A short study on ignition timing and combustion

WZ507 said:

...let's consider an example - the ignition timing employed on the often hot-rodded small block Chev V-8 engine. This engine has a wedge combustion chamber that is nothing special in terms of efficiency, and has historically employed ignition timing of ~ 36 deg BTDC. Why is this range of ignition timing chosen? A wedge combustion chamber is not particularly efficient, thus it requires significant spark lead to get the ignition event started.

Click to expand...

The "old" small-block Chrysler (LA; '64-87?) runs around 32 BTDC since the plug sits more centrally located in the head, so this helps affirm your statement.

As for the airplanes dropping rpm on a single plug, as someone pointed out earlier, the airplane engine's timing is optimized for the dual plug configuration. When you run on one mag, you're effectively running with retarded timing. You could probably recoup some lost performance by advancing the timing, but then you'd be getting into the realm of detonation again. This has already gone around the horn in this thread, and I'm sure numerous others.
BTW, Chrysler did go to dual plugs in their 3G hemi, but it only fires one on the compression stroke. The other fires the cylinder that's 180 cam degrees out (exhaust stroke, of course), using double coils. They say it's to help with emissions. Stupid Americans! There's a wiring kit out there that corrects that, but, without the associated ignition timing change, I wonder if there'd be anything gained.

Does this look familiar? Our Nortons should have so much squish area! 2004 Dodge Hemi out of my Durango.



Nathan

P.S. Thirteen pages! I think that breaks even Phil's record.

acotrel said:

The better way to fire twin plugs might be with a double ended coil for each cylinder and no wasted sparks. That way if one plug fouls the motor would cut out on one cylinder and you would be aware that one plug has stopped working and can do the necessary.

Click to expand...

If you connect both leads of a dual-lead coil to one cylinder, you'll need a bloody powerful coil. It takes almost no voltage for a plug to spark on the exhaust stroke. To fire across both plugs on the compression stroke takes almost twice the voltage of a normal coil. Your typical Boyer,Pazon etc will burn out if you use a coil that draws twice as much current to get that stronger spark. That's why the wasted spark, and each dual lead coil connected to a plug on either cylinder.

If one plugs fouls from oil or carbon, it still conducts and completes the circuit, and the other plug still fires. Both plugs don't stop working. The current gets through the fouled plug easier; it takes the low resistance path through the fouling, instead of the high resistance path across the plug gap.

The following crude analysis demonstrates that the combustion process is far more complex than the simple Spark- Burn- Exhaust model.

Take a 3 inch (1/4 ft) diameter cylinder bore, place a spark plug dead center in the head, so that the flame front travels only 1/8 ft from ignition point to the periphery of the cylinder. So far, this is not too different from many engines.

The flame front travels at the speed of sound, which for air = 49 x SQRT (Temperature) ft/sec. Temperature is in degrees Rankin = 460 + Temp deg F. For air at 100 deg F, sonic speed is approx 1150 ft/sec. The temperature after compression is higher than 100 deg, but as I said this is crude, and the higher actual temp. makes the following analysis even more significant.

Now the distance the flame front must travel, divided by the flame speed is 0.125 ft/1150 ft/sec = 0.1 millisecond. If the engine RPM is 3000, the time to rotate 180 degrees representing the power stroke (again crude...neglecting timing lead BTC and exhaust opening BBC) is 10 millisecond. This analysis shows the time available for complete burning in the cylinder is 100 times greater than the time required for the flame front to completely engulf the compressed mixture.

An exhaust gas analysis will show unburned hydrocarbons and carbon monoxide (incomplete combustion), which in consideration of the time available from the above, should be convincing that something far more complex is happening. How is it that with so much time available relative to flame propagation time, that fuel is still incompletely burned in the power stroke?

When I studied combustion as a graduate student in the late 60's, this question was largely unanswered, or some mostly unproven theories were floated about. Since then, the focus on emission abatement and improved fuel economy, has no doubt produced much research on the matter. My career went a different way, and I do not have first hand knowledge of what has been published in this area. Until someone does a literature review, we are all speculating.

My studies approached combustion using molecular gas dynamics. Mole gas considers the fuel carbon and hydrogen atoms, and the oxygen and nitrogen atoms as atomic "billiard balls", colliding in 3 dimensional space. Carbon and hydrogen atoms must collide with oxygen atoms before they can react, with the inert nitrogen atoms getting in the way. The reaction is not the simple chemical equation: C + O2 > CO2; 2H + O > H2O. In short, the reaction kinetics is a scientific study itself, and not an area which I have any particular expertise.

Before any exothermic reaction can occur, the petrol molecule must dissociate, that is, the carbon and hydrogen atoms must break their bonds and become free atoms, or ions. Dissociation requires elevated temperatures to cause the molecule to vibrate so vigorously that the intermolecular bond is broken. This temperature is created by the spark, and probably dissociates a very localized population of petrol molecules, raises the local temperature to above that necessary for chemical reaction, and a local environment is created for the billiard ball collisions to occur. This is the initial "kernel". The chemical reactions (burn) generates heat, raising the local temperature further, causing more dissociation of petrol molecules and the flame front spreads.

Mole gas theory explains unburned hydrocarbons as the result of statistical probability that some carbon and hydrogen atoms never collide with an oxygen atom. Another theory explains unburned hydrocarbons occur because the flame is quenched in the boundary layer where fuel/air contacts metal surfaces. To my knowledge, neither theory has been validated, and there are undoubtedly more theories than I am aware of.

It is not simply Spark-Burn-Exhaust.

Slick

Slick ,
Great reply..leaving more questions than answers...as a complete novice..but an interested one..could the in-complete combustion be due to the fact oxygen is consumed faster than the fuel can totaly "catch on fire" ..when Red was sent out to put out the oil well fires after the war,he used dyamite to absorb the oxygen ..is this the case in a cylinder ? the flame front is due to enriched oxygen being consumed allowing a "good" burn ,but this causes the lesser oxygen "mix" to remain dowsed? As you state fuel lying on surfaces tends to ressist burning , once the oxygen above this surface fuel is spent..no more can be burned. proberly why diesels are more effective..there fuel is burned in a "air burst" not after coating the surfaces? Plus the diesel is atomised via hundred's of bar.each droplet is fired through space burning like a metorite.picking up oxygen and leaving the spent gas behind.
Just a laymans responce .

@JohnRobertBould: Oct 03, 2014

John, I know enough about this complex process to know I don't know enough to be considered an expert, only knowledgeable...and only a little at that.

Yes, my discussion raises more questions than answers....because I understand the complexity but do not have the answers. As I said, I am sure much work has been done on internal engine combustion for emissions and fuel economy. For those interested and qualified, a literature review would be helpful.

Re: dynamite to extinguish oil fires - my understanding is the dynamite, which does not require oxygen to detonate, blows the air supplying the oxygen away from the flame momentarily, creating a vacuum. All sources of re-ignition (hot metal pipe and oil field detritus) must be removed from the vicinity or the fuel with re-ignite when the vacuum bubble collapses.

Re: Red Adair - did you know that after Red died, he went to hell, but the devil banished him to heaven after he put out the hell fire? Then Brown and Root put in air conditioning and they two were banished to heaven. Now the Devil refuses to accept any Texans.

Re: Diesel efficiency The Diesel injects fuel over part of the power stroke which keeps the power stroke mean cylinder pressure higher than the Otto engine. This results in the high torque of the Diesel.

Got to go to work...May comment more later.

Slick

So it sounds like the bottom line is that we can speculate forever, but we'll never know if dual plugging a Commando head offers any real world benefits until someone finds a satisfactory way to convert one, and then does enough dyno development to optimize the ignition for it. We'd have to compare the results to those from the same engine prior to the twin plug conversion. Any volunteers?

Ken

I don't think that because a Triumph or Norton, once dual plugged, doesn't show a performance gain on a dyno is a reason to believe the process isn't worthwhile. I'm dual plugging my big bore Triumph so as to be able to use the performance that is already there by keeping detonation at bay, especially with our lousy Claif. fuel. Pazon dual ignition advises setting the advance from 38* down to 32* with his set up and a stock engine, I believe reduced ignition lead will give combustion less time to heat soak pistons/head on each stroke and further toughen the engine against detonation, this has to be a good thing all by itself and worth the modification regardless of lack of hp gains on a dyno. My only concern is what happens if you foul one plug and now have one cylinder essentially running retarded, would I be able to tell without an EGT or CHT in each cylinder? Any thoughts?

lcrken said:

So it sounds like the bottom line is that we can speculate forever, but we'll never know if dual plugging a Commando head offers any real world benefits until someone finds a satisfactory way to convert one, and then does enough dyno development to optimize the ignition for it. We'd have to compare the results to those from the same engine prior to the twin plug conversion. Any volunteers?

Ken

Click to expand...

Since Comnoz has already made significant inroads in reducing ignition timing with his high energy multi-spark system (on a par with what dual-plugging might achieve?), we may already have a partial answer to the question, if….. Comnoz had a side by side dyno comparison of a low energy and his high energy ignition system on his current bike. In the event he doesn't have such data I'm sure he wouldn't mind totally dissembling his current system and installing a low energy ignition system on it and then going through the entire optimization of timing/fuel injection, etc for the new low energy ignition system so we can have an answer. Then of course as soon as he posts the results someone will point out that a high energy single plug system is not a dual-plug system so no correlation can be drawn between the two.

As I read the above, can't help but to think of the book "If I Ran the Circus", and the myriad chores little Morris McGurk (me/we) had for Mr Sneelock (Comnoz). Apologies to those outside the US that may not be familiar with Dr Seuss books.

After all, Mr. Comnoz is one of my friends.
He might even help out doing small odds and ends.
Doing little odd jobs, he could be of some aid…
Such as dynoing Nortons, to shed light in the dual-plug shade

marksterrtt said:

I don't think that because a Triumph or Norton, once dual plugged, doesn't show a performance gain on a dyno is a reason to believe the process isn't worthwhile. I'm dual plugging my big bore Triumph so as to be able to use the performance that is already there by keeping detonation at bay, especially with our lousy Claif. fuel. Pazon dual ignition advises setting the advance from 38* down to 32* with his set up and a stock engine, I believe reduced ignition lead will give combustion less time to heat soak pistons/head on each stroke and further toughen the engine against detonation, this has to be a good thing all by itself and worth the modification regardless of lack of hp gains on a dyno. My only concern is what happens if you foul one plug and now have one cylinder essentially running retarded, would I be able to tell without an EGT or CHT in each cylinder? Any thoughts?

Click to expand...

I concur with you on the benefits of multi-plugging or other means to take ignition timing out of a system while maintaining or increasing engine output.

I really like the looks of that Dodge Hemi combustion chamber, only thing missing are a couple of Singh grooves for Hobots sake...I wonder how close Dodge set up the squish?

texasSlick said:

The following crude analysis demonstrates that the combustion process is far more complex than the simple Spark- Burn- Exhaust model.

Take a 3 inch (1/4 ft) diameter cylinder bore, place a spark plug dead center in the head, so that the flame front travels only 1/8 ft from ignition point to the periphery of the cylinder. So far, this is not too different from many engines.

The flame front travels at the speed of sound, which for air = 49 x SQRT (Temperature) ft/sec. Temperature is in degrees Rankin = 460 + Temp deg F. For air at 100 deg F, sonic speed is approx 1150 ft/sec. The temperature after compression is higher than 100 deg, but as I said this is crude, and the higher actual temp. makes the following analysis even more significant.

Now the distance the flame front must travel, divided by the flame speed is 0.125 ft/1150 ft/sec = 0.1 millisecond. If the engine RPM is 3000, the time to rotate 180 degrees representing the power stroke (again crude...neglecting timing lead BTC and exhaust opening BBC) is 10 millisecond. This analysis shows the time available for complete burning in the cylinder is 100 times greater than the time required for the flame front to completely engulf the compressed mixture.

Click to expand...

The combustion process is admittedly complex, and I too am no more than a curious novice interested in learning a bit more about the process. Of the many different aspects of combustion you mention, I wanted pick up on one of them - flame speed - and add a bit more to it.

I'll start with a question. I have never heard of flame propagation rates in IC engines being anywhere near the speed of sound, but rather a decade or more lower than this, so perhaps you could enlighten as to how/why the flame speed is related to the speed of sound? Thank you.

Regarding flame speed in internal combustion engines, it is for sure a fascinating topic, as flame speed changes throughout a single combustion event as well as throughout the various operation speeds of the engine, thus flame speed is a complex continuum of speeds.

As previous submissions here have made clear, in a single combustion event, flame speed is very slow initially, i.e., during the "lag angle", where very little flame travel or pressure increase occurs over 10-20 deg of crankshaft rotation. Then during the "effective burn angle", typically ~ 40 total crankshaft degrees distributed evenly on both sides of TDC, flame speed is high as pressure and temperature both increase dramatically to provide peak cylinder pressure. Following maximum cylinder pressure, the flame speed is slow again in this "after burning" stage as pressure drops, due to the piston moving away from TDC, and the majority of combustion has occurred.

Regarding flame speed changing as a function of engine speed, it has always fascinated me how, serendipitously, flame speed increases proportionately with engine speed, so that essentially no change in ignition timing is required even if engine speed increases by a factor of 3X. This increase in flame speed is due to the accompanying increase in turbulence that occurs with higher piston speeds.

Using the above information we can actually back out the average flame speed over a range of engine operating speeds for a generic internal combustion engine. For this exercise we will consider only the effective burn angle, where productive burning occurs, and disregard the slow burn events occurring during the lag and after burn angles. The effective burn angle occurs over 40 crank deg, therefore we need to know the time interval required for 40 crank deg at various engine speeds. Per Slick's earlier comment regarding the flame propagating 0.125 ft to complete combustion in a 3" dia bore with centered spark plug, we then need to divide the distance (ft) by the time interval (sec) to back out the flame speed (ft/sec). I have done this for a range of engine speeds and have provided the data below in tabular and graphical formats. Please check my math so I don't mislead you or myself.

As you can see the flame speed is a continuum based on engine speed. And although we talk about the long drawn out combustion process with it's various fast and slow stages, you can see that at 7000 rpm the entire effective burn occurs in less than a millisecond (0.001 sec). There, while you just blinked your eye the engine fired somewhere between 100 and 400 times (depending on how fast a blinker you are).

Tex :
'My studies approached combustion using molecular gas dynamics. Mole gas considers the fuel carbon and hydrogen atoms, and the oxygen and nitrogen atoms as atomic "billiard balls", colliding in 3 dimensional space. Carbon and hydrogen atoms must collide with oxygen atoms before they can react, with the inert nitrogen atoms getting in the way. The reaction is not the simple chemical equation: C + O2 > CO2; 2H + O > H2O. In short, the reaction kinetics is a scientific study itself, and not an area which I have any particular expertise.

Before any exothermic reaction can occur, the petrol molecule must dissociate, that is, the carbon and hydrogen atoms must break their bonds and become free atoms, or ions. Dissociation requires elevated temperatures to cause the molecule to vibrate so vigorously that the intermolecular bond is broken. This temperature is created by the spark, and probably dissociates a very localized population of petrol molecules, raises the local temperature to above that necessary for chemical reaction, and a local environment is created for the billiard ball collisions to occur. This is the initial "kernel". The chemical reactions (burn) generates heat, raising the local temperature further, causing more dissociation of petrol molecules and the flame front spreads.'

My experience outside of motorcycling has for several years been in improving the way things go 'bang'. I think what you have said in the above paragraphs explains the effect of chemical composition on octane rating. However with a racing motorcycle, you have what you have - the permitted fuel is often controlled quite well at events. With chemical reactions in combustion processes and explosions, a finely divided, well dispersed fuel air mixture is more likely to give the situation where every bit of combustible is consumed. That is why I could never understand why some guys removed the squish band in Norton heads.
I haven't raced my commando much, however I've had a lot of experience with Triumphs. Much as I love Triumphs, I think the Norton is far superior. With the earlier Triumph 650s, I think the splay port head used on the pre-unit T100C will fit - the combustion chamber will be smaller and remove the need for the high crown pistons if you want high comp.- also give a squish band.
Another thing that works quite well - I once made a metal template of the combustion chamber in a 650 Triumph and used it as a guide to machine the edges of the crowns of two Hepolite pistons from a 350 BSA Goldie. I ended up with about an 8mm squish band right around the combustion chamber. The plug was a bit close to the crown, however the pistons were much lighter, and even with the flat crown I still got a good increase in comp. ratio. That was a very good motor.

All rise and Hail

Note compression-pressure heat ignited diesel engines are some what rpm limited d/t only able to start the combustion with piston about TDC so less time to pressurize in early zone of power stroke. If we were limited to just flame front speed we'd still be in hit-miss era engines, its the turbulence mixing that gets the heat-activation energy spread out in time. Not all engines benefit with squish * if the turbulence at TDC is active enough, even in Norton hemi's w/o any squish.

This is how old b/w moving and TV taught some of us, with some the best real time inside views.
https://www.youtube.com/watch?v=gyv91PwlMR8

For the methanol burners, real time insights
https://www.youtube.com/watch?v=OgZqAMKcGEk

And for some plasma power - skip crap to over 3 min point.
https://www.youtube.com/watch?v=P1S5_rRDkMY

marksterrtt said:

I really like the looks of that Dodge Hemi combustion chamber, only thing missing are a couple of Singh grooves for Hobots sake...I wonder how close Dodge set up the squish?

Click to expand...

Sadly, that seems to be where they missed, although it may be more for emissions than power. As pointed out, fuel won't burn in the tight squish areas, so Dodge left the pistons as flat tops. Add the gasket thickness (.010"), and there's really no active squish area.
It's hard to argue with results; 345 rear wheel horsepower from a 345 CID.

@WZ507

First off, a definition of terms is in order.

Flame front: A theoretical plane on one side of which no combustion occurs, and on the other side, some combustion occurs.

Flame speed: I would like Alcotrel to concur on this, as he is a chemist and been professionally involved in propellant combustion. I would define flame speed as the speed in which a combustible substance self ignites adjacent material to effect a moving flame. I will defer to Alcotrel's definition, as this subject is more in his professional realm, than mine.

A pressure disturbance in a gaseous medium, such as that caused by the ignition kernel, travels at the sonic velocity. This is a scientific given. Mole gas theory assumes if the reaction kinetics (speed at which a single carbon or hydrogen atom can combine with oxygen) is sufficiently fast, some reaction will keep pace with the sonic wave. NOTE: this does not mean every fuel atom reacts...only some. Mole gas theory is based on statistical probabilities, and the probability that all the fuel atoms will react in a very short time is nil, while the probability of a small number of atoms to react is high. Thus, the sonic wave propagates into a region where no combustion is occurring, with a small amount of fuel atoms reacting behind it....the definition of a flame front. It should then be obvious that after passage of the sonic wave, or flame front, further reactions (combustion) occurs in the wake.

The reaction in the wake region is not uniform; as you point out pressure and temperature quickly rise to a peak, then rapidly fall as the cylinder volume increases with crankshaft rotation. Mole gas theory accounts for this rise and fall as the probability of molecular collisions is high when the piston is near top, and the probability of molecular collisions falls as cylinder volume increases. With high collision probability, fuel atoms burn at a high rate, then the burn rate falls as the atoms must travel longer distances before meeting with an available atomic candidate for combustion. You may have notice that I have introduced the term "burn rate". This is the mass rate of combustion, i.e. pounds fuel / sec, and is more appropriate to mole gas theory than flame speed, as the latter is two dimensional, and molecular gas dynamics is three dimensional. This is consistent with your comments that "flame speed changes during the cycle", but coming from a molecular gas dynamics background, I would say burn rate changes.

Let's be perfectly clear, all I have said regarding molecular gas dynamic theory as applied to internal combustion engines is just that....a theory. It does adequately explain the combustion event and the pressure / temperature relationships in the engine, but just because "it fits" does not make it gospel.

I would like to add one word to my earlier post. Where I wrote "....divided by flame speed..." , make that "divided by flame front speed".

I must admit, your graphs and tables of "flame speed" vs rpm leave me clueless. This appears to be someone's computational data. Can you PM me a reference so I can study it further.

@Alcotrel

Alan, you write fine atomization and dispersion of fuel leads to more complete fuel combustion.

Absolutely! Before the petrol molecule can dissociate, it must first vaporize! Fine atomization and dispersion enhances vaporization.

Slick

All chemist give hail to Amedeo Avogadro who gave us the mole count.


Flame speed is how fast combustion progresses in set conditions w/o any extra mixing assistance. Flame front is the thin layer or sheet of combustion, which travels at chemical flame speed. Sound waves are compression/ratification waves and do not transport materials, though more molecules can collide in the compression part of sound wave passing by and echoing. Turbulence is what moves the flame speed flame front sheets around in time to matter. An ascending piston can slap a mixture compression/shock wave ahead of physical piston crown for local CR zone higher than just mechanical and intake flow kind. Mixture around TDC can get dense sluggish as a liquid rather than a gas. Initial temperature of the reactants comes into play too as some learn the hard way. With enough spark/plasma ignition you can burn water and Aluminum like rocket fuel.

I'm about to light off a big brush fire, cheated, built the pile around a bag of dry paper waste...

If the combustion velocity becomes sonic, that is detonation. You find that even with nitro compounds the burn accelerates, and if the critical height is exceeded even an unrestricted situation the burn velocity can reach the speed of sound, e.g. when burning waste gun propellant, if it is stacked too high, you can be bowled over - not something to near. (it is quite something to see, you can actually watch the burn accelerate). In a combustion chamber you have a controlled situation and the main consideration is the rate of pressure rise divided by the total time taken to reach maximum pressure. In the case of gun propellant this acceleration rate measure is called brisance. With liquid fuels, octane rating is probably the equivalent concept. If you mix fuel and oxygen very well in a combustion chamber in the stoichiometric (or close to it) mixture, you would usually get detonation. If the mixture is slightly richer, the process becomes slower and the energy is usable. This is the reason that those LNG gas tankers are usually safe if they catch on fire and BLEVE until they are almost empty. Then the ends come off the tank and it opens up down the seams. We actually had one of those happen here in Australia a few years ago - very spectacular.
What we talking about is getting the highest amount of energy from a given amount of fuel. I suggest the mixture in an engine needs to be homogeneous fuel and oxygen mix and just on the slightly rich side of the stoichiometric mixture, and it might depend a bit on the load conditions. And in the case of our engines the timing that max pressure occurs determines the amount of torque generated. It is about angularity of the internals.
I suggest that all increasing the comp. ratio does is move the game up a notch and more fuel is used. With methanol fuels in a bike engine, it moves the mixture requirements to an area where normal petrol carbs do not cope well. In both cases there is more energy produced, however in the case of many guys using methanol for racing, a lot of fuel is wasted because the carbs cannot cope.

A while ago I was involved with a company making gas fired aluminium melting furnaces. There was one strict rule - you cannot allow a confined space to stand with the explosive mixture inside - you will always get the spark. It is the reason the Honeywell controllers are so widely used to control the purge cycle in gas appliances. We had one guy who was standing on top of a gas fired oven and as he climbed down it attempted to change from a cubic to a spherical shape.

In early college dazes, I joined the cave club - one of its members a pryo nut case would bring a few grams of this and that to meeting to touch off, most would just go puff or spit a flash but a few would CRACK! with a supersonic shock flash from just a thimble loose power. Wonder about nano metals added to liquid fuels for a hotter flash and might even help conduct spark out further.

Check out this zero gravity combustion for an eye full.
https://www.youtube.com/watch?v=qQQ1OHW1_F4

Best view yet of gasoline spark ignition combustion in lab piston engine.
https://www.youtube.com/watch?v=p_P1-2octb4

hobot said:

In early college dazes, I joined the cave club - one of its members a pryo nut case would bring a few grams of this and that to meeting to touch off, most would just go puff or spit a flash but a few would CRACK! with a supersonic shock flash from just a thimble loose power. Wonder about nano metals added to liquid fuels for a hotter flash and might even help conduct spark out further.

Click to expand...

Sure, why not. After all, they've built engines that run on drops nitroglycerine (yes, I mean nitroglycerine, not nitromethane or nitrous oxide). You can build an internal combustion engine around almost anything that burns rapidly enough and produces enough energy. I remember guys experimenting with hypergolic fuels like hydrazine in race cars back in the '50s, but that doesn't mean it's a good idea!

Ken

I started a draft of this post prior to watching Hobot's post of a visual combustion chamber. I now complete the post, and will further comment at the end.

Most of us have our minds wrapped about the "macro" model of combustion. This model views the flame spread as that which would occur if Hobot set his grass pasture afire by setting a small fire at one edge, the flame moving as a wave across the field, fresh grass ahead of the flame front, and only ash behind.

The "micro" model as given by molecular gas dynamics, views the ignition source as "creating a wind which sweeps across the pasture at sonic speed", and then leaves in its wake micro combustion nuclei representing the collisions of combustible atoms. The result is like looking down on the pasture from a hovering helicopter and seeing the pasture sprinkled with specs of light, each light the result of atomic collisions that chemically react (burn). From each of these nuclei, flame spreads as in the macro model, meanwhile, more atomic collisions occur, creating new nuclei, and so on.

Which model represents the true events in the cylinder? I do not know, and do not know if anybody really knows. The truth may be some combination. I am only offering an alternative way of looking at the combustion event.

Irken wants to know if dual plugs will benefit combustion. If the macro model rules, then it should be intuitively obvious, dividing Hobot's pasture in half, and simultaneously setting a pilot fire at the edge of each half, that the entire pasture should be consumed in half the time.

The micro model does not offer the same intuitively obvious benefit, but my academic training allows me to visualize dual ignition sources setting off sonic waves that collide and reflect from each other, enhancing the micro combustion nuclei. I am sure Ircken is running high domed pistons and with an asymmetrical plug, the dome shades one side of the combustion chamber. Alcotrel has commented in this and other threads, that he has found the far side (away from the plug) of the piston to be coked up. This convinces me that symmetrical dual plugs would be of benefit with a high domed piston.

Furthermore, my experience as a small plane pilot, has shown me that dual plugs (with independent dual magnetos) always produces an increase in power. This has been noted in this thread and others on this forum. I stand that thermodynamics says if you get more power, you must burn more fuel. Hence, dual plugs provide more efficient combustion.

Molecular gas dynamics predicts all the equations of macro thermodynamics, and this theory is accepted science. The application of mole gas to the combustion event is more fuzzy.

Mole gas explains the detonation event, as well as how the addition of tetra ethyl lead, TEL, inhibits detonation.

Under conditions of high compression and/or higher than normal engine temperature (low coolant level or impaired cooling fins), the temperature of the compressed fuel charge may be high enough to dissociate nearly all of the available fuel atoms. With high compression, the atomic collisions are increased due to the dense packing of the gas mixture, and when the ignition event occurs, a higher frequency of atomic collisions results in a greater number that produce combustion nuclei. The "sprinkles of light" in Hobot's pasture is so dense, it is nearly saturated. The fuel charge combusts nearly completely in a instant....the definition of detonation.

The addition of TEL introduces intermediate reactions in the conversion of carbon and hydrogen atoms into CO2 and H2O. If the probability P of a fuel atom reacting with a molecule to produce an intermediate compound X1, is P1, and the probability of X1 reacting with another atom or molecule to produce X2, and finally compound X2 to form the end product of combustion is P3, then the overall reaction probability is P1 x P2 x P3. These probability numbers are small, less than one, (one represents 100% probability) thus the product of two or more small numbers is very small. In this way, the probability of combustible nuclei to form is less, and the saturation density of "specs of light" in Hobot's pasture is greatly reduced, holding off detonation.

Now back to Hobot's video.

I see the combustion as fitting the micro model. I do not see a flame wave spreading as that in the macro pasture analogy. I would like to see the event further slowed. Does anyone else see it this way, or am I just mole gas prejudiced?

Slick

Slick,

If you start the fire in the middle of the pasture would it spread outwards on two front as quickly as it would spread inward from opposite ends?
I could be way off base on this but it is something to think about.

Pete