Combustion Pressures (not another head gasket discussion)

[cygnusx1]

Ever since I popped out my first head gasket I have been thinking about combustion pressures. In all my engineering work so far, I have dealt with low pressure systems in manufacturing and processing equipment design. I never really though much about the PSI numbers that were achieved in the combustion chamber other than more boost makes more power. I did some google'ing and discovered that inside the combustion chamber, pressures reach on the order of thousands of PSI and, during detonation, factors of thousands of PSI. Has anyone done a combustion pressure analysis of their engine?

 

http://www.tfxengine.com/

 

Also, how does intake boost translate to combustion pressure. It is definitely NOT a 1:1 correlation. Ultimately, no matter what cam, boost, porting, ignition timing, mixture, or whatever we do to the motor, combustion pressure is what produces the torque on the crankshaft. Maximum combustion pressure is achieved during detonation. To maximize safe torque, an engine should be running as close to detonation as possible, without detonating. The more frequently, or the faster you can create, combustion pressure continuously, the more horsepower you can make.

 

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

 

Inside the chamber, the ignition occurs, the pressure is greatest at a singular point for a instantaneous amount of time. Then the pressure propogates from that point and spread out over the area of the piston and chamber. As it spreads it's force over the piston, it is not evenly distributed at first. Little by little it becomes more evenly distributed as the piston begins changing direction from compression stroke to power stroke.

 

Any engine gurus care to elaborate, clarify, correct, contribute to an intelligent discussion?

[Tony D]

What you are talking about is altering the BMEP (Brake Mean Effective Pressure) if you attach a Keniscope (osciliscope tracking a piezio transducer) to the combustion chamber (they make spark plugs for this) you can see a characteristic curve that looks somewhat like a 'flexible egg' that someone stuck their finger into on one side.

 

When you are N/A, the curve on the bottom of the 'egg' dips below zero on the x-axis, when you boost, the whole 'egg' moves up along the y-axis accordingly.

 

Boosting makes the egg 'less poked in' and makes for more effective average pressure on the head of the piston, which you said increases torque. The head gaskets, and the head studs will hold enormous pressures in that nice, steady rise to peak cylinder pressure (which is a ocmbination of the burn, burn rate, and piston position).

 

When you detonate, or run lean even, the burn rate is MUCH faster. Lean burning makes the flame front rise pressure much faster, and to make a poor analogy you get a 'water hammer effect' like when you shut of flowing water---the pressure in a lean burning cylinder rises so fast the strains induced on the joints and gaskets tend to start seeping. If you ever quickly turned on a large refinery pipeline, you see that gaskets that held during slow bleed valve pressurization sometimes weep when rapidly pressurized---this is what affects sealing integrity, and once the process starts it weakens the area so it's prone to do it again.

 

Detonation is plainly an explosion (not a burn)---technically the burn rate in an explosion is superfast. Tens of Thousands of feet per second, versus a few thosand feet per second (watch a long length of Detcord light off and you can actually see it go, highspeed phootography reveals a discernable flame front on the cord!). This superfast rise to pressure, as well as the much higher pressures produced (the burn rate determines the rate of pressure rise---for a combustion chambe that may be decreasing in size due to an upstroking piston that can make for exponential rise) simply overwhelms the gasket joint, and it fails catastrophically.

 

SB Fords were known to lift the heads off due to stud stretching. You can do that from a normal burn at enough BMEP (peak pressure specifically)...when that clamping force is affected, then you get blown head gaskets as well. But without the 'detonation signs' of broken piston rings and lands...or burned/holed pistons.

 

I haven't Keniscoped a Datsun L, but spent hundreds of hours on Lean-Burn Staionary Industrial Engines (Specifically the Cooper/Superior SGTA Series). What happens to the curve when you play with temperatures of incoming air, gas admission rates, even spark plug gap (for cylinder to cylinder comparison) is amazing.

 

The key is to get the peak pressure event to happen at the proper portion of the crank position. It's not simply making more pressure, it's making pressure when the crank is ready to make use of it! Spark timing comes into it bigtime, as well as fuel formulation. Slow burning fuels will limit the horsepower you can make (there was the 'high octane makes lower horsepower' argument, and this is why) but because you have such a range of spark timing adjustment you can usually compensate for the slower burn by running more advance. Faster burning fuels will tolerate less advance (lower octane), and the spark timing can be less for the same peak pressure---this is the point where you have to make the tradeoff between slower peak rise to pressure, versus a quicker rise. The less time the heat is in the chamber, the better, so a fast event is marginally better than a slower, prolonged event. I would not say you want to detonate --defined as uncontrolled fast burn. You want the 'fastest burn' that you can reliably control without it going to an uncontrolled state. This is where preignition comes into the game. Preignition is different than detonation...you can melt the ceramic inuslatiors on the plugs, burn the groundstraps off them...all sorts of nasty stuff. You don't hear it, it's like detonation but much more insidious and damaging---you can see it in the keniscope trace as a 'big hump' on the oeak pressure. Instead of the rapid detonation spike you can see, it's maybe 3/4 as high as detonation, but then 'humps at top' and is rounded. Basically all the effects of detonation, without the outward signs of a blown headgasket. Things just melt and break apart silently. This is the realm between controlled burn and outright detonation. Engines can run like that for quite a while damaging things without any outward signs. Leanburn stationary engines are suscptible to this---they run 22:1 AFR's and have precombustion chambers at 14.7:1 with 'flame holes' to keep the lean burn chamber ignitied. This probably illustrates the difference in burn times as you can literally ignite in the precom chamber with Stochiometric, and literally light off and relight several times during the pressur cycle a lean burn cylinder that burns so fast it can burn and go out and relight up to three or four times before the pressure peak target is reached.

 

Man, I'm going all over the place on this, is it making any sense---there is so much to discuss it's not funny!

[Tony D]

I'll add this. The N/A BMEP of say an SG8 Superior is something like 48, where the SGT8 (turbocharged) running 21psi of boost is something like 148. It was a big rise. Of course specific fuel consumption went up, to compensate for doubling the horsepower, but you can see the EFFECTIVE PRESSURE of the engine rose by almost 3X compared to the N/A rating of the same engine.

 

There is a calculus formula for finding pressure inside that Keniscope Trace, and that is how BMEP is calculated. Raising that egg above the 'Zero Axis' starts to make some real power!

[cygnusx1]

In Engineering School, I skipped an elective class called "Combustion Engine Theory". I have always regretted it. This is the heart of all the things we know and do here at Hybridz to make more power and I think it's definitley an area that all of us should educate ourselves on. TonyD, excellent and very informative post that just makes me want to learn even more. I am going to shop for some good text books on the topics involved. The more I learn, the more I want to learn which is why I don't really care about breaking car parts. Each experiment and failure is an opportunity to learn.

[naviathan]

Tony, I have to say in the last few years I have learned more from reading some of your posts than I have from any teacher I've ever been instructed by. Ever think about writing a book?

[WizardBlack]

Tony, I have to say in the last few years I have learned more from reading some of your posts than I have from any teacher I've ever been instructed by. Ever think about writing a book?

Have you read Corky Bell's book? It's a great one to take a look at, too.

[Tony D]

What you have is a balancing act of getting the fuel to actually keep expanding as it burns sliiiightly faster than the rate at which the piston is advancing down the bore and giving it room to expand.

 

This is why racing fuels have different formulations---different engines have different mean piston speeds and dictate a different burn rate to keep the pressures up. This is why I'm not a proponent of AvGas, it is formulated to burn in large diameter cylinders with very low piston speeds. VP makes several different formulations for 'import' engines now realizing that the 105 for a Chevy V8 really won't produce the same peak power in an Import with pistons outrunning the flame front!

 

Indeed, with a Keniscope you can actually see these phenomenon happening, and it is neat to watch.

 

Want to watch something boring, watch the trace on a Diesel...Same every time...

 

But gasoline and gasseous fueled engines? They are all over the place! LOL

[thehelix112]

Couple of points:

 

As I understand it, pre-ignition is when the engine `diesels'. Ie, the charge ignites under the increasing temperature as the pressure increases whilst in the compression stroke. The few articles I have read on this (for racing engines), said that by the time you notice it, your engine is already broken.

 

Also, a wizardblack suggested, I also suggest you read Corky's introduction to this. IIRC he says that the peak pressure does not increase that much, what happens is that the pressure is a little bit higher for a lot longer (which gives you the noticable increase in brake *mean* effective pressure).

 

To use a simple analogy, imagine riding a bike but only being able to push the pedal for half the down stroke. Obviously its going to be a lot harder to ride. The longer period of high pressure is analogous to being able to push on the pedal for longer during the down stroke.

 

Or something like that.. I dunno, i'm not a mecheng.

 

Dave

[Tony D]

Yes, that is the difference and why I refer to the 'characteristic curve', the pressure really doens't 'rise' much, but the 'shape of the egg' really gets altered, as does the relative position on the O-Scope relative to the X-Axis Baseline.

 

An N/A Pressure Curve (Trace) has most of the bottom below the axis while 'intaking', and a pronounced 'indent' under power.

 

A Highly Turbocharged Curve (Trace) is above the X-Axis due to the presence of pressure in the cylinder from the intake---it may barely touch the X-Axis during overlap/scavenging, but usually it will be above it, and the power stroke portion of the trace will be much less "indented" comparatively to an N/A Trace.

 

Yes, preigniton is compression ignition, but chances are it's inaudible and you will never notice it. The key is that it happens before you are igniting the mixture via the spark---(hence 'preignition'). Detonation happens after normal combustion has started, and is usually the result of secondary ignition sources working on terribly lean mixtures which burn WAAAAY faster than they should, and result in a real high pressure peak. You can see these on the scope as 'spikes' in the 'indented portion of the egg'.

I am looking over some stuff now regarding the new computer modeling and pressure tracing software that is out there, and the traces are WAY different than the equipment I used even 15-20 years ago! They aren't 'round CRT O-Scope Traces' anymore! LOL Anyway, this is also a good site that discusses Detonation and Preignition as well: http://www.contactmagazine.com/Issue54/EngineBasics.html

 

You can hear detonation happening, preignition--especially in stationary engines---is referred to many times as a 'silent killer'---you see the aftermath, and just scratch your head 'because we never heard anything'.

 

Here is a link to a paper that has some graphical traces of cylinder pressures, it's not the same thing you see on an O-Scope, but you can see the differences in curves somewhat---sorry it's a boring DD 60-Series, but it's a good engine... Page three has some nice graphs relating directly to different spark timing scenarios, and you can see what I was trying to explain above much easier.

http://me.engin.umich.edu/autolab/Publications/Adobe/P2004_05.PDF

[cygnusx1]

I am bringing this thread back into the daylight because it is so informative and we have a lot of new people here.

[Tony D]

I was going to invoke 'Third Dungeon Necroposter' nomenclature till I got to the bottom to see who dredged it up from the bottom of the heap!

[jasper]

From what I learned,

preignition is the spontaneous combustion of the fuel/air mixture, due to a higher temperature than the octane specification permits. This flame front (shock wave), then collides with the flame front (shock wave) from the normal igntion event.

The collision of the shock waves is the cause of the broken parts. Physical damage.

 

My understanding of detonation is not as good.

However, I personaly drove my 1978 Ford Fiesta (A GREAT CAR !!!) down the New York State Thruway, everyday, with my foot on the floor. About 90 mph top speed. (1100 cc if I rember correctly).

One day the coolant temp went well over full scale, And I lost all power, maybe 35 mph. It took a while to cool, and I regained power.This happened a few times. The end result was a cracked head. I beleive the cause was fuel starvation, a very lean condition.

 

Take it away Tony D............