Running Lean and "Melting Pistons" 2

[dangnm]

There is an old motorhead rule that goes something like this

"Never run lean or yer gunna burn a hole in yer pistons"

To avoid confusion, when referrig to this post, the definition of LEAN will be considered A/F greater than stoich and RICH will be considered less than stoich

In the application I'm working on a lean burn with max lambda of 1.1 would be very advantageous however before breaking this rule I want to ensure that I understand the problem entirely.

Emissions aside, With instrumentation of a N/A engine I have found the following.

While running rich:

Power increases as A/F in decreased until roughly 13.5:1 (.9 Lambda)
Exhaust temperatures decrease
From A/F of 13.5 to 14.7 the amount of ignition timing that could be added without detonation did not change.
Increasing intake air temperature requires retardation to prevent detonation

While running lean:
Power decreased
Exhaust temperatures ???? <-- (no instrumentation at time)
Ignition timing required retardation to prevent detonation.
Increasing intake air temperature requires retardation to prevent detonation

From this testing I believed that the motorhead rule evolved due to the following two reasons.

1) Decreased resistance to knock may result in detonation causing engine damage.

2) Motorhead carburated systems typically run rich .8-.9 lambda to take advantage of the power gains running leaner causes EGT's to increase which may lead some to believe that if you go too lean you will "melt the pistons"

I have read the following which seems to be a reasonable explanation of change in knock resistance.

A lean air/fuel mixture burns with most efficiency, so much that the insulating boundary layer also gets consumed and the flame front touches the metal walls. At those locations, there is a dramatic rise in temperature, high enough to cause subsequent charges of air and fuel to spontaneously ignite.

Question 1: is this explanation of the decreased knock resitance correct

Question 2: If detonation and EGT's are controlled will damage still occur.

Question 3: What is the chemical/thermodynamic reason for the increase of power when running with enrichment?
On the surface one would expect that the complete combustion of a stoichiometric fuel/air ratio would produce the maximum power.

 

[ivymike]

Question 3
You're limited by how much air you can get into the cylinder, so you want to burn as much of it as possible. Mixing (+etc) is not going to be ideal, so to ensure that you're burning as much air as you can, it's worth it to dump in a bit more fuel. The additional fuel absorbs some energy, but up to a point the numbers will work out in favor of having the extra fuel.

 

[franzh]

To question 3:
When running rich, the evaporative effect of excess fuel tends to absorb combustion chamber heat, which will thus lead to lower combustion temperatures and help reduce detonation/pre-ignition. Running rich is commonly seen on small air cooled gasoline implements (mowers, generators, motor cycles, etc.) for this very reason.

When using a liquid fuel, running rich allows for a more ideal air-fuel saturation in the cylinder. This is less than Lambda=1, where the ideal thermochemcial process is complete. When running greater than Lambda=1, there is excess oxygen which can lead to higher combustion temperatures. One clear example is operating with Methanol. It has a much lower energy density than gasoline, almost half, but as it is mixed with air in the cylinder, it evaporates quickly and leads to a much cooler and denser air-fuel mixture.

As a side note, much of this theory is tossed out the window when running gaseous fuels since there is no evaporative effect of the liquid fuel.
Franz

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.

 

[Rob45]

"A lean air/fuel mixture burns with most efficiency, so much that the insulating boundary layer also gets consumed and the flame front touches the metal walls. At those locations, there is a dramatic rise in temperature, high enough to cause subsequent charges of air and fuel to spontaneously ignite."

I think this to be incorrect: the detonation produced while running lean without retarding the spark disturbs the insulating boundary layer - it is not "consumed" - and this dramatically increases heat transfer to cylinder walls, combustion chamber, and piston crown.
This is what melts pistons.

 

[drwebb]

A simple thermodynamic reason why more power is made at rich mixtures is that hydrocarbons release most of their energy on combustion to CO; so for any given displacement/compression ratio/engine speed, etc. complete combustion doesn't lead to highest power output.

Consider the combustion of methane as an example:

CH4 + 2O2 --> CO2 + 2H20

If you took introductory chemistry, recall that standard enthalpies of formation (Hf) allow calculation of the enthalpy (or heat output) of reaction:

Hrxn = Hf,products – Hf,reactants

Tables of Hf can be found in most general Chemistry textbooks. So for methane combustion we have:

Hrxn = [-395 + 2(-242)] – [-75 + 2(0)] = -804 KJ/mol

Now for incomplete combustion the reaction is:

CH4 + 3/2O2 --> CO + 2H2O
Hrxn = [-110 + 2(-242)] – [-75 + 3/2(0)] = -519 KJ/mol

So essentially 2/3 of the energy available from hydrocarbon combustion is generated in the first step (CH4 --> CO), leaving only 1/3 in the second step (CO --> CO2). Conceptually at least, for a given displacement/compression ratio/engine speed the most power would be produced IF we could stop the reaction after the first half of combustion, dump the CO and pull in a fresh charge of CH4. This is hardly practical, but does occur to a certain extent when the A/F ratio is slightly stoichiometrically rich. There are plots of emissions and power vs. A/F ratio in the literature showing that often maximum engine output and maximum CO emission (untreated) coincide.

Clearly in a fired engine both kinetic and thermodynamic effects are operative. But the size of the heat of combustion difference suggests to me that the chemical effect may be the dominant factor.

 

[btrueblood]

One more point to consider regarding running "lean" - if you have more free oxygen available, you are more likely to literally "burn" the iron in the pistons. With a rich or near-stoiciometric mixture, there is little free oxygen to oxidize the iron.

BT

 

[Bourke30]

Russell Bourke invented the perfect lean burn engine in 1938, in which detonation is harnessed and not shunned. "The Bourke Engine Documentary" manual will put all of these issues into perspective. If you want the facts behind his amazing engine, please see <

http://bourke-engine-project.com>

 

[ICman]

Speaking of "melting pistons":

I have a friend that ran his Harley at Bonneville a couple of years ago. On the return run, he noticed he lacked power. That was his last run of the week, so he packed up the bike and went home.

When he removed his front cylinder, there was NO piston. He had a rod with a wrist pin, but nothing else. He later found his piston in the exhaust, carburetor, crankcase, intake manifold, and pretty much everywhere except where it was supposed to be. He figured he had an intake leak on the front cylinder, causing a lean condition. Got so hot it vaporized the aluminum piston.

I know this has nothing to do with the basis of the thread, I just thought it was an interesting (true) story.

 

[TTmotors]

Thats 4271 degrees F. You sure this really happened??? For it to instantaniously vaporize the entire cylinder in one stroke (what had to happend because the next stroke wasnt building heat) you would of had to have one REALLLY hot explosion in that cylinder. 6000+ I would imagine to permiate the metal like that. It couldnt be a concurrent thing either. If it heated up in the precuring cycles to the melting temp it would have already disentegrated too much to build pressure and heat on the next cycle.

So in a couple milliseconds while the piston was comming up the bore it had to absorb 3000 degrees of heat to do what your saying. Also since the piston had to do this on the top of tdc how would the air get pumped back out into everything else? The piston would be gone so unless it was an effect of hot air rises (unlikely because of crankcase vacuum)?

Wouldnt the scorching air EAT EVERYTHING away?? No way anything in that cylinder would be recognizable after something like that happened. The car would not run correctly for very long although the piston event is gone air would seep in via the crank case pressure and the gas would still fire in. Contaminating the oil and having an open burn in the block. You sure he wasnt using uranium and lasers?? Or mayby just bsing?

-Travis-

 

[schwee]

Hmmm. I hope this is related.

I want to run an engine very lean (AFR~40:1 or so). I think I can get away with this in a pure detonation (HCCI) engine in a free piston configuration. Let's see, other conditions:

Burn rate should be very fast, maybe ten times as fast as spark ignition -- so fast there really is no flame front. (It is detonation ignition, after all - it essentially "runs on knock.").

Piston speed is very high - say around four times as fast as a "normal" crank engine. So pistons don't hang around very long to soak up heat from combustion? I don't really know.

So, I wonder -

1) Will this thing have the problem of the lean burn melting pistons (and cylinder walls/head)? Seems to me there's no flame front, so no insulating boundary layer at all, and no excess fuel to soak up heat. So it should melt everything -- but it apparently doesn't, in prototypes of related designs I've looked at. and,

2.) what about water injection? Other threads seem to say it's not worth it because of lost thermodynamic efficiency, but there also seems to be a concern of lots of excess heat, and water would seem to be the answer to soaking up heat and providing, like, internal cooling, whilst also making steam for added power.

So what's gonna happen? I plan on making a prototype and finding out, but I'm looking for some speculations to try to reduce the number of prototypes I will break or the amount of injury I do to myself. .

 

[crysta1c1ear]

schwee said:
I want to run an engine very lean (AFR~40:1 or so).
....
So, I wonder -
1) Will this thing have the problem of the lean burn melting pistons (and cylinder walls/head)?

Crystalclear:-
The stoichiometric AFR is about 14.7:1 I believe.
Three forteens are 42, so you're talking roughly about burning 1/3 the amount of fuel of a normal engine.

So releasing only 1/3 of the petrol energy of comparable engines, why are your worried about heat? Is it going to rev three times as fast making the total fuel burnt about the same? Is it going to have a wicked supercharger and three times the air, making the total fuel burnt about the same? If not, I'm missing something about the heat concern. I can see more heat from compressing to the point of ignition (higher compression ratio), but given that some engines are close to knocking anyway, I'm not even sure that that is a valid argument until one looks at fine detail.

 

[obanion]

HCCI, because of the nearly instant ignition across the entire air/fuel mix, causes a VERY rapid heat and pressure rise, compared to spark ignition. Which is why HCCI runs, at least so far to my knowledge, at lean to ultra lean burn. If you were to attempt to compression ignite a stoichometric air/fuel mix, the pressure and heat would blow apart any engine made for normal spark ignition.

But since you will be running about 1/3 of stoichometric, is shouldn't be a problem.

What I'm interested in is a mixed mode engine. HCCI for light load with a low octane fuel, switching to spark ignition for high load with a normal octane fuel.

 

[schwee]

crystal clear wrote:
"Is it going to rev three times as fast making the total fuel burnt about the same?"

Yes. It is, in fact, going to "rev" at least four times as fast as a standard crank piston engine, and it is going to fire at every stroke. At stoich it could conceivably deliver 16 times as many power strokes per minute as a 4-stroke crank piston engine. So, the total fuel burnt can be the same or even quite a lot more (2*, 3*?) even at 1/3 of stoich.

"Is it going to have a wicked supercharger and three times the air, making the total fuel burnt about the same?"

Yes. It is going to use a standard one- or two-stage centrifugal compressor from a small gas turbine engine (such as an Allied Signal JFS 100-13). Specs on this engine give a pressure ratio of 3:1, or at least three times the forced induction of a standard turbocharger.

obanion:
I'd totally agree with you about running an HCCI at stoich blowing things apart, if the engine had a crank and connecting rods. It doesn't. And I will in fact likely be running the engine at stoich during peak power demands. The original inventors did -- in fact, they even recommended running it a bit rich. They did not run it lean - or if they did, they didn't say anything about it.

So, I guess maybe I'll reframe the question(s):

1) What the heck is an "insulating boundary layer?" Is that air? Oil? what? Am I going to kill it with HCCI detonation at stoich, such that I heat up my cylinder walls?

2)Is the,um, "inconceivable" heat and pressure spike of full stoich HCCI combustion going to cause intolerable temps on the piston head and cylinder walls? Answer, apparently empirically, is no. So if not, why not?

And what about that water injection plan??

 

[Azmio]

Everyone,

I just want to share my opinions here. The old phrase "Never run lean or yer gunna burn a hole in yer pistons" may be right and may be wrong depending on where we define the reference point..

The reason is, detonation is unlikely to happen if the charge is either too lean or too rich. The reference point here is the stochiometric point of 14.7 (slightly less in the real application though). This is due to the fact that the combustion as a whole gets weaker if the charge is lean or rich. As a result, the combustion rate gets slower thus reducing the peak temperature and pressure.

The gasoline direct injection which is quite popular among the Japanese and European automakers runs around 40-50:1 air to fuel ratio. Some have even gone to 60:1. They all reported that the knocking tendency for their engine is less and they can either boost it up, runs on lower RON fuel or advance the ignition. This is why I think lean burn GDI engine is not likely to detonate.

On the other hand, back in the 80s where turbocharged F1 cars reached 200mph on the long straight at Silverstone, it has been reported that some of the cars run as low as 8:1 Air to fuel ratio. The engine guys purposely did that to cool of the piston top, intake charge, exhaust valves, exhaust valve bridge and the spark plug using the extra rich fuel.

Well to these F1 engine guys, they have the reason to believe in "Never run lean or yer gunna burn a hole in yer pistons". Perhaps the next question we should ask, where is the reference point, is it in the lean side or rich side of the stochiometry point?