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

Exhaust linearisation in a moving piston

  1. Jul 16, 2010 #1
    Hope someone is able to assist. I have been musing about generating greater
    efficiencies in ICEs and pose the following:

    * Consider the exhaust phase of an internal combustion engine (diesel or gasoline)
    in which the exhaust stroke is pushing hot gases and particulates out of a simple valve

    * Consider the valve configuration as being either asymmetrical (one side of the centre
    line of the cylinder head) or sysmmetrical (both sodes of the cylinder head).

    * Consider an availability of a high pressure air source (derived by independent means) at
    a pressure of up to 300 or 400 psi.

    *Assuming that the exhaust gases are in a highly disordered state and the objective is
    to have them exist the valve configuration as quickly as possible....

    In what way could introduction of high pressure air assist in this goal (presumably as
    a high pressure)?

    Would a very fine, high pressure jet down the middle of the cylinder (just astride the
    spark plug or injector say) impose any kind of more ordered flow out of the exhaust valve(s)?

  2. jcsd
  3. Jul 16, 2010 #2

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    I am not an automotive engineer such as many on this forum..and i am sure they can give you more emperical data
    when measuring on-car (or -truck) exhaust backpressures, that a very good simple approximation of the exhaust gas flowrate is to use the engine volume x rpm x .5 (for a four-stroke) then correct for gas temperature. The error in assuming 100% volumetric efficiency about equals the error in neglecting the products of combustion.
    For example, a ZR-1 Corvette engine running 6000 rpm with a gas temperature of 1400 F showed an exhaust flow rate around 2000 cfm.
    This worked well for sizing exhaust systems, and as a "reality check" when comparing bench to vehicle backpressure data. I suspect that 400 psi of air is slower than the exhaust psi if we wanted to g o thru the math. also it takes power to compress air,,where will this come from?
  4. Jul 16, 2010 #3
    Car engines are already able to push out pretty much 100% of the exhaust gases anyway, there is no need to try to increase this. Infact many engine attempt to retain some exhaust gas in cylinder to create a cleaner burn. See: EGR http://en.wikipedia.org/wiki/Exhaust_gas_recirculation

    Pressures in the cylinder already exceed 400 (ish)bar, adding a few psi will not make any difference. Not only that but as Mike pointed out it takes energy to compress the air. So you are spending lots and adding complexity to gain very little.
  5. Jul 16, 2010 #4
    Very interesting subject. Not all engines around today are moving 100% of the exhaust gas. Heavy diesel engines have exhaust reversion from pressure differences of the cylinder and intake tract when the intake first cracks open. I wouldn't doubt the same from a regular automotive vehicle to a point. Turbochargers do not help much since it takes pressure to rotate the turbine wheel. Depending on what your camshaft attributes are, you can check how the blow-down phase itself is occurring. If the exhaust valve opens sooner after TDC, we have a stronger blow-down effect and in turn produces less work for the piston to exhaust the combustion gasses by pushing. Opposite for a later opening after TDC. More gas stays in the cylinder because the CP is becoming lower as our piston moves towards BDC before the exh. valve opens, there will be a some evacuation even with the lower pressure against the valve.

    Things you must consider when evacuating a cylinder.
    Exhaust valve design
    Exhaust valve modifications (if used)
    Gas velocity in exh. port and in exh. pipes
    Port length (exhaust port is generally short)
    Port Cross Section Area
    Port quality (turbulence will lengthen the evac time)
    Turbocharger size
    Exhaust headers or manifold
    Exhaust piping diameter
    Exhaust piping length
    Muffler design/flow capacity

    In regards to 4-valve gasser heads, I do not know how much more adequate the exhaust is removed.

    What is the engine you are experimenting with?
  6. Jul 16, 2010 #5
    Thanks to all who contributed. Fascinating topic. I'm not working with any specific engine rather a simplistic, idealized model of a piston compressing a hot gas out a small number of cylindrical valves.

    The general problem i'm addressing is exhaust efficiency. My off the wall reasoning is that in perfect efficiency each exhaust molecule should effect a linear transit in the direction
    of piston motion directly to a valve and transit the valve open with minimal delay.

    This contrasts with the feddback that indicates that exhaust retention can aid other
    aspects of the combustion process, however i merely want to focus on exhaust efficiency
    rather than overall engine performance.

    In a real piston the gases are chaotic in motion.

    Question: Can a configuration of injected gas impose any better structure to the
    piston which minimises the time of transit for any given gas molecule to the exhaust
    opening and out of it.


    Would a 300-400 psi stream of air injected at the bottom, mid point or near top of the exhaust stroke at any orientation, impose a more orderly flow of exhaust gases. e,g
    1 or 2 angled in streams started ahead of the exhaust wave proper as the piston
    comes past BDC?


    Would injected air streams at the mid or near top of the stroke do the same or at least
    pre-load the piston with air just prior to the intake stroke and hence act as a form
    of micro supercharging?

  7. Jul 16, 2010 #6
    As the exhaust stroke proceeds i believe that cylinder pressures drop exponentially and
    are less than 10bar at the end of the stroke. A 300 psi stream should at least be able to
    enter the chamber and my question is whether this mechanism can effect more efficient
    extraction by imposing less chaotic flows.

    For the moment i'm ignoring the issue of the energy source of the compressed air.

    One might expect that a more complete exhaust process leaving residual air rather than
    waste gases in the cyclinder could pr-load the chamber prior to the intake stroke
    effectively acting as a micro supercharger.
  8. Jul 16, 2010 #7

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Read post in mech engineering forum

    Horsepower - Please help - Confused!
    Feb14-09, 04:36 AM
  9. Jul 16, 2010 #8
    Well, to improve evacuation efficiency, you should R&D the single cylinder with a exhaust pipe and one without because downstream system characteristics effect things upstream. It should all be seen as one complete system, but you can focus on the cylinder to end of the exhaust port itself. When I work on improving flow, I observe what I can do to the valve itself first since it is the first thing the exhaust gas contacts.

    At what point is the exhaust valve at? seated still? Opened slightly prior to injection?

    This would also depend on the valve events. If the injection was early on an induction stroke, depending on your induction charge velocity & inertia level etc, it could cause a (if carbureted or fuel injection is done prior to the chamber) reversion of the air & fuel and cause a lean condition and misfire or detonate possibly. We don't have to worry about this with direct injection which is nice. If the air injection is later, after majority of the induction charge has filled the cylinder, then I would say yes it would present a more dense charge since the cylinder is almost if not closed already and ready for ignition. I suppose you would have to calculate the amount of air injected at the level you are planning to use.

    Just remember, if you have an injection that doesn't come through the intake port, there is that chance of the flow stopping and reversing itself because the pressure injection could very well create a higher pressure within the cylinder. This doesn't mean you will never get reversion with forced induction either.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook