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

New Piston Head Concept

  1. Feb 21, 2005 #1

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    I want to share and discuss with every forum posters here a new concept on piston heads, which by the way is not as brainy as you may think.

    The new concept is concerned with spark ignition engines. Everybody know that combustion process have two main stages in this kind of engines.

    Once the mixture has entered into the chamber with moderate turbulence, the spark ignites it and enhances a premixed flame. The very first instant of ignition is governed by laminar considerations, and it is the so called "laminar" stage. The process of ignition depends upon local thermodynamic variables and has low dependence with turbulence as it was checked by Taylor et al. There is a finite ignition energy threshold which has to be reached by the spark energy inflow in order to burn the mixture. The crank angle occupied by this stage will be [tex] \theta_l[/tex].

    Once the ignition has occurred there is a turbulent transportation of the flame to the rest of the chamber. Because of this transportation there are effects of "cyclic dispersions", as the cycles represented in a PV diagram vary continously due to successes and failures of the flame. Each eddie transports the flame to parts where there are reactants. So that, this stage is strongly influenciated by turbulence, and is the so-called "turbulent" stage.
    We will call the crank angle occupied by this stage [tex] \theta_t[/tex].

    There are many ways of increasing turbulence in order to exploit the combustion process and make it to last the least. Pollutant generation and power bonus would be some of these profits. New concepts, like the GDI Mitshubishi engine makes use of the "tumble" movement of the mixture, designing a piston head which points directly the intake mixture towards the plug.

    The turbulence is modified by the angular velocity [tex] n [/tex]. In fact, the more [tex] n [/tex] the more [tex] \theta_t[/tex] but the time of combustion remains the same, so nothing exceptional is achieved.

    You may now open the figure attached. I only have drawn a quarter of the piston surface due to the axilsymmetry.

    My concept is concerned with the next:

    i) the entire piston has two movements: one of spinning around its axis of symmetry and the typical up and down movement.

    ii) the head piston is shaped and mechanized in such a form that there are small blades disposed radially. So that, as the piston spins, the flow is forced to swirl, and so increasing strongly the turbulence and mixing.

    iii) Therefore, this design has two main consecuences: (a) it increases the global volumetric efficiency because of the suction effect of the blades. Remain that the volumetric efficiency of a heat engine is defined as [tex] \eta_v=m_a / (\rho_{at}Q)[/tex] the quotient between the air mass accepted by the combustion chamber and the reference mass adopted (here it is the multiplication of the atmospheric density and the total volume of the chamber). Such increasing of volumetric efficiency will enhance an increasing of total power, ceteris paribus. (b) It decreases the time of the turbulent stage, because [tex] \theta_t=nt[/tex] and [tex] n [/tex] is constant. This decreasing enhances a less combustion duration, so that the gas expansion stage will take away less power to the chemical reaction. Also, the pollutant generation due to further flame cooling by means of the expansion will not occur, because the flame will end well before the point it nowadays does. More power will be available.


    What are your opinions? Negative comments are allowed, only and only if they are not very offensive :tongue2: .
     

    Attached Files:

    Last edited: Feb 21, 2005
  2. jcsd
  3. Feb 21, 2005 #2

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    I seem to remember that two-stroke engines often incorporate a swirl chamber into the top of the cylinder head to promote more complete mixing of the fuel, along with a 'bump' on the top of the piston head to help direct incoming flow, but obviously this is partly to do with trying to minimise the mixing of fresh fuel/air and exhaust gases.

    There's also a hell of a lot of work being done at the moment on improving atomisation effects of fuel injectors, particularly with regard to being able to control the break-up point of the fuel film, but I haven't seen anything like this!

    I think the first thing which I would ask is just how good Mitubishi's GDI concept really is? It's been around for a good while now, I'm surprised that we're not seeing the same concept used by other manufacturers yet.

    I would probably give a lot of attention to the issue about the use of thin, fine 'swirl vanes' in such a harsh environment in your proposal, these would be experiencing the very hottest temperatures seen in an internal combustion engine (and much hotter than that experienced within, say, a turbocharger).

    Other than that, my worry would be concerned with the mechanics of getting your piston to rotate axially. When under the 'work' stroke, the piston and con rod are under a phenomenal amount of compression, I would imagine any bearing capable of transmitting this load axially would be pretty hefty.

    Anyway, those are the two concerns which spring to mind, haven't done any numbers though! A nice thought, is this some project you're working on or just an idea?
     
  4. Feb 21, 2005 #3

    Astronuc

    User Avatar

    Staff: Mentor

    Interesting idea. However there would be some design/performance issues.

    1) A rotating piston head means additional moving parts. Then one must consider how the piston is connected to the connecting rod to allow rotation.

    2) On the combustion side of the piston head, I would be comcerned about deposits in the bottom of the 'serrations' (grooves?, i.e. root of small blades). It might be more desirable to offset the intake entry, and perhaps opposing entry intakes (i.e. dual valve intake), but that may introduce issues with cam design/performance.

    3) The grooves could be a location of stress concentration and fatigue life might be an issue.

    You might want to look at Honda's CVCC and VTEC technologies.
     
  5. Feb 21, 2005 #4

    minger

    User Avatar
    Science Advisor

    Yes, the first thing I thought about when ready this was how the piston would be connected to the connecting rod. Typically, a simple wrist pin goes through the piston and connecting rod. With the piston spinning about the connecting rod, there would need to be some bearing. It would take a helleva bearing, taking into account the tremendous force that is being exterted on the piston during combustion.

    Also, explain how the spinning piston would increase volumetric efficiency. I guess I'm just not understanding this "suction" of the blades. If they are spinning, it would seem it would in essense create 'high' pressure. Something spinning with vanse sounds like a pump/compressor to me. The point of a pump is to create pressure. Being that there is a significant pressure drop across the air flow sensor (mass air, volume air, for the cars/trucks that have those), elbows, throttle body, and that pressure hog air cleaner, you want as low pressure in your combustion chamber as possible after exhaust.

    The last thing I was thinking about was in to Astronuc's point on deposits. Every car in the world burns oil, that's just part of the necessary engine design, seems to me like it would just take time for these vanes to become filled in, further reducing efficiency.

    And, also, I'm not sure how much VTEC technology applies here. The engineering is wonderful (I used to work for Honda) and as simply as it is, its brilliant. Basically, it was originally designed to allow a car to essentially have two sets of cam profiles. The simple method of locking and unlock rocker arms is what makes the system brilliant. After time, they came out with the "efficiency" VTEC system. This is the system that is put on Honda's SOHC engines. What it does is under a certain RPM (oh jeez, I used to know this....~5000 rpm) it swiches from it's normal cam profile to basically a car with 4 valves per cylinder. Under that RPM, the cam profiles for the individual valves are different. This makes one valve open more than the other, creating a swirling effect during intake, and I believe this is what you referring to.

    Either way, if you happen to come across a book called, "Performance Tuning for the 4-stroke Engine", be sure to read it, especially if you are into cars and performance. It goes beyond what you will ever get out of any magazine, citing formulas and real engineering principles (not just, dude, im gonna crank my booost up and hit the nos). In that book though, there is an entire chapter devoted to combustion chamber design that I believe covers a little on piston design. It is very good reading.
     
    Last edited: Feb 21, 2005
  6. Feb 21, 2005 #5

    GENIERE

    User Avatar
    Science Advisor

    The piston would not be able to rotate after a short running time as the cylinder erodes to an oval shape. This occurs due to the pivoting action of the connecting rod, side thrust if you’re looking down the crankshaft axis. A wobble plate design to eliminate the crankshaft would have to be used. Sharp edges in the combustion chamber may not be able to dissipate heat quickly enough and ignite the next fuel charge prematurely.

    A rotor recessed into the piston may do what you want or perhaps a rocking device.
     
  7. Feb 22, 2005 #6

    FredGarvin

    User Avatar
    Science Advisor

    Geniere,

    The rotating part could be undersized on the diameter enough to account for the out of roundness, assuming that the lower piston component has the appropriate sealing capability (rings). That being said, you do have a very good point regarding the possibility of predetonation. That may have to be overcome by running rich, which in turn opens up a bunch of other problems.
     
  8. Feb 22, 2005 #7
    When examining the situation, the idea sounds brilliant. It integrates a turbine property into a combustion engine.

    There could be a good side and a bad side to the idea. A good idea would be to place a pump in the piston which circulates coolant through the piston walls, cooling the piston head and piston sides. I know this sounds impossible since the pressure in the cylinder is so high in the combustion stage. But with a little bit of engineering, and some heavy duty parts, it is doable.

    The major setbacks of the idea are the frictions caused in the cylinder. With the addition of a spinning cylinder head, the tolerances of the construction still have to be within 0.0001in. This spinning head with this kind of tolerance would create tremendous pressure and heat from the friction of it spinning against the cylinder walls. This friction in turn would eventually wear down the sides and the piston would start burning oil, and the engine would loose efficiency. Then you would have to think about the cooling systems you have cooling the extra heated cylinder.

    Overall, the idea sounds like a good one, but the engineering of the idea needs to be worked on.

    Regards,

    Nenad.
     
  9. Feb 22, 2005 #8

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    This would be much harder than you think. Circulating coolant through the piston walls? What, and out into the combustion chamber? Or is this not what you meant? Getting coolant feed into anything which moves like a piston moves would be an incredibly hard challenge, and don't forget that pistons are made solid for a reason.
     
  10. Feb 22, 2005 #9

    FredGarvin

    User Avatar
    Science Advisor

    Not only would it be incredibly difficult to get coolant into the piston walls, then you would be effecting the combustion chamber temperature which I think would negate any benefits from the improved turbulence.
     
  11. Feb 22, 2005 #10

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    Quoting the Mitshubishi GDI I was trying to underline the efforts made in increasing turbulence and mixing. The key with this engine is the shape of the intake manifold (which goes vertically straight into the chamber) and the shape of the piston head (which promotes the "tumble" or vertical swirl movement).

    Yes, here I totally agree with you. But, I mean, the small blades are like the blades of a saw, they would be mechanized over the piston in the manufacturing process. No exceptional material will be needed. The proper piston head material is the recquired one. Everybody know that compressor blades in a turbocharger have an accurate function and a low tolerance of design because they are designed according to fluid motion considerations.

    Yes!! You hit the nail on the head by saying that!. That's the main difficulty. And a mechanics challenge, isn't it? I have been thinking it for a while and I haven't come to a conclusion of how the movement can be communicated to the piston.

    It was only and idea.
     
  12. Feb 22, 2005 #11

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    Yes it is one of the main disadvantages and design challenge, as Brewnog has pointed out too. I haven't realised how the hell the spinning movement could be extracted from the crankshaft.

    Maybe you're right. You talk like an engineer :rofl:, didn't you realised?

    [/QUOTE]
     
  13. Feb 22, 2005 #12

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    As far as getting round the issue of turning the piston go, why not just have a small spinning rotor on top of the piston? Make it out of Nimonic 90 or something, and such that the force of expanding gas still presses on the top of the actual piston (which would not itself rotate) rather than the 'swirl disc', and it could be sufficiently small in diameter so as not to go anywhere near the cylinder liners. It might not need powering; with some clever fluid mechanics (sponsored by Clausius, of course) it could be set spinning just by the fluid flow in and out of the combustion chamber.

    At TDC, what kind of space is there between the piston head and the cylinder head, for a diesel engine, and for a spark ignition engine? This is where I see the problems arising, - compression ratios are such that frequently there isn't even enough room to allow for a broken cam belt without the valves getting twatted by the piston, let alone enough room for a lightweight spinny whirly thing.
     
  14. Feb 22, 2005 #13

    Clausius2

    User Avatar
    Science Advisor
    Gold Member


    Imagine there are no blades in the piston head, and assume that the piston is spinning around it's axis of symmetry. Due to viscosity, the movement of piston head will produce shearing stresses into the fluid, communicating progressively the drag over the rest of the intake mixture. If the process is well designed, there won't be enough time available to raise pressure due to an excess of mixture an accumulation. The vanes collaborate in this movement trying to enlarge the shear stresses inside the fluid. It is not a question of pressure decreasing (although the high turbulence into the chamber would decrease the static pressure), but of viscoous drag of the intake mixture towards the chamber.
     
  15. Feb 22, 2005 #14

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    The combustion chamber MUST NOT dissipate any heat. That would be a loose of energy and efficiency.
    Edit: Sure there is a cooling system, but its mission is to conserve the materials.
     
  16. Feb 22, 2005 #15

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    That's a severe problem. The blades could make the mixture to knock.
     
  17. Feb 22, 2005 #16

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    Yes, friction would be another severe problem. But may be a bearing expert could design a friction annulus which works too with spinning movement.
     
  18. Feb 22, 2005 #17

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    That idea sounds good, although the spinning rotor had to be energized with electricity or hydraulics, and the connections to the source could be another problem. The sponsoring also sounds good. :smile:

    Diesel engines use to employ compression ratios of 14-20, while spark ignition engines use to employ compression ratios of 7-12 or so. We know that the more compression ratio the more efficiency and the more power. But in spark ignition engines, the more compression ratio the more average temperature after compression, and so the more tendence to undesired knocking (detonation). The design I showed is very rough, maybe it could be improved in some way the intake mixture is directed also towards the plug to enhance better the ignition. The effect of turbulence and blades is benefitious once the combustion has started, but before the ignition it could be dangerous due to knocking.

    The main point with this design is the use of a spinning piston. It is one of those things which surprises people, because we are usually so inmersed in nowadays invented technology that we are unable to "look at different possible movements" inside such a traditional machine.

    Be sure if it has not been improved by engineers up to date, there is something wrong with this design.
     
  19. Feb 22, 2005 #18

    minger

    User Avatar
    Science Advisor

    You agree that there isn't enough time for the negative effects to happen, then how is there enough time for the positive effects to happen? Given the time for 1 single stroke of the piston, how fast must the piston be spinning for any kind of swirling effect to take place?

    Perhaps maybe the spinning of the fuel mixture can be accomplished with simpler techinques. Example is the SOHC VTEC system I mentioned earlier. With 2 sources of fuel mixture entering the combustion chamber at very different velocities and flow rates, this automatically creates a great swirl, increases fuel atomization, and improves efficiency tremendously. In addition to this, possible the idea "rifling" the intake runners right before the valve, perhaps even a swirling design on the valve itself. There is also the idea is twisting ellipital shaped intake runners.
     
  20. Feb 22, 2005 #19

    FredGarvin

    User Avatar
    Science Advisor

    I was thinking along those lines but for the piston itself. The rifling could pose a sealing problem. However, with, say a ball and socket as the connecting rod attachment, it theoretically could rotate the cylinder in the bore.
     
  21. Feb 22, 2005 #20

    brewnog

    User Avatar
    Science Advisor
    Gold Member

    With all that said, I still think that near-future advances in improving the efficiency of internal combustion engines are those related to improving atomisation of the fuel into the cylinder. I'm currently working on a project looking at how sound waves can precisely control the breakup point of the liquid film, and thus perhaps accurately control the combustion rate. No complicated mechanisms this way...!
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: New Piston Head Concept
  1. Rotary Piston (Replies: 22)

  2. Horizontal Pistons (Replies: 4)

  3. Piston seals (Replies: 22)

Loading...