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

Crankshaft balance

  1. Apr 3, 2014 #1
    So for the last 3 years (off and on) I have been designing an engine. I have gotten much more serious lately as it could help me get a job that I seriously want, but somethings have stopped me in my tracks. On of the biggest things is the g.d. crankshaft. It is a pain because it does so much for so little... so, so little.

    Here is the question: is it possible for a v8 crankshaft to be angled to 45°? There are the old/mundane crankshafts that are angled to 180° and they were horrifically unsmooth, and then there is the 90° which is smooth but if it could be smoother that would be nice.

    The engine is designed to run up to 15000 rpm and put out about 550 hp with n/a and 13000 rpm and about 750 hp (really mild tune) with twin turbos. Now in my mind 45° crank would work but could it get ahead of its self? Could it be too close, possibly causing misfire or under rotational fire? Now what if it is pushing in nearly 3 bar of air? Could that cause problems?

    Thank you for any and all help.
  2. jcsd
  3. Apr 3, 2014 #2
    Why on earth do you think 45 throw offset would work?

    Flat plane crank is the only way to go at high rpm. Less mass, much less inertia, even firing on each bank so no need for a crossover exhaust. Only 'problem' is 2nd order vibrations, but they can be solved with balancer shafts.
  4. Apr 3, 2014 #3
    It would run fine a low RPM, but at these high revolutions you would find better stability if it were set at 90 degrees. 45* crank running at 15,000 RPM just seems way to tight.
  5. Apr 3, 2014 #4
    For the 180° bank, I thought that a crank angle would cause too much instability. Would that be wrong? So 45° is out for what I thought I it would be out for... but the 90° bank should still work.
  6. Apr 4, 2014 #5

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    For high rev engine flat crank is only way to go. Once you stick long con rods and light weight pistons in it the 1st and 2nd order harmonics can be resolved with various dynamic crankshaft vibration dampers on the marker today. Of course you will have NO low end and most likely will need a lot of gears in the transmission...
  7. Apr 4, 2014 #6


    User Avatar

    I have to admit, I'm scratching my head about what a 45 degree crank would even look like for a V-8. I'm also confused why you would describe a flatplane V-8 as "horrifically unsmooth" - dynamically, it's just a pair of inline 4 cylinder engines running together and sharing a crankshaft. On the couple of occasions I've had to drive a car with a flatplane V-8, it has seemed just fine to me (and clearly, an inline 4 works just fine, since it seems to be the engine of choice in most economy cars). The flatplane's lack of the heavy balance weighs on the crankshaft that are necessary for a crossplane V-8 also mean that the flatplane is much better for a high-rpm design, since the lower rotational inertia allows it to change speed much more quickly (which is important in a high-revving design). If you need a high-revving design that is inherently more balanced than a flatplane V8 for high RPM use, I would look at an inline 6, flat 6, or V12 design rather than trying to invent a new crankshaft for a V8.

    Interestingly, as a side note, the one manufacturer I am aware of that uses a flatplane design in all of their 8 cylinder engines is far from being old or mundane (as you describe for a 180 degree crank), but is instead internationally renowned for their high performance, high-revving cars. Their name? Ferrari.
    Last edited: Apr 4, 2014
  8. Apr 4, 2014 #7
    A 45 degree crank would give uneven firing. Itd be like a crossplane I4 crank with 45 degree offsets.

    Not sure if you could use flying arms or would need 8 throws.

    Either way, not woth boiling your brain over.
  9. Apr 4, 2014 #8
    Yeah, I know that ferrari does it (and I love me some sweet red ponies) but I am looking for smoothness to the ninth degree. The engine I am designing would be a street engine. Obviously it could be/would be used for racing but it should be usable on the street. You should be able to kick it into sixth and have it revving a little under 1000 and not have the whole car shake (also I know that the whole car doesn't shake but I have been in plenty of modded high performance cars that do).

    I guess what I need to ask is: Is the main important piece to engine stability the firing order being even (i.e. left right left right) and a damper or is it the crankshaft? I have heard arguments for both but you guys are engineers.. right?
  10. Apr 4, 2014 #9
    That is exactly what has been happening for the last few days. My head hurts. :biggrin:
  11. Apr 4, 2014 #10


    User Avatar

    Honestly, I don't think you're going to get an engine that is both happy at low revs (especially under 1000) and willing to rev to 15k (which is absurdly high for a street engine). Why do you want it to rev that high anyways? That having been said, I repeat what I said before - if you want a smooth engine, a V12 (best from a power and vibration standpoint, but large, most complicated and expensive), flat-6 (very good, kind of wide, so it can be difficult to fit in a normal engine bay), or inline 6 (pretty long for its displacement/power, but very smooth) is the way to go, not a V8.
    Last edited: Apr 4, 2014
  12. Apr 4, 2014 #11
    Okay, so 15k is a little high. I am realistically shooting for 12k. 15k would be nice but 12k is where my design feels best. To answer your question with a short answer: I need it to be functional in a race situation as it would be marketed as an engine that can be run in a race or on the street. Heres the other thing, although I am saying 12k would be nice, I have been thinking about cutting it back to a solid 10k (which is good and would feed the turbos more than enough air) to keep my engines from blowing up.

    [/QUOTE]That having been said, I repeat what I said before - if you want a smooth engine, a V12 (best from a power and vibration standpoint, but large, most complicated and expensive), flat-6 (very good, kind of wide, so it can be difficult to fit in a normal engine bay), or inline 6 (pretty long for its displacement/power, but very smooth) is the way to go, not a V8.[/QUOTE]

    I like the idea of a v12 but the cost WELL outweighs any benefits. I would love to do a boxer engine but once again, the cost outweighs the value and I need it to be much more compact. Now an inline six was a thought but my design incorporates something that would save me cost and heartache/headache so that was out.

    Essentially my design is a lot like the h1v8 which I had never heard of when I started my original design (which incorporated hayabusa heads). After showing it to a friend of mine, he told me of their product and I (in spite of them) revised my design to incorporate another head. There are other differences between my design and theirs but I felt like I would be stealing if I used the head I wanted to use initially.
  13. Apr 4, 2014 #12


    User Avatar

    Well, in that case, I don't know what to tell you. The features that make an engine work well at high rpm (long duration, high lift cams, short stroke, large bore, stiff valve springs, super low rotating/reciprocating mass) aren't conducive to the engine being happy at low RPM (nor are they conducive to having a usable amount of torque at low RPM, which makes their smoothness at low RPM somewhat irrelevant anyways). Those same features also aren't great for efficiency, which is increasingly important in street engines.

    For a high power engine happy on both race and street (including purring along at low RPM on the highway), a crossplane, large displacement V8 seems to be a better design than a super high revving engine to me. The Chevy LS series is a perfect example of what I'm talking about - it sees plenty of use in streetcars, has a great torque curve for usability, and even gets pretty good gas mileage with the right gear ratios on the highway, but a racing built motor can easily get over a thousand horsepower.

    (I'm also curious how you can say your design "feels best" at any RPM when you haven't even made simple decisions like final engine configuration or crankshaft shape/firing order yet)
    Last edited: Apr 4, 2014
  14. Apr 4, 2014 #13
    This is all a bit haphazard. You seem to have picked a set of criteria almost arbitrarily.

    15k a little high, ok 12k, or maybe 10. Why 10k, why not 9k, why not 2k?
    But it could be naturally aspirated putting out 500hp, or boosted by 2 bar to produce 700.
    It must work at high rpm, and low rpm.

    How have you arrived at any of your decisions for figures? This is not a criticism, its just not how things are designed.

    It's like you want a pair of shoes that you can run a marathon in, would be good for wading through rivers, yet look smart enough to get married in. But you can't decide if they should be slip on or laced. But they must be blue.

    It doesn't exist because it doesn't really make sense.
  15. Apr 4, 2014 #14
    I agree with Chris. When I read v8 engine running at 15k RPM I immediately thought this was a formula 1 prototype engine, not something that would be suitable for daily driving.
  16. Apr 5, 2014 #15
    honestly, if you want engine reving 15k, most likely it would not even IDLE below 1500 rpm...

    second thing, you want 550hp at 13000rpm. you need 300Nm for that. most engines produce around 100Nm/l so you want the 3.0l V8 ?
    that really don't sound very low-rpm streetable. cam profiles, intake and exhaust manifolds etc required for these extremely high revs are incompatible with low-end torque.

    and so on. I'm not sure that what you want (engine working "efficiently" from idle to F1 rpm territory) is even possible.
  17. Apr 5, 2014 #16
    Its impotant not to feel negative about this thread.

    I think you just need to follow a design process. And define a realistic operating envelope, then then defines the architecture of the engine and the parameters you need.

    You've effectively been making it up as you are going along.

    The important thing to realise is that race engines and road engines are hugely compromised. An engine designed to do both will do neither very well.
  18. Jun 2, 2014 #17
    I should let this thread die and bury my shame along with it but I cannot. I feel that I must explain. I have not yet made this design into a cad model. I only have what I have which is raw data. Till I can test it and I know its real limits, it revs to 15k, has 1m horses, and loves street driving.

    This was a theoretical idea and I thank everyone who posted onto this, thanks for the help. I will make a new thread for the future.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook