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

Increase RPM by decreasing stroke?

  1. Aug 19, 2013 #1
    To begin I'll just give a summary of what I'm looking to do, and then explain further for those who are looking for more information. The goal of the engine is having a forced induction motor capable of revving to higher RPMs (more power at the higher RPMs with FI) while still being somewhat dependable. Thinking inline-6 for inherent engine balance (Think BMW S52/S54), or going V8 (GM LS1 or LS3) for more displacement which would bring more balancing issues into play (as we all know there's no replacement for displacement :tongue2:). To do this in summary, shorten rod length, increase component strength, raise RPM limit, apply forced induction to correct compression. Ignore cylinder head work for now, I will get to that if this method is logical.

    Firstly, replacing all internals (crank, pistons, con rods etc) to forged 4340 steel (maybe a billet crankshaft?), with 8740 chromoly steel rod bolts that way they are able to handle a much more demanding load. Then, my idea is to decrease the stroke of the engine by reducing the length of the connecting rods.

    The reason for decreasing the connecting rod length is to reduce the amount of force put onto the rod bolts occurring during compression and exhaust strokes (along with rod flex during intake and power stroke). By reducing the length, essentially my understanding is I would be able to achieve a higher piston speed because of all the other force factors on the stress v strain curve being reduced, therefore higher RPMs could be sustained. I understand decreasing stroke would alter the compression ratio, and alter the stroke/bore (bore/stroke?) ratio but I would like to figure out if my understanding is true before tackling the rest.

    Let me know your thoughts! Or if I'm crazy (could be likely). Questions are welcomed, and appreciated. All is theory, measurements are left out.

    My knowledge of physics is limited, but I do understand a decent amount. This is all based off of some research I've done into motors and my experience working with them.
  2. jcsd
  3. Aug 20, 2013 #2
    Welcome to the forum. Do you have a master plan for this engine? The vehicle it's going into, normal usage, expected life, desired level of maintenance, budget constraints, etc.? The engine is only a part of the equation. As for the stroke...it is not affected by the rod length. The crank throws determine the stroke of the engine. You can run any length rod you want as long as it will fit in the engine. Short rods do increase piston speed, but if you're looking to de-stroke an engine for high rpm, you would change the crank and then select a rod length based on piston design and the engine characteristic you desire....which could mean using a short rod.
    You're on the right track, just keep gathering more information...
  4. Aug 20, 2013 #3
    You're on the right lines.

    Reducing stroke means you lower the mean piston speed for any given RPM. Since it is usually piston speed which governs the maximum speed any engine design can run at, it's perfectly sound reasoning that reducing the stroke will increase the RPM ceiling of an engine.

    However, do not count on this to increase power. Let me introduce some maths.

    NASP, the torque you can produce is dependent on the volume of the engine. At maximum, you can expect 90lbft/litre (and that is F1 levels of performance, 80-85lbft/litre is more realistic of road engines).

    The RPM ceiling is dictated by the stroke. Factor 25m/s as your maximum.

    For any given stroke, your RPM ceiling is

    Max RPM = 30000*(25/Stroke)

    Stroke in mm

    Next the torque

    Max Torque = 0.0004048*216*Capacity

    Capacity in cc.

    Thus we can find the maximum theoretical power for any engine design as:

    Max Power = Max RPM * Max Torque / 5252

    Bear in mind that this figure is not really achievable as you'll never hit both figures at the same time. It's merely an indicator.

    Now if we write that out fully:

    (30000 * (26/Stroke))*(0.0004048*216*((∏*Bore2*Cylinders*Stroke)/4000)/5252

    We can simplify the whole thing to


    You see we have cancelled stroke out of the equation. This is because as you increase the RPM ceiling, you also reduce the potential torque at the same time.

    Now, that's for NASP engines, and FI does certainly make things more interesting.

    If you know the boost pressure, you can work out the effective capacity as

    Effective capacity = Swept volume * (1 + Boost pressure in Bar)

    So a 2000cc engine boosted to 1 bar would be effectively 3000cc.
  5. Aug 20, 2013 #4
    Currently everything is all theory and there really isn't a master plan yet. I'm still kind of figuring things out in terms of a total engine build. I might be getting a GM 350 block which if I did anything it would be based off that. I have a lot more to think about before I would blue print anything.
  6. Aug 21, 2013 #5

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    welcome to the forum..couple of notes ..the small block chevy is a great a high rpm engine..
    it really like a longer connecting rod. The best set up is the 5.7 con rod which does a few things..the ligther piston ( and you will need custom pistons) can use a smaller wrist pin, has a light lighter mas and you can go with super thin piston rings. The ratio of stroke to rod length gets better as well as the rod angularity. The piston spends more time at TDC , more dwell time and has more effective power making capability do to the angularity.
    Look into the benefits of H rod as opposed to I rod construction. There is a difference in longevity but been so long I forgot the specifics.
    The engine has to live so oiling is important and a rework of oil system is absolutely necessary at the high rpm. The old version block only oiled 4 main bearings with filtered oil. The rear main is not oiled by filtered oil. The valve train is the biggest weakness of the SMC and valve spring life is the weakest link. There are so many intake, cylinder head packages developed that you have a lot of selection....turbo, supercharged the like. On any high rpm application you must have 4 bolt mains and a girdle would help.
    Anyway, welcome again and do your homework as this is the most important step to a good build that will live.
  7. Aug 21, 2013 #6
    Hey Aquaticbob, I have been learning about how the length of the connecting rod affects the mass inertia forces and hence the piston speed from TDC to mid-stroke. Its mind-boggling how much effect just the geometry of the 4 bar mechanism has on piston speed.(most of these guys in the forum have taught me half of it :approve: ).

    You see,since the connecting rod is at an angle for most of the stroke,the piston has higher velocities from TDC to mid-stroke than from mid-stroke to BDC. The amount of variation of these velocities determines the inertia mass forces and hence the balance masses you need to weld opposite your crank throws.

    Website for simulating this kind of stuff on MATLAB:http://www.mathworks.in/products/symbolic/examples.html?file=/products/demos/symbolictlbx/Piston_modeling/Piston.html [Broken]

    Books for reference: Heinz Heisler,Advanced Engine Technology
    Last edited by a moderator: May 6, 2017
  8. Aug 21, 2013 #7
    Just to add to this, a more realistic power ceiling can be given by using 95% of the maximum RPM and torque values, which simplifies to

    Max Power = 9.2x10-3 x Bore2 * Cylinders.

    This seems almost universally applicable to NASP engines running on gasoline.
  9. Aug 21, 2013 #8

    jack action

    User Avatar
    Science Advisor
    Gold Member

  10. Aug 21, 2013 #9
    Awesome information from all! I love engines and like knowing all the technical details. I think with a good amount more understanding I could build up a nice engine. I need to know a lot more about oil supply and flow though. Same goes to coolant. There is a lot more for me to know! Is there a solid site that has a lot of information to read, or is asking questions here a good solution?
  11. Oct 11, 2013 #10
  12. Oct 13, 2013 #11
  13. Oct 13, 2013 #12
    You are correct!
  14. Oct 13, 2013 #13


    User Avatar
    Homework Helper

    The 7.0 liter (427 cubic inch) "small block" V8 used in the 6th gen Z06 Corvette's has a redline at 7,000 rpm (rev limiter kicks in at 7100 rpm), and is tuned to make 505+ hp at the crank before being installed into a Z06.

    First gen Honda S 2000's had a 2.0 liter engine that revved to 9,000 rpm, but the second gen's were changed to a 2.157 liter engine that revs to 8,000 rpm, with peak torque increased to maintain the same power, around 240 hp.
  15. Oct 15, 2013 #14


    User Avatar

    Sure, but 7000rpm is awfully high for a 7L V8 though. It's a lot easier to make a 2L inline 4 rev than a 7L V8, simply due to the size of all the components.
  16. Oct 15, 2013 #15

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    exactly where did you come up with that idea? Have you ever driven a race car running 7000 RPM with a 4 cyl engine? Easier? How so? expense wise or building it to last? please qualify your answer?

    i was going to edit my comments but let us se where this line of thinking goes...
    Last edited: Oct 15, 2013
  17. Oct 16, 2013 #16


    User Avatar
    Gold Member

    @OP, please look up Offenhauser and explore what a well-designed 4 cylinder engine can do. There is no need to reinvent the wheel when there are decades of racing experience to rely on. Good luck finding one to rebuild, but they were once regarded as "bulletproof".
  18. Oct 16, 2013 #17
    What's the stroke length on that 7.0 LS engine?
  19. Oct 16, 2013 #18


    User Avatar

    Race car? There are a lot of ~2l 4cyl street cars that run 7krpm reliably. It's nothing terribly special.

    (My street car even runs 7200rpm, and it's a 3.4L 6cyl)

    As a general rule, smaller displacement (especially smaller displacement per cylinder) engines can be made to rev higher more easily. Similarly, shorter stroke engines rev higher (all else equal), and long rods help as well (a high rod to stroke ratio allows for higher revs due to lower side loading of the pistons). All of this should be common knowledge to anyone with at least a passing interest in high performance engines.
    Last edited: Oct 16, 2013
  20. Oct 16, 2013 #19

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Sure, but 7000 rpm is awfully high for a 7L V8 though. It's a lot easier to make a 2L inline 4 rev than a 7L V8, simply due to the size of all the components.

    Really??? 7 liters = 427 cubic inches. Chevy big blocks were running 7000 right off the show room floor in 1965 when the 427 was introduced, Chrysler 413, 426 in 1962 were busting 7000 daily as did 440 cid wedge and Hemi engines could do it in Stock mode. Ford SOHC 427, DOHC 427, tunnel port nascar 427, low medium and high rise 427 Ford engines and the 424 Ford hemi and 428 cobrajet engines ran over 7000 rpm. Stock..out of the box. The problem is the cam and valve train would float over 7000 rpm except for the hemi’s . I built and raced all of these except for the Ford hemis and DOHC indy engine.

    A 4 cylinder race engine is a maintenance nightmare. The darn thing vibrates your teeth when revved over 7 grand. The reason is simple. it is firing every 90 degrees..I think it was invented to vibrate! A 6 cylinder fires every 60 degrees and is a little better but the hot set up for years is the venerable V8. The V8 fires every 45 degrees and com,parted to the 4 banger, it is a lot smoother running. Granted a short stroke 4 cylinder can be made to run 7000 rpm a lot cheaper but required a lot more maintenance and expense and usually is limited on compression ratio due to only 5 bolts holding the head gasket. The V8 usually has 5 bolts surrounding the head gasket ( Mopar being the exception). The small block Chevy was cranking out over 7000 RPM since introduced in 1955. This version was going 9000 rpm in 1959. Valve train and springs being the major liability of the design. We raced all of these at one time or another and if I recall the 2300 cc Ford 4 cylinder was going 8500 rpm on circle track min stock. I will take the V8 from horse power and reliability stand point any day.
    Last edited: Oct 16, 2013
  21. Oct 16, 2013 #20


    User Avatar

    And that still doesn't change the fact that 7000rpm is fairly high for a 7 liter engine. For a race engine, it's nothing extraordinary, but for a street engine, very few go that high. There are very different compromises that go into a street engine vs a race engine from a reliability and performance point of view.

    4 cylinder race engines run much faster than 7k all the time. Very few car classes run 4cyl engines, but many racing motorcycles run inline 4cyl engines of around 1L displacement, and they run upwards of 12000rpm easily (with some going a lot higher than that). Many street-going 4 cylinder engines run well over 7k as well - the FA20 in the BRZ/FR-S runs 7400, pretty much every honda/acura vtec engine ever made runs at least that high (with many over 8k), with the S2000 going all the way up to 9000rpm, the turbo inline 4 in the Lancer Evo runs over 7k, the Toyota 2ZZ-GE ran around 8k, the new BMW 4 cylinder turbo runs 7k, and all of them are far more reliable at that speed than the older V8s you're talking about.

    Finally, pretty much all of the numbers you're talking about aren't on a factory stock engine as it was delivered in a street car. Most of them had redlines in the 5800-6800 range, and in most cases, the power was falling off by that point anyways. Different cams (and quite possibly significant modifications to the entire valve train to prevent valve float) were needed to make top-end power on most of those engines, wheras many of the engines I mentioned above make power up to 7k or above.

    Reliability will also suffer with a high-RPM engine, due to the much higher rotational loads on the internal components. Making an engine rev to 9k for the duration of a race is very different than making an engine with a 9k redline that will last for 100,000 miles in a street car (which is why 9k capable cars are so expensive - off the top of my head, the only ones I can think of excluding rotaries are the S2000, which was close to $40k for a 4 cylinder powered Honda, the new Porsche 911 GT3, which is $130k, and the Ferrari 458, which is $300k+). Many motorcycles do rev that high and higher, but they typically have a much smaller displacement per cylinder, a shorter stroke, and much lower expected lifespan.
    Last edited: Oct 16, 2013
  22. Oct 16, 2013 #21

    I have to whole heartedly agree with cjl. Ranger mike, I am not familiar with the engines you seem to be familiar with of that era, but as you would be well aware engine component strength is the single biggest limiting factor in rev ceiling in any engine. Given that the average piston and rod out of any of these exceptionally large engines would be literally twice the weight of it counterpart I4 item from the same era (outside America), tie that with the fact that the old tech V8s that you speak of all run OHV which meant massive inertia in the valve train making it a requirement to run massively over on valve spring tension, anything but manufacturer bolting a partially built track engine in a stock car would see the revs you refer to from the factory floor. If you are going to compare engines, make sure you compare apples to apples!

    BTW, an I4 fires every 180deg, and doesn't require bob weights to balance!

  23. Oct 16, 2013 #22


    User Avatar

    Ahh - I just remembered 1 other engine that revs to 9k+ in a stock production car: the Toyota 1LR-GUE. 9000RPM redline V10, 4.8L, 560hp. It's used in the Lexus LFA.

    (Again, tiny per-cylinder displacement, high cost).
  24. Oct 17, 2013 #23

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    statement was -

    Sure, but 7000 rpm is awfully high for a 7L V8 though. It's a lot easier to make a 2L inline 4 rev than a 7L V8, simply due to the size of all the components...........................

    i believe i addressed this statement. A proper 4 cylinder rod/piston combo would weight the same. i will give you the fact that the crank is a lot lighter than the V8. And i can tell you the camshaft used on the stock V8 fell off after 6500 rpm but that is not to say the beasts could not rev tot 7000 rpm...re-read the statement..

    Sure, but 7000 rpm is awfully high for a 7L V8 though. It's a lot easier to make a 2L inline 4 rev than a 7L V8, simply due to the size of all the components...
    all you had to do was walk into the show room and buy the car and you could do 7000 rpm..universal statements fal apart if you can show one exception..i believe i have shown numerous
  25. Oct 17, 2013 #24
    The statement hasn't been addressed, I can't think of a single 3.5ltr 4cyl in existence, which means "A proper 4 cylinder rod/piston combo" which matches the 7ltr v8 for weight also doesn't exist. Most 4 bangers are less than 2ltr in capacity which would make the piston/rod assembly at LEAST 30% less than the 7lt V8.
    Regarding smoothness, I4 manufactured in japan after early 80's which are 2ltr or larger, run balance shafts to cancel 2 order harmonics.

  26. Oct 17, 2013 #25

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    damo have yo u ever weighed a piston/con rod when balancing them?
    stock 2000 liter Ford 4 cylinder con rod weighs 595 gram, piston weighs 482
    big V8 Ford 429 cid rod is 765 piston is 623..don’t have wrist pin or ring weight
    2000L con/piston = 1072 g while V8 totals 1388 g or a little over 20% more. A quick scan of my old balancing tables shows this to vary form 18 to 24% depending upon the manufacturer..Chrysler being heaviest. Properly prepared race engine will reduce this amount significantly. The stock pistons have huge rings and wrist pins..bulletproof but way too heavy, the forged steel con rods are junk and should be replaced.
    the point is 7000 rpm is high for any production automobile engine. I got no reason , desire or need to look at motorcycles. The statement that is is a lot easier to make a 4 cylinder rev simply due to size is a universal statement and one can not make sweeping assumption based on component size. Easier is a word with unlimited meanings in this context.
    this is a forum dealing with specifics and empirical data is a must.
    Last edited: Oct 17, 2013
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook