there, because modern emission systems rely on stoichiometric air/fuel ratio. No point finding a way to fire a lean mixture if you can't use said lean mixture anyway because it won't pass emission certification.

 

[dcasto]

go here and read about the big boys in engines running natural gas engines :

http://www.cat.com/cda/layout?m=38842&x=7,

toward the bottom is the G3616.

And the big big boys are here.

http://www.wartsila.com/,en,product...868B1BE-EC55-49CB-BC55-13D23DEC0E89,,8003.htm

Look at the video on their lean burn engine.

 

[RossABQ]

Just for fun, here's a piston removed from a turbo engine that had an injector that was dirty and the cylinder went lean under boost. The "pebbly" material coating the crown is melted aluminum that sprayed around... Lean miuxtures are rough on components!

 

[SonyAD]

I feel a far better pursuit for fuel economy would be materials engineering to be able to build a diesel or DI petrol engine that requires no cooling and the block, head, manifold, injectors, fuel lines, pump, everything is at, near or in the proximity of combustion temperatures.

This naturally precludes all indirect injection engines.

 

[TDIMeister]

Adiabatic engine are quite 70s and 80 technology. The driver for this was in the military where they ideally wanted engines that would not leave a significant heat signature that can be traced with infra-red detectors. Well, they found out quickly that Carnot's rule meant that a lot of heat will be rejected anyway even in a perfectly thermally-insulated engine.

Further investigations into adiabatic engines focused on potential efficiency improvements and obviating the need for complex cooling systems. However, recently research in to this area has all but gone by the wayside because the technology will not really hold up now with NOx regulations. But for non-automotive engines (e.g. industrial and marine applications), this might still be worthwhile, what with recent advancements in design methodologies, materials and coatings.

 

[patprimmer]

I would feel very nervous if the fuel line was near combustion temperature.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.