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

Power of an Engine

  1. Dec 5, 2013 #1
    I always wonder how BMW engines of smaller size are powerful than other OEM's engines despite other OEMs haveing engine of larger size.
    To be very clear i have following questions..
    1. Doesnt size of engine affects the power of the engine?
    2. What influences the power of engine...? Bore? Stroke? CC? Material?
    3. Does the power of a car is mainly predicted upon the power of engine or its Drag?
     
  2. jcsd
  3. Dec 5, 2013 #2

    cjl

    User Avatar

    They aren't.

    BMW engines are very much middle of the road when it comes to power output. That isn't to say they're bad - far from it, they make some very nice engines. They aren't dramatically different from any other manufacturer's efforts though, at least when it comes to specific power output (power output for a given engine size).

    As for your questions?

    1) Of course. A larger displacement engine will, all else being equal, make more power than a smaller one
    2) All of the above, as well as compression ratio, engine layout, forced induction vs natural aspiration, and a variety of other factors.
    3) At reasonable speeds (let's say, 80mph and less for example), drag doesn't make a lot of difference in performance, at least for higher performance cars like most BMWs. It makes a large difference in fuel consumption, but the performance is based on other factors. Power is important, of course, as is weight (light weight is better), and weight distribution. Tire compound makes a large difference too, as does suspension geometry, as does drivetrain layout (front vs rear vs all wheel drive). It's difficult to narrow it down to just one factor - many factors play into a car's overall performance.
     
  4. Dec 6, 2013 #3

    rcgldr

    User Avatar
    Homework Helper

    Looking at BMW specs: 3 series - 1.6 liter engine, 100 kw / 136 hp, 2.0 liter engine, 135 kw / 184 hp, or 328i, 2.0? liter engine, 180 kw / 245 hp, 3.0 liter, 225 kw / 306 hp. 5 series, 4.4 liter engnie, 330 kw / 449 hp.

    The numbers are a bit higher than those for USA built cars like the Mustang and Camaro, but the difference seems to be a deliberate choice made by the car makers as opposed to some techincal issue. Some of the Mustang engines use dual cam 4 valve per cylinder design, but the power output isn't much more than Chevy's push rod 2 valve per cylinder designs. Chrysler gets 345kw / 470 hp from a 6.4 liter engine.

    On the other hand motorcycle engines produce a lot of power, getting 142 kw /190 hp from 1.35 liter engines.

    Since the smaller engines have the advantage of higher rpm, comparing power per liter could be considered a bit "unfair". It would be a bit more "fair" to compare peak torque per liter. Using this as a basis, motorcycles and high end sports cars like Ferrari get 80 to 85 ft lb torque per liter. Other sports oriented cars get around 75 ft lbs of torque per liter, while the early 4 valve per cylinder Mustangs were only getting around 65 ft lbs of torque per liter, resulting in less power than should be expected from such engines.

    Another comparson is the power to weight ratio of the engine. Chevy got 505 hp from a 7.0 liter pushrod V8 engine used in the Corvette Z06, while Porsche got 480 hp to 520 hp from a turbo charged flat 6 3.6 liter engine, but the Porsche engine weighs more than Chevy's 7.0 liter V8.
     
  5. Dec 6, 2013 #4
    Thanks for taking time to replying for my silly childish questions. I have a doubt
    "How cc plays the role in determining engine performance or more precisely power?"
     
  6. Dec 6, 2013 #5

    cjl

    User Avatar

    It all depends on what you compare them to though. Several of the BMW engines listed above use forced induction (turbocharging), so they can't be compared directly to naturally aspirated engines. Among naturally aspirated engines though, most high-performance engines fall in similar power ranges (typically around 85-110hp/L). For example:

    BMW inline 6 from the older M3: 3.2L, 338 hp
    BMW V8 from the new M3: 4.0L, 414 hp
    Subaru FA20 from the BRZ: 2.0L, 200hp
    Honda F20C from the S2000: 2.0L, 240hp
    Porsche H6 from the Cayman: 2.7L, 275hp
    Ford Mustang Boss 302: 5.0L, 444hp
    Chevy Camaro V6: 3.6L, 323hp
    Porsche 911 GT3: 3.8L, 475hp (wow!)


    This is true, to a great extent - this is why a lot of the highest power naturally aspirated engines have a lot of cylinders (typically a V12). The small cylinder dimensions allow for the engine to rev very high, but the large number of cylinders still allows for a large overall displacement (for example the 618hp, 6.1L BMW S70/2 V12 engine used in the McLaren F1).

    As for the torque? New Mustangs actually have among the highest specific torque available, aside from super-exotics. The Coyote 5.0L engine in the Mustang GT puts out 390 lb-ft of torque, which is a specific torque of 78 ft-lb/L. The V6 is similar, putting down 280lb-ft from a 3.7L, or 75.8 ft-lb/L.

    This is also important to note - the Z06 engine is large displacement, but its actual physical dimensions are small, due to the pushrod design, and it doesn't weigh very much due to the aluminum block and naturally aspirated design. As a result, many "smaller" engines are actually larger and heavier. It is often overlooked though among people looking to compare car and engine specifications.

    (I'm curious what numbers you're using for that comparison though - from what I can find, the 3.8 flat 6 in the Turbo is of a pretty similar weight to the LS1, rather than being significantly heavier).
     
  7. Dec 6, 2013 #6

    etudiant

    User Avatar
    Gold Member

    Is there some SAE standard for measuring engine weight?
    It would be helpful to have an idea of the engine all up weight, including cooling fluid, radiators, turbos etc to achieve a reasonably even comparison, but I'm not sure anyone provides that.
     
  8. Dec 6, 2013 #7

    rcgldr

    User Avatar
    Homework Helper

    Exotics and sport motorcyle engines are in the 80 to 85 ft lb / L range. Due to the high rpm, Suzuki's 1 liter GSXR 1000 makes 195 hp at 12,000 rpm at the crank.

    The comparasons were based on the turbo charged 3.6 liter engines in in 911 GT2's, which I assume includes the weight of the turbo chargers. The earlier GT2's were making 480 hp, while the newer ones are making 520 hp. I don't recall the actual numbers, only that the Z06 engine was a bit lighter than the GT2 engine (with turbo chargers). Both cars overall curb weight is about 3160 lbs (the Z06 with some fuel, the GT2 with no fuel), which is less than the 3360 lbs of a Subaru WRX STI.
     
  9. Dec 6, 2013 #8

    cjl

    User Avatar

    Yep. I should have included motorcycles in my statement there, but it doesn't change the fact that the Mustangs are making rather surprising amounts of torque for their engine size and price range. Their specific power is less impressive, mostly because they aren't terribly high revving, but they're certainly excellent engines. As for impressive motorcycle engines, I think my favorite is the 1L 4 cylinder in the BMW S1000RR and HP4. 193hp from 1 liter at 13,000 rpm, with 13:1 compression.

    Ahh - the numbers I was looking at were for the regular Turbo - I think the GT2 has larger turbos, which explains the difference. I'm also not completely sure what was included in the engine weight I saw. It definitely illustrates the point though - engine displacement alone isn't a great way to judge an engine's size.
     
  10. Dec 9, 2013 #9
    OP, you might find this article interesting, the effects of bore and stroke on torque, rpm and power.

    How tuned is your engine?

    Basically, bore diameter and number of cylinders are the biggest contributors to power. Stoke and capacity are less important because as you increase/decrease the stroke, the capacity and thus torque change proportionally but so too does the maximum RPM. If you increase the stroke, you get more torque but at a lower speed so the power stays constant.

    The PI rating system in the article is comparable over any NASP four stroke piston engine. You'll see that BMWs S52 M3 engines are some of the best available in a production car, with Honda's VTEC engines close behind. The best is the Ferrari 458, but that still leaves a lot on the table compared to an F1 engine...
     
  11. Dec 9, 2013 #10
    I 100% agree with this, power per litre IS unfair. The PI score in the article I linked above cancels out the capacity (read: stroke and thus RPM) aspect and instead does it based on bore diameter and number of cylinders, which is a far more universal measure. You can compare a Mustang with a motorbike with an F1 car using that score.
     
  12. Dec 9, 2013 #11

    cjl

    User Avatar

    That's a fascinating article. Another engine which scores extremely well on that metric is the 911 GT3's flat 6 - with a 102mm bore and 475hp, it handily beats the BMW engines (but still falls short of the Ferrari).
     
  13. Dec 12, 2013 #12
    Ferrari managed to take the best (?) out of their engines for the 458 Italia by extremely fine surface finishing inside the cylinders. Surface roughness means a lot in such areas.
    Moreover, generally speaking, the way in which the fuel is mixed with the air affects the combustion process which in turn have its impact on the performance so a lot of optimisation is being done on this.
     
  14. Dec 12, 2013 #13

    jack action

    User Avatar
    Science Advisor
    Gold Member

    X2

    For more info, you can also check HPWizard.com. There is also a horsepower calculator.

    1. As said previously, absolutely not. The stroke acts as a lever only (or like the radius of a wheel). So the longer the stroke, the higher the torque but the lower the maximum rpm. The power always stays constant.

    2. Basically 2 things influences power:
    3. The power needed is predicted by the drag, rolling resistance and acceleration. The engine power is the power available. More info with this acceleration simulator and the theory behind it.
     
  15. Dec 13, 2013 #14
    Power will give you a better value to compare how quickly each car accelerates. If both cars have 200 Nm of torque, one could create it at 4000 rpm and another at 2000 rpm and they'd be vastly different. However if both had 200 bhp, one at 4000 rpm and one at 8000 rpm, they'd be very close in performance (assuming same weight etc).

    For examples check out diesel and petrol versions of the same car. eg BMW 120i and 120d (not the latest ones, haven't checked those figures), both around 180 bhp and almost identical 0-60 times. However the diesel generates way more torque. Now compare the diesel with the 130i which has similar peak torque, and the 130i is waaaaay faster.
     
  16. Dec 13, 2013 #15
    Power and power reporting is dependant upon the perceived needs of the manufacturer.

    If ford needs to get 25+mpg on the highway out of the mustang to bring the corporate fleet under a certain number, then they will do what is necessary since you can only sell so many escorts to offset the other cars in the fleet.

    Longeivity is also a concern and companies have been known to ignore it to gain quick fame for high numbers by possibly compromising engine life.

    I'm pretty sure honda gambled with the F20c that owners wouldn't run it up to the redline every single shift as a daily driver and it paid off.....kinda.......unless you were a honda technician replacing them under warranty.

    Notice that after honda made it's HP bragging rights that the later F22c was a lower rpm engine........partially because of higher displacement but also for engine life.

    Toyota did the same thing with the 2zz in the celica GTS where it was introduced at 8250 (maybe higher) rpms for the first two years and lowered the redline to 7800 rpms.

    Porsche did it in 1967 with the 911r which made 210hp from 2.0 liters.

    Oldsmobile did it for a limited production run of W41 quad 4 engines that won what's now called the pirelli world challenge 3 years in a row, made thier mark, and lowered the redlines and hp of the engine.

    BMW did it with the E30 M3 in which the evoIII made around 230hp.

    MBZ did it with the 190e-16v evoII made around 224hp.

    Manufacturers muddy up the waters and have been known to take a gamble on low production cars because the lower numbers of them out there means less will come back for warranty work and they get thier bragging rights.

    They make thier reputation on relatively low production drivetrains and they make thier money on the rest of the product line.......except for the exotic car manufacturers but you don't see guys driving around thier ferrari 288gto to work every day.


    You can also look at the junkyards for an indication of relative engine life and you will surprised which cars end up there at a pretty young age as compared to others that are still running at 250,000+ miles.
     
  17. Dec 16, 2013 #16

    cjl

    User Avatar

    This isn't really true, from everything I've heard about the S2k (and from what I know about the engines). A lot of people race the S2k with the F20C, and it's a very reliable engine even under racetrack conditions (redline all the time, high power output) if well maintained. From everything I heard, the move to the 2.2L design was more to increase low end torque, rather than to improve reliability, since the F20c wasn't a great engine for just putting around town unless you revved it up quite a bit. This is supported by the fact that they stroked the F20c to make the F22c, rather than boring it out. Stroking increases the piston speed at a given rpm, and increases the stresses on the components at a given rpm, so the lowered peak rpm wasn't to lower stress, it was to bring them back down to the same level they had been on the F20c, since stroking the engine increased the stresses.

    This is verified by looking at piston velocity - if you look at the maximum piston velocity, it's nearly identical between the 20c and the 22c, with the lower RPM almost perfectly counteracted by the increased stroke.
     
    Last edited: Dec 16, 2013
  18. Dec 16, 2013 #17
    Good post this. As it highlights the ultimate practical goal of an engine, which is to be drivable.
    The F20c made it's peak torque after most engines had run out of revs.

    Which is juxtaposed by this overly simplistic view of driving life.
    The assumption is 10/10ths driving. When giving it 'the beans' the same car with more power will indeed be quicker (obvious??). However the x20d is the faster car in the real world (definately compared to the x20i and possibly in some cases the x30i), and acutally slightly more relaxing to drive becuase of the low down in gear grunt.

    It's the folly of using 'peak' figures rather than looking at the engines performance as a whole.


    On saying that, I'd rather have the petrol, less hassle and just 'nicer'. And the BMW 6pots are lovely engines. Got an N52 in my 3 series, though you do have to really give it some to get the best from it. 5000rpm+



    BMWs raison d'etre has never been to make the most powerful engine, the fastest 0-60, or any other crap like this. Their goal was to make 'The Ultimate Driving Machine'. Which meant the most balanced car they could, and the amount of cars you see compromised to give a certain stat is just silly.
     
  19. Dec 16, 2013 #18
    I guess you weren't replacing them at the dealership under warranty and getting paid 1/3 of what you earn if it were customer pay.

    I was there in the early 90s fixing quad 4s, and was there at Honda replacing f20c, moved to Toyota and experienced replacing Toyota 2zz engines so I've actually experienced it.

    In order to keep master technician self from being outearned by first year lubetechs handed gravy such as services and brake jobs while I was replacing the above mentioned engines, I had to spend time doing detective work like lurking on forums where the customers bragged they pulled one over on the dealerhip.
     
    Last edited: Dec 16, 2013
  20. Dec 17, 2013 #19
    I know from experience that it's very easy to become jaded and over critical when all you see is broken ones day in, day out.

    No matter where you set your durability limit, someone is always going come along, exceed it and make the engine go pop. It's just trying to get the balance right so that replacement costs (including reputation damage etc) are less than the savings made on unit cost.

    It's the type of people that rev engines hard from cold, use chip fat for lubrication, then complain when it's totally shagged after a couple of years. Nature of the beast.

    Compare and contrast to say, the Renesis engine in the RX-8. I don't think I've ever seen one that hadn't had an engine swap. So much so that UK Mazda dealers very rarely take one as part ex.


    PS. If stuff didn't break, there'd be no need to employ people to fix it at all.
     
    Last edited: Dec 17, 2013
  21. Dec 17, 2013 #20
    My point is that I hardly ever see a broken engine at my lexus dealership job.

    In fact, I only saw 2 in two years and one was because somebody ran it without changing the oil for 40,000 miles.......actually never checked it, either.

    But, when I worked at Toyota, we got at least 2 celica GTS engines per month........sometimes 4 while getting maybe 1 corolla engine in a year there which again.......was there because the owner never once check the oil after 35,000 miles of driving and ran it dry.

    You can't compare rotary engines in this discussion because the configuration is completely different with very different methods of sealing whether the "water seals" or the apex/side/corner seals.

    I drove rotary cars from 1981 to 1990 exclusively in various stages of tune from a stock 12a with fat apex seals in my RX-2 to a full on J-ported 13-b in my RX-5.

    Rotary doesn't apply to "grenade engine" discussion because their difficulties don't arise from rev limits but rather the basic design.

    The point is that "grenade engines" delivered by manufacturers experience failure rates substantially higher than their close assembly line cousins and it even driving them by the book (owners manual) still results in said higher failure rate.

    Manufacturers know it and this is why most "grenade engines" or the cars in which they delivered which are intended to produce bragging rights on power production or RPMs are of limited production so the higher failure rate can be absorbed.

    In using "absorbed", I am referencing how manufacturers are very careful to use the term "campaign" instead of recall and the cooking of warranty books to dilute the higher failure models into the fleet as a whole.

    As a former service manager, I have plenty of exposure on how the manufacturer wants the warranty department to classify failed parts to lessen the percentages as perceived by the public.

    This is why some manufacturers can get away with it..........but not always.
     
    Last edited: Dec 17, 2013
  22. Dec 17, 2013 #21

    cjl

    User Avatar

    That's not entirely a fair comparison though - the Lexus owners probably take better care, on average, than the Toyota/Honda owners did, and very few of them probably rev anywhere near redline on a regular basis (and the ones who do are likely to be the ones who care about, and take good care of their cars). This is especially true with something like a Celica GT-S, since it was sold as a sporty, cheap car (which increased its popularity among younger drivers, and the low cost makes the drivers more likely to abuse it), and it didn't have a lot of power unless you revved the heck out of it. I'd be willing to bet a lot more Celica GT-S drivers would get in their car and rev it to the redline immediately (even when cold) than Lexus drivers.

    Also, I was just running the numbers on the engine in the Porsche 918 Spyder in that calculator on the prior page, and I think it beats the Ferrari 458 in engine performance. 608 hp from a 4.6L V8 with a 95mm bore. It's not even talked about all that much - most people seem to focus on the hybrid aspect of the car instead, and I haven't seen much talk about the fact that it seems to have one of the finest naturally aspirated piston engines ever made. The overall engine weight is just 300lb too, which is astonishingly light for a 600hp engine. I'd really be curious to see more about that engine, since as I said above, it seems to be a somewhat overlooked part of the car (and a real engineering marvel).
     
  23. Dec 25, 2013 #22
    Engine power output is dependent upon: displacement, volumetric efficiency, compression ratio, and rpm.
     
  24. Jan 7, 2014 #23
    BMW piston engines tend to use advanced composite alloy materials that can withstand much higher internal cylinder temperatures & pressures than many of their competitors.

    Because of this BMW can make their piston engines much smaller while compensating with higher internal cylinder pressures & temperatures upon the pistons during each powerstroke. Force on piston = Pressure on piston x Area of piston head.

    This allows smaller BMW piston engines to match the same power output of engines that are larger which operate at lower internal cylinder pressures & temperatures upon the pistons during each powerstroke.

    Parameters that affect a piston engine’s power output are: Brake Horsepower, Brake Mean Effective Internal Cylinder Pressure (BMEP) during each powerstroke and Torque. All three are directly proportional to each other.

    Other parameters that affect a piston engine’s power output are Volumetric Efficiency, Compression Ratio and whether or not the piston engine is naturally-aspirated or is turbocharged. If naturally-aspirated, higher the elevation the piston engine operates at above Sea Level the lower its power output.

    The power output lost in a piston engine moving an automobile is from: Friction Horsepower, Aerodynamic Drag, Tire Rolling Resistance and Naturally-Aspirated Elevation above Sea Level.


    Indicated Horsepower = Total theoretical power involved with the entire piston engine at a certain RPM which includes the power lost to keep the engine running plus the usable power available at the crankshaft.

    Brake Horsepower = Usable power at the crankshaft of a piston engine at a certain RPM.

    Friction Horsepower = Power lost to keep the piston engine running at a certain RPM.

    Indicated Horsepower = Brake Horsepower + Friction Horsepower.

    BMEP = Average gas pressure exerted upon piston head area during each powerstroke measured in PSI.

    Torque = Twisting moment of a rotating plane measured in Ft-Lbs also known as Work.

    Power = Ft-Lbs Work per 1 sec. 1 HP = 550 Ft-Lbs/Sec = 746 Watts = 746 Joules/Sec.

    Volumetric Efficiency = Ratio of the actual amount of airflow entering a piston engine divided into its theoretically calculated intake airflow times 100. Naturally aspirated piston engines at Sea Level without a turbocharger or supercharger operate around 80% Volumetric Efficiency and even lower when farther above Sea Level. Not all of the fuel charge burns in a naturally-aspirated piston engine because it relies solely on atmospheric pressure in order for air to enter piston engine to burn the fuel charge.

    Compression Ratio = The ratio between the total cylinder volume including the clearance volume above the piston head when it is at top dead center divided into the clearance volume above the piston when it is at top dead center.


    USEFUL FORMULAS FOR GAS PISTON ENGINES:

    Volume Conversions =

    1 Cubic Inch = 16.387 CC
    1 CC = 0.061 Cubic Inches
    1 Liter = 1,000 CC

    BMEP in PSI (4-Stoke Piston Engine) = [(150.8) x (Ft-Lbs Torque)] / [(CID)]

    CID = Total Cubic Inch Displacement of Piston Engine
    Note: If 2-Stroke Piston Engine then replace 150.8 with 75.4 instead

    Brake Horsepower (BHP) = [(BMEP) x (L) x (A) x (N) x (K)] / [(33,000)]

    BMEP = Brake Mean Effective Pressure in PSI
    L = Piston Stroke Length in Feet
    A = Piston Head Area in Square Inches = [(Piston Diameter / 2)² x (3.14159)]
    N = Number of Powerstrokes per minute ===> 4-stroke = [(RPM / 2)]
    ===> 2-stroke = RPM
    K = Number of Cylinders

    Note: Brake Horsepower is directly proportional to Torque. To find maximum Brake Horsepower use the RPM representing maximum Ft-Lbs Torque. Half of this RPM is the 4-stroke piston engine’s powerstrokes per minute (N).

    Atmospheric Pressure PSI (Between 0 Ft MSL and 36,089 Ft MSL Only) =

    [(1) – (H / 145,442)]^5.255876 x [14.7]

    H = Elevation in Feet MSL Above Sea Level

    Slugs/Ft³ Air Density = [(29 x Atmospheric Pressure PSI x 0.031) / (10.73 x °R)]

    °R = [(°F + 460)]
    °F = Temperature of outside air in degrees Fahrenheit

    BHP Lost To Aerodynamic Drag = [(Cd x A x P x V³) / (2)] / [746]

    Cd = Car’s Drag Coefficient
    A = Frontal Area of car in square feet
    P = Air Density in Slugs/Ft³
    V = Vehicle Speed in Ft/Sec
    MPH to Ft/Sec = [(MPH x 22) / (15)]

    BHP Lost to Rolling Resistance = [(W) x (V) x (0.012)] / [(550)]

    W = Weight of car in Lbs
    V = Speed of car in Ft/Sec on dry paved road

    Naturally Aspirated BHP Lost To Elevation = [(BHP) x (H / 1,000) x (0.03)]

    H = Elevation in Feet Above Sea Level

    Cubic Inch Airflow/Min into 4-Stroke Piston Engine (CIM) = [(RPM / 2) x (CID / K)]

    RPM = Revolutions per minute of piston engine
    CID = Total Cubic Inch Displacement of piston engine
    K = Number of Cylinders
    Note: For 2-stroke piston engine do not divide RPM into 2.

    % Volumetric Efficiency = [(Va / CIM)] x [100]

    Va = Actual measured Cubic Inches of Airflow into piston engine per minute
    CIM = Total calculated theoretical Cubic Inch Airflow into piston engine per minute

    Note: Naturally-Aspirated Piston Engine has 80% Volumetric Efficiency at Sea Level. A turbocharged Piston Engine has 100% Volumetric Efficiency or more depending on PSIG Turbocharger Boost and Altitude Operating Limits of Turbocharger.


    You might want to look up various automobile and piston engine makes and models online. Then plug in their specifications into all these formulas and learn how piston engine and automobile parameters affect their power output and overall performance.


    Regards,

    - MisterDynamics -

    January 07, 2014
     
  25. Jan 8, 2014 #24

    cjl

    User Avatar

    I agree with the rest of your post, but (as I mentioned earlier in the thread), I disagree with this - BMW engines are not really any more advanced than any other car manufacturers' engines designed at a similar price point for a similar purpose (Audi, Mercedes, Lexus, etc). If anything, a lot of BMW's engines (especially the turbocharged ones) have a relatively low specific power output - many of their turbo engines only put out around 100-120 hp/L, which isn't all that much for a modern, direct-injection engine with forced induction.
     
  26. Jan 8, 2014 #25
    Very true.

    Most of the specifications of late-model American cars out on the market 2011 to the present have very high compression ratios and BMEP pressures with much lower displacement and better composite alloy materials than they did in the past. Not just the German & Japanese cars.

    But we have to admit that during World War II when BMW was an aircraft manufacturer they were making smaller engines with a whole lot of BMEP. Same with Mitsubishi on the Japanese end during World War II.

    I figure that in the past resources were not as tight and capitalism dominates America compared to America's competitors so why not make bigger engines with lower BMEP and higher gas consumption for profit sake on the American end?

    But things have changed, everything is getting expensive, the global economy is global not just American any longer, everyone is getting in line to make higher power-to-weight ratio piston engines that are smaller, with better materials and have much higher BMEP.

    The EGR Valves and variable valve & cam designs are making it so that 86 octane gasoline can be fired in high compression gasoline piston engines pushing in excess of 11.3:1 without detonation or "knock".

    The statement I made regarding BMW piston engines was a very generalized statement to prove a point to the person that asked the thread question. I was meaning to say that engine size doesn't necessarily mean more power when better materials can make a smaller engine with higher BMEP and compete with bigger engines for the same power.

    cjl very true what you stated I totally agree.


    Regards,

    MisterDynamics

    January 08, 2014
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook