Turbulence Induces for increasing burn rate 4

[automotivebreath]

Without turbulence in the combustion chamber we would burn the mixture at the laminar burning rate which is ten to twenty times slower than the turbulent rate. Numerous designs beyond common squish areas have been developed to shorten burn times. Does anyone have experience with turbulence inducers and knowledge of the potential impact on the combustion process?

 

[Fabrico]

automotivebreath
A consensus won't give you concrete answers. Have you considered using real time combustion pressure analysis equipment like TFX or ?

http://www.tfxengine.com/software7.html

 

[automotivebreath]

Al1,
Thanks for the clarification, I understand what you’re saying and its a very good point. I would think the window of opportunity for detonation at high RPM would be very small.

Fabrico,
For the most part this has been good aside for a few distractions, for me it's all educational. My main objective at this point is to try to learn what to expect with increased in cylinder turbulence intensity especially at low RPM. Eventually I will be attempting to reduce burn time at low RPM to some sort of minimum. If known hurdles exist, I’d rather learn about them now. I'm content with my testing methods for the time being.

 

[automotivebreath]

To those who may be following this thread, Rob wrote this on another forum. I love the way he simplifies the whole process.

Spark ignition (SI) engines all work on the principle of a turbulent flame front (TFF) consuming the air-fuel charge. A normal combustion event would be considered one where the spark ignites the air fuel mixture (a complex process in itself) and the flame propogates throughout the air-fuel mixture with this turbulent flame front. Skip the next paragraph if you know how stuff burns

I think most people have a pretty good idea of how a TFF works, but lets go through some of the details. First, think of a chamber (say a cube or a disc for simplicty) full of nice calm, still air-fuel mixture. A spark in the middle ignites, and begins to consume the mixture. Ideally, the flame front (in this case a laminar flame front, since the mixture is still) would form a spherical shell, as it progresses. Now, this flame front is propelled by a couple of forces. First, the mixture in the wake of the flame front is obviously heated by the combustion. This heat translates to an increase in pressure. This higher pressure burned gas compresses the mixture ahead of the flame front. Since the volume of the burned gases expands it helps accelerate the flame front. Think of blowing up a ballon. The compression of the end gas also raises its temperature. Flame speeds are higher in higher temperature mixtures.

For a turbulent flame front, consider instead of a calm chamber, a chamber full of turbulent eddies, of all size scales. As the flame front approaches one of these swirling eddies, the flame edge is 'torn' and spun around by the eddy, into fresh mixture. This helps to shred up the flame front, and helps to progress the burn of the mixture. In short, this is really why SI engines work at all, lol.

Now, this burning action is really a race. As you compress the combustible end gas, you get ever closer to the auto-ignition temperature. Auto-ignition is a process by where a series of branching chemical reactions (which mainly all have a very strong dependence on temperature) result in the combustion of a mixture, with no flame front. These reactions begin to oxide the mixture simply due to the thermal energy available from the high temperature. Now don't get me wrong, this is a VERY complex series of chemical reactions, but some of the basics help give a good understanding.

The auto-ignition is very dependent on the time history of the mixture. If you hold the mixture at a low temperature, it may not auto-ignite for a long time. Raise the temperature, and it ignites sooner. So if the TFF takes a long time consuming the mixture, the compression effects of the TFF are present for a longer time, and hence the temp. of the end gas is higher for longer.

If the end gas does auto-ignite before the TFF reaches it, or before the relatively cool cylinder wall quenches the flame, the end gas auto-ignites, or detonates. Now, if one corner of this chamber is the last to receive the flame front, and indeed does detonate, what is the result? The auto-igniting mixture essentially 'explodes' and sends a pressure wave across the chamber at the local speed of sound in the cylinder. This pressure wave is what is heard as 'knock' Oddly enough your ear picks up short pulses of a tone, as a 'knock', and not a ringing sound.

Hopefully this helps a bit, one of the prof's here also has a small program to help you hear how the short duration pulses are heard as knock. I will try and post it soon.

Sorry for the novel, I hope it is somewhat informative

-Rob

 

[MWPC]

Cleverlever,

I reviewed the link provided and took a brief look at the patent. I did not see specific figures for the intake valve closing point BBDC. If you are at liberty, would you please provide information on just how early you were closing the valve? I would be interested in fully closed and 0.020 or 0.050" angles.
Ken

 

[cleverlever]

Ken

The SAE papers mentioned in the patent give lots of the information you are looking for.

What you are asking is like asking what the valve timing is for an Atkinson engine. Its a ratio thing, the more restrictive the VE the higher the expansion ratio that can be tolerated.

The difference is you can get a lot more detonation limited max low speed power from an early close than a late close intake valve closing which is good for max power. However the slow burn is bad for light load fuel economy so you have to revert to the late close for faster burn at light load, requires a valve timing scenario similiar to i-VTEC.

Some of MS comments about combustion are completely the opposite of what I observed on 3 different dynos. Fast burn is not desireable at low rpm heavy load conditions because the rapid presure rise will cause an uneven combustion rate which will cause detonation.

Brief periods of detonation are tolerable without destroying an engine. Desireable Presure rise characteristics that impact detonation can be manipulated with twin spark plugs which can fire at the same time ,one at a time or at staggered timing.

 

[mattsooty]

Cleverlever,

My recommendation to you would be to go away and buy a decent book on IC engines (Heywood would be a start), read it and then try and understand it. Then maybe join the SAE and read some of their papers and try and get a better understanding of these things.

At present your knowledge seems very flawed and the complete opposite to what is a well documented, understood & even modelled mechanism.

Unless of course you think that your experience on a whopping '3 dynos' was somehow groundbreaking and brought fresh understadning to the IC engine community - in which case I suggest you get your paper in quickly!!!

MS

 

[patprimmer]

Come on guys play nice, otherwise this thread might disappear, and that would be a shame as it contains some very interesting information.

cleverlever.

You are mistaken re octane rating and burn rate.

Octane is a measure of detonation resistance and is independent of burn rate.

Try this link for some well researched data.

http://naca.larc.nasa.gov/

http://blizzard.rwic.und.edu/~nordlie/cars/gasoline.html

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.

 

[automotivebreath]

My suggestion is to try to understand the stated claims and the reason for the benefits. If statements like these are fact, it would be desirable to understand why.

…You can get a lot more detonation limited max low speed power from an early close than a late close intake valve closing which is good for max power…

… for light load fuel economy so you have to revert to the late close…

Assuming these statements are factual:

Why is an engine running under heavy load at low RPM is more detonation resistant with early intake valve close?

Why is it advantageous to revert to late close with an engine running under light load at low RPM?

 

[cleverlever]

Why am I being repeatedly asked to answer questions that are explained with relevant SAE references in my patent 4,961,406?

Obviously most of you are younger and far more educated than I am. That being said I am the first person who ever applied for a patent on the concept of using a detonation sensor to control intake valve events for the purpose of altering combustion rates. Have any of you read Toyotas patent 6,848,422? Pretty much says the same thing I patented
over a decade before their patent was issued.

Many years ago Champion Spark plug would host seminars and demonstrate the different burn rates of low octane and high octane fuel. They would use two clear tubes about 50 feet long and put low octane in one tube and high octane in the other. They would light them both at the same time and you could watch the low octane burn to the end of the tube faster than the high octane.

Maybe somebody could explain to me how this doesn't demonstrate the different burn rates of low octane and high octane fuel?

 

[turbomotor]

Automotivebreath,

I must confess that I have spent a lot more time modeling and testing CI engines than SI engines, so take my comments with that in mind. However, as complex a subject as detonation can be, with many variables, I think that some of the variables coincide with NOX generation in the CI engine. (The most basic variables are time and temperature, as one is working with involved chemical reactions and partial equilibrium of various hydrocarbon species.) So a consideration of Caterpillars ACERT engines, that use a modified Atkinson cycle to reduce NOX generation, may be instructive. I think that the dominant effect of early intake valve closing is the expansion and cooling of the charge prior to BDC, thus reducing the temperature profile throughout the compression stroke and initial combustion process. The cooler profile effectively retards the preparation of the end gases for detonation (or NOX generation in the CI engine). This may partially answer your first question above.

I also have a question. Considering all the interesting comments in this string, do any of the authors have experience testing or modeling HCCI engine technology? Those data may help answer some of the questions.

Vincent

 

[cleverlever]

Vincent

Bless you for your comments.

Clyde Bryant has several diesel patents that are basically the same thing I am saying about gasoline engines (6,279,550)

So I would like to know how many of these critics that are saying I don't understand the fundamentals of engine design have ever built an engine from scratch. Modeling is no better than the assumptions you plug into the front end of your model.

I built those engines and have seen the low speed high load detonation problems associated with Intake valve throttling

Moore (4,280,451) built several early close Atkinson cycle engines and his data was exactly the same as mine. The early close Atkinson cycle engines made more low speed torque than the base line engines.

Moore and I actually exchanged engines and retested them and confirmed each others data.

One thing that nobody talks about in late close Atkinson engines is the additonal heat dumped in the mixture when you over fill the cylinder and then push it back in the intake manifold to be consumed in the next intake stroke. Not a real bad deal at light load but the extra heat is a detonation disaster at heavy load.

 

[automotivebreath]

I think that the dominant effect of early intake valve closing is the expansion and cooling of the charge prior to BDC, thus reducing the temperature profile throughout the compression stroke and initial combustion process. The cooler profile effectively retards the preparation of the end gases for detonation (or NOX generation in the CI engine).

Turbomotor, Thanks for the insight, this makes good sense. The way I look at it is regardless of the timing for intake valve closing; once the valve is closed it is now recent history and all that remains is the charge that has been captured in the cylinder and how the contents and condition of the charge are related to present and future events leading to ignition. This charge has qualities that are related to the induction process as well as the early intake closing. In addition, as you stated, more time is available relative to crankshaft rotation for "expansion and cooling".

Why am I being repeatedly asked to answer questions that are explained with relevant SAE references in my patent 4,961,406?

Two reasons I have questions.

1. I am interested in learning more about the advantages of early intake closing and its relationship to the quality of combustion.

2. Some of your reasoning goes against current scientific belief. These people are very knowledgeable and can help us understand what is happening in the diminishing cylinder that contains a charge prepared for combustion with early intake valve closing.

 

[cleverlever]

"Reasoning against scientific belief"

I have no education but Mr Moore was a college grad. In his patent with the early close Atkinson cycle it says "With the sub-atmospheric compression,less than a full charge is used with each cycle. Thus,a larger engine would have to be used to obtain the same power as would be obtained on a full charge on a smaller engine."

However when he built a prototype engine and put it in a car he experienced an unexplainable increase in low speed torque without a trace of detonation. So we both, working without knowing about each other thought we could do what Fiala did 4,364,353 did and just limit low speed VE with a detonation sensor hooked to the throttle valve. That was a detonation disaster.

So what was it we didn't understand that was giving us power gains where we expected power losses?

Tuttle SAE 820408 had observed "similiar combustion duration" with early intake closing. Hara SAE 850074 observed an "increase in combustion duration" when he studied early intake closing. Neither tested early closing with an atkinson cycle and variable valve events.

Obviously Hara identified the illusive factor that was giving us the increase in low speed torque. Effectively we were increasing the octane of the fuel at low speed high load conditions with early intake closings.

What amazes me is this info has been known for 25 years and there are still educated people challenging its validity.

Let me close with these words of wisdom from Charles A Amann ex head of engine research , who refused to test my patented technology with an Atkinson expansion ratio - "As history is unfolded,it is seen that the experimenter sometimes finds the right answer for the wrong reason" (1985)

I couldn't have said it better with my own lips.

 

[automotivebreath]

…Obviously Hara identified the illusive factor that was giving us the increase in low speed torque. Effectively we were increasing the octane of the fuel at low speed high load conditions with early intake closings….

Although it’s not possible for an early intake valve closing event to increase the octane of the fuel, the early intake event may have an effect on the state of the mixture that decreases detonation sensitivity.

As turbomotor stated earlier “dominant effect of early intake valve closing is the expansion and cooling of the charge prior to BDC”. There could be more, the early closing allows for more time for the KE to be absorbed by the mixture to decrease turbulence in the compressed charge, this would effectively increase combustion duration.

 

[mattsooty]

Turbomotor,

Agree with a lot that you say but: -

The NOx production in a CI engine is a function of the amount of free 02 in the cylinder, the rate at which heat energy is imparted upon it and the max temp the Nitrogen Oxygen mixture attains. Whereas in an SI engine this is offset by the rate at which the O2 is utilized in hydrocarbon reactions, so it acnnot actually react with the N2 because it has already formed H2O, CO2, CO, SOx etc.

Also, wont the cool intake charge actually be heated by the combustion chambers temps more than the 'cooling' effect of expansion can occur with IVC BBDC (I dont know????)

As an aside, in an CI engine with CAT ACERT a lot of the NOx reduction strategy is injection retard, close coupled pilot injection and the 'bumpy camshaft' introducing EGR from the exhaust manifold. That's why this system suffers from terrible 'real world' transient reponse and pretty poor fuel economy.

Although, I totally agree that the propensity of knock is heavily reliant on the preparation of the end gases, ie temp upon combustion initiation. Thats what Octane is all about - the reduction of an engines propensity to knock.


Cleverlever

The seminar that you describe has nothing, in my mind, to do with octane rating - and everything to do with RVP.

To put it this way; liquid gasoline is pretty inert but gasoline vapour is easily combusted.

Take a fuel, such as RFG2 with 9 psi RVP and compare it with a 'lower temp fuel' of lets say 15psi RVP and it's obvious that at room temp there will be more vapour available to burn for the higher RVP fuel, hence the rate at which the liquid fuel vapourizes & combusts at _room temp_ will be much faster. Salesmen will show you anything.

By the way RVP is unrelated to Octane Number.....

As for the push back into the intake manifold that occurs with late IVC; this is a common tecnique in a VVT GDI engine, reducing pumping losses by drastically decreasing the actual compression ratio - hence better part load SFC.

In a PFI engine this results in terrible cycle to cycle repeatability and an awful idle, part of the detonation that you speak of is the result of inaccurate air prediction within the cylinder - spark advance is calculated on the amount of air entering the chamber, not the amount that stays in there.

Both of these are more pronounced with a larger residual gas fraction. Although this 'hot' residual will not increase knock propensity because of the action of the inert working fluid.

MS

 

[patprimmer]

Aromatic hydrocarbons are often used as octane boosters.

By coincidence, they have relatively high vapour pressures.

Therefore a test of more volatile low octane vs a test with less volatile high octane will show the high octane burning slower.

Try the same test with low octane diesel fuel and high octane methanol and see the result. It will be consistent with vapour pressure and exotherm from the reaction, and inconsistent to octane rating.

Much of the data necessary to work this out and understand how your observations work is contained in the links I provided.

As I understand it, a patent does not need to actually work to be issued in the USA, therefore issue of a patent is not proof of principal.

Also, test data can be misleading if we do not understand every variable that might be influencing the results, like in the Champion burn test.

One purpose of these fora is to question results of tests and established dogma, so as to understand and learn from each other.

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.

 

[automotivebreath]

MS
I am aware of large natural gas fired turbines that use an O2 rich environment to cool combustion temperatures to reduce NOx emissions. H2 is added to the combustion mixture to prevent flame out.

Is this sort of rational possible with the CI engine?

 

[automotivebreath]

I meant to ask, is this sort of rational possible with the SI engine.

 

[mattsooty]

automotivebreath

Basically what you are talking about is a very lean burn engine. I dont have any experience with this sort of application but can imagine that with an SI engine it would be neigh on impossible to initiate combustion, unless as you suggest another 'fuel' were added, in which case things start getting really complicated!!!

This combustion initiation problem is one reason that HCCI look attractive.

MS

 

[automotivebreath]

Mattsooty,
What air/fuel ratio would you consider very lean. I'm thinking in the 17:1 range.