Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Pushing the piston.

  1. Nov 22, 2009 #1
    I do not know if this is the correct place to post this subject, I apologize if it is not.

    This is something I am really confused about:

    What is really the gas that pushes the piston down in an car fuel engine ?

    Is that the flame front gas, or that still unburnt gas which is ahead that flame front which is compressed by the expanding burnt gas?.

    I understand, within the combustion chamber, right after the ignition, we have gas which spreads out as it gets burnt. This gas travels with a speed what we can call "low speed", as I heard it is around 40 cms/second?.

    But, on the other hand, we have gas pending of being burnt which is compressed, and it seems that "pressure wave" moves at sound speed, more or less.

    Ok, I rather shut up now, and leave any expert or experts to let me know more about this.

    Thanks a lot in advance.
     
    Last edited: Nov 22, 2009
  2. jcsd
  3. Nov 22, 2009 #2
    Welcome.

    It's a combinatoin of both of those things.

    The burnt gases are at much higher pressures, which pushes down. As you say as the flame propogates the unburnt mixture also pushes down and its pressure rises.

    Majority is due to high pressure combusted products.
     
  4. Nov 22, 2009 #3
    Thanks a lot xxChrisxx for your quick response.

    So, shall we say that those pending to be burnt and pressurized gases act like a shock absorber?.

    In other words, the piston is initially pushed by a low pressure of unburnt gases and it is when the flame front reaches the piston when we definetively apply the combustion force to it.

    Any idea about the crankshaft angle (from TDC) when this flame front reaches the piston?

    Thanks again in advance.
     
  5. Nov 22, 2009 #4
    Well the flame front reaching the piston is of no consequence really. The flame progates radially outwars from the spark plug.

    The peak combustion event (pressure peak) depends on the engine, typically its a few degrees aTDC (maybe 5 - 20). This is really just a guess, as it can vary depending on timing, temperature etcetc.
     
  6. Nov 22, 2009 #5
    I get the feeling that your mental model shows the pressure variation within the cylinder to be happening at on a slower time scale than the motion of the piston itself. Consider this: a cylinder with a 5 inch stroke turning 5000 rpm has an average piston speed of 4170 ft per minute or about 70 ft/sec. Now compare that to the speed of sound, say on the order of 800 ft/sec. So you see, as far as the piston is concerned, the entire volume is essentially at the same pressure (I mean the pressure at the 'burning' mixture near the spark plug and the 'unburned mixture' near the piston equalize over a very small time compared to the time it takes for the piston to descend from top center to bottom center...
     
  7. Nov 23, 2009 #6

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    The pressure waves propogate faster than the flame front.
     
  8. Nov 23, 2009 #7
    Thanks you all for your responses.

    So, am I wrong if I say that there is no a well defined frontier between the burnt and unburnt gas and that the flame has a ramdon "advance" in that chaotic distribution due to the gas turbulence created by the head intake design?.

    Thanks.
     
  9. Nov 23, 2009 #8

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    Hmmmn. The boundary is quite well defined; ahead of the flame front is a definite area of completely unburned mixture. The area behind the flame front is a mixture of combustion products, and species undergoing combustion. The advance of the flame front isn't random (though its motion will be somewhat chaotic), but it's controlled heavily by inlet port and valve design, and by the shape of the combustion bowl.
     
  10. Nov 26, 2009 #9
    The air pressure inside an internal combustion engine with a compression ratio of 10:1 is 147 psi, or 10 atmospheres. Naturally, the air-fuel mixture isn't plain air, and there are heat issues to deal with.

    Interestingly, in an ideal gas, the speed of sound depends on temperature alone, as the pressure and density have equal but opposite effects.

    It's the heat issues, or more precisely, the temperature, which determines the speed of sound in a combustion chamber.

    I couldn't find a good source for that info, but I do know combustion temperatures for most fuels range between 3,500 and 4,000 deg F, or about 2,100 deg C.

    Plugging that into this hand online app gives us a speed of sound of 3203 ft/s. Since the standard day speed of sound is 1145 ft/s, that/s 2.8 times greater!

    I would agree that the flame front is sub-MACH.
     
  11. Nov 26, 2009 #10

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    This isn't really true. Your statement assumes that the throttle is wide open, barometric pressure is 14.7psi, and a volumetric efficiency of 100% (these conditions are rarely met). Also, the OP was asking about what pushes the piston; the pressure you've quoted is just the compression pressure. The peak cylinder pressure is far, far higher than this.
     
  12. Nov 26, 2009 #11

    Ranger Mike

    User Avatar
    Science Advisor

    Brewnog and XXChrisXX are THE ones with all the answers , in my opinion...they have access to all the empirical data. One aspect that should be considered in the force to push a piston conversation is the actual configuration of the piston and combustion chamber.But it is probably the wrong time to bring this up so will let the OP who started this post to decide.
     
  13. Nov 27, 2009 #12
    Compression ratios are what occur in a flameless isothermic cylinder, brewnog. When combustion occurs, the pressure becomes much greater.

    No kidding. The OP also asked about the whether the flame front travelled at the speed of sound, and that's the portion of his post I was addressing, not "what pushes the piston."
     
  14. Nov 27, 2009 #13

    Ranger Mike

    User Avatar
    Science Advisor

    The process of burning the fuel in the combustion chamber is not simple.

    We better agree that when we are talking about the whole question of "GAS" pushing the piston refers to the mixture of air and fuel. Lets look at what happens in this process.

    Burn Duration- The ideal burn duration is when the peak combustion pressure occurs at about 15-17 crankshaft degrees after top dead center (ATDC). This is when we have the greatest pressure force on the crankshaft at the optimum crankshaft angle and for the maximum possible power stroke duration. When the engine achieves this, then Maximum Brake Torque is produced. Shift this position and less Maximum Brake Torque is produced. Because the gas combustion is designed to burn at a constant rate, ignition must occur long before the peak combustion pressure 16 crankshaft degrees ATDC.

    Now throw in the fact that faster the engine is turning, the shorter the time for the crankshaft angle to reach that 16 degrees ATDC position. The burn time of the gas is controlled by the chemical makeup of the fuel itself, the temperature of the fuel, and how well it is mixed with the required Air..really the good stuff in Air...oxygen. Octane additives do not change the burn rate of the gas. Racing fuel has a different chemical design so that it will burn faster to keep up with high RPM required in race engines. Octane rating is NOT involved in this fuel burn time. Combustion chamber shape will also affect burn time ( hemi, semi hemi , wedges etc). As the engine RPM increases, the ignition spark must be advanced many crankshaft degrees BEFORE TDC to have the peak combustion pressure occur at the desired 16 degrees AFTER TDC. This is why it is critical to know the advance curve of the distributor ( old days) ignition timing curve ,,today's lingo.

    Burn Rate or specifically Gas Burn Rate. Several factors affect the burn rate (flame speed) of the gas. The air-fuel ratio (a/f/r) affects burn rate. Mixtures with a/f/r of less than 11:1 are too rich and way down on power, a/f/r greater than 20:1 are too lean and may burn valves, pistons..not good . The fastest burn rate is at 17:1 but that is far to lean for and way to lean for maximum power. Best power is achieved at a/f/r of 12.6:1. Now when we look at restrictor plate racing ( like Daytona, Talledega) High RPM lean downs have become popular because Leaner a/f/r of 13.5 to 14.5:1 can deliver more power at high RPMs, but combustion temperatures will be higher. This improves the chances for detonation. Forget about this unless you want your engine in multiple pieces..

    Lets look at what we are really doing when we talk about Compression Ratio. We are talking about squeezing the a/f/r to make a denser charge..why??? to effect the gas burn rate. A higher charge density burns faster. Charge density is a function of gas pressures and gas temperature. As charge density increases, burn rate also increases. (Compression Ratio of 11:1 . will burn faster than compression ratio of 8:1). Gas burn rate will increase exponentially with pressure and temperature.

    More a/f/r problems to ponder. In any IC we have always had a problem with nonuniform distribution of air and fuel molecules within the gas mixture. This is caused by intake runner lengths , fuel puddeling, not being uniformly atomized, all which effect the a/f/r which does effect the burn rate. If the a/f/r in one cylinder is different than another where the spark plug is located , then we do not have max HP. The more uniform the a/f/r at each spark plug, the greater the probability of consistent ignition for each power stroke. Inert effects like nitrogen gas in the air we breathe, effect burn rate but not much we ca do about it. Cold chamber walls tend to reduce gas temperature which can quench the gas from burning, or at least slow it down due to temperature drop.

    Recent engine improvements are CD ignitions that produce a long spark over many crankshaft degrease during low RPM and multiple spark pulsed over many crank degrees at HIGH RPM to more effectively burn the a/f/r. BTW, Multiple sparks will not make the combustion gas burn any faster. Another trivia point - top fuel drag engines designed around the classic Chrysler Hemi use two spark plugs per cylinder are too close together to form two flame fronts. The twin plugs are used to assure Ignition of the cylinder. The spark plugs are totally eroded 1/8 mile into the run and the engine is " dieseling " or all practical purposes on the verge of hydrostatic lock. The dragster drive (pilot) shuts off the fuel at the end of the run...

    Another trend recently is the turbulence and swirling actions due to the intake port shape ( D intake ports, Swirl CC heads, and piston quench areas all in an attempt to replace that lean mixture with a normal mixture while the spark is still arcing.

    Mechanically combustion chamber design will affect burn rate. A hemispherical chamber with a high surface to volume ratio, will cool the gas more, and make it burn slower (reduced charge density). Those engines need more advanced ignition timing to compensate for this slower burn time. This slower burn time also reduces pumping efficiency. The spark plug location also affects burn time as mentioned above. To use extremes as examples, if the spark plug was located at one edge of the chamber, ( wedge design cc) it would take twice as long to burn all the gas across the chamber as a spark plug located in the center of the chamber. The hemi uses centered spark plug location and is most effective shape to distribute the pressure uniformly onto the piston.

    Ok now to the rat killing!! Small block Chevy V8. ignition advance set at 23 crankshaft degrees BTDC at 3000 RPM.At 23 degrees BTDC, the ignition coil fires, and the high voltage ionizes the gas between the spark plug electrodes. At some point of ionization, the ignition spark arcs across the gap and starts the burn process. This happens while the piston is still moving towards the cylinder head. Cylinder pressure is now increasing because of both the piston advancing towards the head (compression) and also because of the expansion of the burning gas. Because the gas is burning and not exploding, this pressure rise remains linear and within the design limits of the engine, while the piston continues to move closer to the head. At about 10 degrees BTDC, the burning expanding gas pressure is about equal to the compression pressure of the piston motion alone. During that last 10 degrees to TDC, we are more than doubling the cylinder compression pressure and charge density, because of the burning gas and its snowball expansion effect. This speeds up the gas burn rate, which makes the gas expand faster, which speeds up the burn rate, which makes the gas expand faster.

    Detonation or when linear burn of the a/f/r goes wrong! Detonation is the biggest producer of multi piece engines. Excessive heat, a random hot spots causing pre ignition, premature fuel detonation will cause the gas to spontaneously explode. These are all bad things and remind me of very unpleasant and expensive earlier racing experiences..ugh lets move forward.. We need to keep the gas burning and expanding as the piston reaches the top of its stroke, and at the same time, never increase the gas temperature to its spontaneous combustion temperature. This is where gasoline octane comes into play. Increasing the octane rating of gasoline, increases the temperature required to promote spontaneous combustion of the gas. As long as the octane rating is high enough, the gas continues a controlled burn and associated linear expansion rate as the piston approaches TDC. While approaching TDC, this cylinder pressure acts as a brake and resists the rising piston motion. This braking action steals power from the engine. This concept is referred to as the pumping efficiency of the engine. The sooner the burn rate starts, the more the engine pumping efficiency will be reduced. At TDC, the combustion chamber shape can also add additional virtual octane to the gas through the process of quenching the temperature of the gas. By the time the piston crosses TDC, we have some pretty serious burn rate and gas expansion happening here. This is due to the effect of pressure rise and temperature rise as the piston approached TDC.

    Now the piston is going down and the cylinder displacement volume is increasing. The tremendous burn rate that has now been achieved the burning and expanding gases are expanding faster than the cylinder volume is increasing, so power stroke force is applied to the piston and pushes it down. In most auto engines, this compression pressure is now approaching 800-1200 psi (depending upon the compression ratio). This burn rate continues to raise the cylinder pressure until about 15 to 20 degrees ATDC (about 1200-2500 psi) which is peak compression pressure . The piston is now receiving maximum force from the power stroke, referred to as maximum brake torque . If all the gas stays below spontaneous combustion temperatures during this time, then the maximum cylinder pressure will power the piston down with great force and for the longest possible duration. As the piston moves further down past peak combustion pressure , and the expanding gas continues to burn, a point is reached where the expanding gases start to burn out, and can't keep up with the increasing cylinder displacement. When this happens, the force applied to the piston by the expanding gas starts to diminish, and the power stroke is rapidly nearing its end. This usually happens at 20-25 degrees ATDC. Ideally, the gas has all been consumed by this time. This concludes a normal power stroke which had no pre-ignition and no detonation.

    As mentioned earlier, the shape of the combustion chamber can help to prevent detonation in two ways. The shape of the piston crown as it approaches the shape of the cylinder head, can create tremendous turbulence in the gas. This squishing of the gas mixture causes swirling and tumbling actions which causes shear tearing of the air & fuel molecules, which results in better homogenization. This improved mixing of the gas makes the gas burn faster. The same gas when burned faster has less time for spontaneous combustion. The faster the burn, the less time that is available for detonation to take place. Another advantage of a faster burn is that ignition spark doesn't need as much advance. With less ignition advance, there is less time to build burn pressures before reaching TDC. This reduces braking action to the piston compression pressure, which increases pumping efficiency of the engine. This results in less power wasted to pump the engine cylinders.

    Quench is a combustion chamber design advantage that lets you run compression ratios one point higher. The a/f/r in direct contact with the metal cylinder walls, piston crown, and the cylinder head surface; is cooler because the metal absorbs heat from the gas (the metal is cool as compared to the burn flame temperature which can reach 3000F degrees plus). Because this thin layer is cooler, it does not burn and results in what is called a boundary layer of gas attached to the metal surfaces. This boundary layer is only a few molecules thick, but acts as an insulator which keeps the burning gas temperature from direct contact with the metal engine parts. This contains the gas burn temperature and prevents imparting excessive heat directly into the metal engine parts, which could melt pistons. Like all insulators, it leaks some combustion heat into the metal parts and the engine cooling system, cylinder walls and engine oil all must absorb that heat. At TDC, portions of the piston crown get within about .040 inch from the cylinder head (quench or squish region), and the close proximity of boundary layers quenches any attempt for gas in that region to burn. The .040 inch gap is hundreds of times thicker than the boundary layers, but the cooling effect quenches any gas trapped there. When that gas cannot burn, it reduces the chamber temperature which results in less heat available to cause detonation during the time from TDC to 16 degrees ATDC (after the squish time). This cooling effect is referred to as virtual octane because the cooler gas escaping the squish area as we leave TDC, steals heat from the burning gas, which reduces the chances of spontaneous combustion. This quenching effect results in a virtual octane increase. It has been found that the squish region has little effect if the piston to head squish clearance is 0.060 inch or greater. The optimum quench clearance is 0.040 inch. One more advantage of the quench area is t squeeze the a/f/r toward the more open area of the combustion chamber at supersonic rates to further compact the a/f/r into the CC. Thus more dense a/f/r, more H.P. at a slightly higher compression ratio.

    This is about all I know about the whole force acting on piston thing, hope it helps. Note below diagram of otto cylce
    ifin I gotta explain 1,2,3 and 4 you got no business on this post!
     

    Attached Files:

    Last edited by a moderator: Sep 11, 2015
  15. Nov 27, 2009 #14

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    Correct.

    Goodo. I was just pointing out that your statement was only referring to cylinder pressure under motored conditions, and is profoundly different to a firing cylinder.
     
  16. Nov 27, 2009 #15
    Thanks for clarifying. :)
     
  17. Dec 2, 2009 #16
    nice explaination ranger mike

    dr
     
  18. Dec 2, 2009 #17

    Ranger Mike

    User Avatar
    Science Advisor

    Thanks Doctor
    Mopars RULE!!!
     
  19. Dec 31, 2009 #18
    hey guys hi

    actually i was reading your conversation and at the instant a question was generated in my mind!!!!and that was if hydrogen is burned in the piston what would be the temperature change inside it and how much pressure will its product-which is water- generate!!!!!!

    anyone of u cud answer it!!!!!thanx
     
  20. Jan 1, 2010 #19
    It depends on the ratio of Hydrogen to fuel, hydren to air and hydrogen and fuel to air. Fuel mixing is tricky and, to be honest I can't remember all of it.

    I also have a New Years hang over,

    Assuming enough air for full combustion of both fuel and H2,

    The hydrogen will burn totally, but the overall power would be reduced and the temperature of the combustion would be reduced slightly also. There would be more water in the exhaust gas.

    Assuming Octane.

    C8H18 + 12.5(O2 +3.86N2) ----> 9H2O + 8CO2 + 48.25N2
    add 2 mol H2 for enery mol fuel.

    C8H18 + 13.5(O2 +3.86N2) + 2H2 ----> 11H2O + 8CO2 + 52.11N2
    So for every mol of Hydrogen you add you need to add 1 mol of oxygen.

    Basically adding H2 just dilutes the fuel, but doesn't make it run lean, as there is enough air for the pertrol to burn stoichiometrically and the h2 to burn stoich.



    Assuming not enough Air for full combustion.
    This case is what you get when you get people who try to run their car on water using little electrodes.

    In this case, there is a set amount of fuel and air drawn in, but you are then dumping H2 in with it. This upsets the stiochiometric balance of the fuel, and the reason why they see marginal increases in fuel efficiency is that they are basically running the engine lean.

    C8H18 + 12.5(O2 +3.86N2) + 2H2 ----> ??????????????

    Becuase we don't know the composition of the exhaust (it will contain lots of crap due to oncomplete combustion of possibly both fuel and hydrogen).

    Running lean long term will cause the temperature to increase, and it ends up ruining the piston, liners, block. (basically a whole manner of problems). It also means they are getting less power.

    With incomplete combustion it's very hard to just work out an equation as the exhaust gas composition changes. You get unburnt fuel, carbon monoxide, NOx, and a whole manner of other crud.
     
  21. Jan 2, 2010 #20
    ok all right i got it!!!!that was a good explanation!!!!
    i had another question to ask that if say we electrolyse a dilute solution of NaCl which will give us hydrgen and oxygen in 2:1!!!for this electrloysis i think using a car battery(which is 50-60 AMP) would be enough to produce sufficient amount of these gases!!!!
    if we introduce these gases into the piston and close the air inlet valve(which will stop nitrogen to come in)the reaction will go as follows

    H2+1/2O2------H2O

    the reaction is exothermic which has an enthalpy change of -286kJ per mole of hydrogen!!!the specific heat capacity and latent heat of vapourisation of water and specific heat capacity of water vapours i suppose is not much high so i think the temperature rise would be great and enough for making those vapours to push the piston and drive a car of say 1000cc or 800cc!!!!!

    am i correct or totally wrong????what do you say
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Pushing the piston.
  1. Piston angle (Replies: 3)

  2. Pascal's law and pistons (Replies: 12)

  3. Car engine piston (Replies: 10)

Loading...