Why is it that if you want to have a car with a deep growling engine. it got to be a V8? There are almost no engines with more or less than 8 cylinders that have that growling sound. The only engine (that I know of)that isn't a V8 but still sounds like one is the Bugatti W16 but that is essentially 2 V8s put together. So why 8? How come a engine if less than or more can't produce that same sound?
Why is it that if you want to have a car with a deep growling engine. it got to be a V8? There are almost no engines with more or less than 8 cylinders that have that growling sound. The only engine (that I know of)that isn't a V8 but still sounds like one is the Bugatti W16 but that is essentially 2 V8s put together. So why 8? How come a engine if less than or more can't produce that same sound?
A V-twin can sound pretty rumbly too, especially on bikes like Harley Davidson's.
Some V8's don't growl that much either, especially European ones (e.g. a Ferrari V8 sounds very different to an American one.)
V6, V12 and V10 engines tend to scream rather than growl.
It is indeed all to do with frequencies and firing intervals. So for an 8 cylinder, you get 8 ignitions in every two crank rotations - i.e. once every 90 degrees of crank rotation. If that were the only frequency involved, you'd get a scream like the HSV-010, GP2 cars etc.
The crossplane V8 has two exhaust banks that usually can be heard separately, and a limitation of the crossplane V8 (as a four-stroke) is that you cannot have even firing within each bank. What you get is the "potato-potato" sound. One short interval, one long one and two normal ones: 90-180-270-180.
A flat-plane four is even firing: 180-180-180-180, and that is the basis for the flatplane V8. But since both flatplane and crossplane V8s are even firing overall, it's possible to get either sound from both with a bit of inventive plumbing (e.g. "bundle of snakes" exhaust).
Knowing all of that, it's easy to construct firing orders that have a similar sound.
For two cylinders, a 90° V-Twin has intervals of 270-450 - kind of proto-cross-plane in sound. Think Ducati, not so much Harley (315-405).
For three cylinders, try 270-270-180 (most triples are even firing; Laverda used 180-180-360 for a while)
Four cylinders, like the crossplane R1, are 90-180-270-180, just like one bank of a V8. Some V4s are the same: e.g. Honda VFR800 - Aprilia's V4 and Yamaha's VMax are very similar. Ducati's new V4 was designed to sound like its old twins: 90-200-90-340 - grouping achieves 290-430, close!
Five cylinders, try 90-180-90-180-180 - that "double 180" is 270° out from the single 180
Six cylinders, try 90-180-90-90-90-180 - similar deal with the five cylinder, except between the triple 90 and the single 90.
With seven, the sound breaks down into an impression of a misfiring 8 cylinder.
You don't have to move the ignition events themselves, just when the pulses arrive at the tailpipe - in the same way you can get the boxer sound out of an inline four.
You can look at it differently, in terms of fractions:
270-450 is 3/8 then 5/8, and that combination of factors must account for the juicy sound, in much the same way that three cylinders (and six, twelve etc.) and five cylinders (and ten) sound unique and appealing.
In fact, you can calculate the precise contribution of each frequency in the same way you can for engine balance, and a pattern emerges. That's what we recognise, somehow
Starting again with the twins, you can go further. For instance, the intervals 288-432, corresponding to a 72° V-Twin, have more than a little bit of the crunchy 5 cylinder growl about them. 288° crank degrees corresponds to two firing events in an even firing five (720 / 5 = 144), meaning the split in fractions is 2/5 and 3/5. You can again scale that up to more cylinders: the triple sounds great, the four just like a misfiring five. Endless fun!
It is indeed all to do with frequencies and firing intervals. So for an 8 cylinder, you get 8 ignitions in every two crank rotations - i.e. once every 90 degrees of crank rotation. If that were the only frequency involved, you'd get a scream like the HSV-010, GP2 cars etc.
The crossplane V8 has two exhaust banks that usually can be heard separately, and a limitation of the crossplane V8 (as a four-stroke) is that you cannot have even firing within each bank. What you get is the "potato-potato" sound. One short interval, one long one and two normal ones: 90-180-270-180.
A flat-plane four is even firing: 180-180-180-180, and that is the basis for the flatplane V8. But since both flatplane and crossplane V8s are even firing overall, it's possible to get either sound from both with a bit of inventive plumbing (e.g. "bundle of snakes" exhaust).
Knowing all of that, it's easy to construct firing orders that have a similar sound.
For two cylinders, a 90° V-Twin has intervals of 270-450 - kind of proto-cross-plane in sound. Think Ducati, not so much Harley (315-405).
For three cylinders, try 270-270-180 (most triples are even firing; Laverda used 180-180-360 for a while)
Four cylinders, like the crossplane R1, are 90-180-270-180, just like one bank of a V8. Some V4s are the same: e.g. Honda VFR800 - Aprilia's V4 and Yamaha's VMax are very similar. Ducati's new V4 was designed to sound like its old twins: 90-200-90-340 - grouping achieves 290-430, close!
Five cylinders, try 90-180-90-180-180 - that "double 180" is 270° out from the single 180
Six cylinders, try 90-180-90-90-90-180 - similar deal with the five cylinder, except between the triple 90 and the single 90.
With seven, the sound breaks down into an impression of a misfiring 8 cylinder.
You don't have to move the ignition events themselves, just when the pulses arrive at the tailpipe - in the same way you can get the boxer sound out of an inline four.
You can look at it differently, in terms of fractions:
270-450 is 3/8 then 5/8, and that combination of factors must account for the juicy sound, in much the same way that three cylinders (and six, twelve etc.) and five cylinders (and ten) sound unique and appealing.
In fact, you can calculate the precise contribution of each frequency in the same way you can for engine balance, and a pattern emerges. That's what we recognise, somehow
Starting again with the twins, you can go further. For instance, the intervals 288-432, corresponding to a 72° V-Twin, have more than a little bit of the crunchy 5 cylinder growl about them. 288° crank degrees corresponds to two firing events in an even firing five (720 / 5 = 144), meaning the split in fractions is 2/5 and 3/5. You can again scale that up to more cylinders: the triple sounds great, the four just like a misfiring five. Endless fun!
Semi-related: I really, really want someone to stick a Vmax engine on the nose of a Morgan 3 Wheeler, it's the only way to make that 'car' even more rude.
It is indeed all to do with frequencies and firing intervals. So for an 8 cylinder, you get 8 ignitions in every two crank rotations - i.e. once every 90 degrees of crank rotation. If that were the only frequency involved, you'd get a scream like the HSV-010, GP2 cars etc.
The crossplane V8 has two exhaust banks that usually can be heard separately, and a limitation of the crossplane V8 (as a four-stroke) is that you cannot have even firing within each bank. What you get is the "potato-potato" sound. One short interval, one long one and two normal ones: 90-180-270-180.
A flat-plane four is even firing: 180-180-180-180, and that is the basis for the flatplane V8. But since both flatplane and crossplane V8s are even firing overall, it's possible to get either sound from both with a bit of inventive plumbing (e.g. "bundle of snakes" exhaust).
Knowing all of that, it's easy to construct firing orders that have a similar sound.
For two cylinders, a 90° V-Twin has intervals of 270-450 - kind of proto-cross-plane in sound. Think Ducati, not so much Harley (315-405).
For three cylinders, try 270-270-180 (most triples are even firing; Laverda used 180-180-360 for a while)
Four cylinders, like the crossplane R1, are 90-180-270-180, just like one bank of a V8. Some V4s are the same: e.g. Honda VFR800 - Aprilia's V4 and Yamaha's VMax are very similar. Ducati's new V4 was designed to sound like its old twins: 90-200-90-340 - grouping achieves 290-430, close!
Five cylinders, try 90-180-90-180-180 - that "double 180" is 270° out from the single 180
Six cylinders, try 90-180-90-90-90-180 - similar deal with the five cylinder, except between the triple 90 and the single 90.
With seven, the sound breaks down into an impression of a misfiring 8 cylinder.
You don't have to move the ignition events themselves, just when the pulses arrive at the tailpipe - in the same way you can get the boxer sound out of an inline four.
You can look at it differently, in terms of fractions:
270-450 is 3/8 then 5/8, and that combination of factors must account for the juicy sound, in much the same way that three cylinders (and six, twelve etc.) and five cylinders (and ten) sound unique and appealing.
In fact, you can calculate the precise contribution of each frequency in the same way you can for engine balance, and a pattern emerges. That's what we recognise, somehow
Starting again with the twins, you can go further. For instance, the intervals 288-432, corresponding to a 72° V-Twin, have more than a little bit of the crunchy 5 cylinder growl about them. 288° crank degrees corresponds to two firing events in an even firing five (720 / 5 = 144), meaning the split in fractions is 2/5 and 3/5. You can again scale that up to more cylinders: the triple sounds great, the four just like a misfiring five. Endless fun!
This thread's in the running for Thread of the Year in this year's GTP Awards, I reckon.
While we're picking your brain @Griffith500, does firing order also play a part in the wide discrepancy in the sound of V6s? I know the angle of the V seems to play a part when it comes to smoothness and different intake and exhaust characteristics will alter how they behave, but is there much difference in firing order across the V6 spectrum? Is say, a highly-regarded V6 like that of a Honda NSX or Alfa's Busso markedly different from something more run-of-the-mill like a Camry's V6 or the widely-used Buick V6?
This thread's in the running for Thread of the Year in this year's GTP Awards, I reckon.
While we're picking your brain @Griffith500, does firing order also play a part in the wide discrepancy in the sound of V6s? I know the angle of the V seems to play a part when it comes to smoothness and different intake and exhaust characteristics will alter how they behave, but is there much difference in firing order across the V6 spectrum? Is say, a highly-regarded V6 like that of a Honda NSX or Alfa's Busso markedly different from something more run-of-the-mill like a Camry's V6 or the widely-used Buick V6?
Alfa Romeo Busso (60°): 1-4-3-6-2-5
Nissan VQ (60°): 1-2-3-4-5-6
Honda "C" (NSX) (90°): 1-4-2-5-3-6
Buick OHV V6 (90°): 1-6-5-4-3-2 (incidently, this is essentially a small bock Buick V8 with two cylinders cut off, hence the 90° vee)
There is a lot more to this, including odd firing V6s (original Buick design), how the exhaust manifolds are arranged, etc, etc. I know some of the basics, but would need to more digging (or wait for Griffith to post again ) to explain exactly why different V6s sound different.
Hah. Thanks 👍 Though I was actually thinking of firing interval (i.e. the degrees at which each cylinder fires) rather than firing order, assuming the two aren't inexorably linked.
A lot of it has to come down to exhaust and intake tuning (sound tuning) Alfa's V6's were always used in their sporty range-toppers, but Lancia and Fiat (i think) also used it some of their cars, but i think it was just a bigger engine option in those. No doubt the Alfa version sounded better even though it was technically the same. The Camry's V6 is just a run-of-the-mill American-market engine choice so they won't bother making it sound anything special. My Audi A3 has the exact same VR6 as the Golf R32, but from the outside they sound quite different. Can only be down to intake and exhaust differences. The R32 was VW's sporty range-topper, where as mine was just an engine choice and visually looks exactly the same as every other A3 of the period with the 'sport' spec box ticked.
This thread's in the running for Thread of the Year in this year's GTP Awards, I reckon.
While we're picking your brain @Griffith500, does firing order also play a part in the wide discrepancy in the sound of V6s? I know the angle of the V seems to play a part when it comes to smoothness and different intake and exhaust characteristics will alter how they behave, but is there much difference in firing order across the V6 spectrum? Is say, a highly-regarded V6 like that of a Honda NSX or Alfa's Busso markedly different from something more run-of-the-mill like a Camry's V6 or the widely-used Buick V6?
I don't know nearly as much about V6s, particularly in terms of individual applications and any subtle variations. This is arguably a bigger question, because it's about more subtle differences than simply "what makes the crossplane sound?"
However, as far as I know, there is only really one V6 firing order - two if you count its reverse, and maybe an array of mirrored versions (left-to-right, front-to-back, rotated etc.) - the shape is the same. It is quite a lumpy order, being that it starts at one end of the engine and visits each cylinder in turn along the crank's length.
The difference noted above is mainly because of differences in cylinder numbering. DIN states you number down one bank, then down the next (like Ford), many manufacturers number along the crankshaft (like GM) and the Italians sometimes do their own thing of numbering down one bank, then back up the next - like a horseshoe.
There are some differences according to whether the engine rotates clockwise or anticlockwise relative to the end of the engine that has cylinder number 1, and whether the "left" or "right" bank gets cylinder 1 as well. But if you draw them all out, they all look the same - and since the order is a cycle, starting from a different point doesn't make it a different cycle.
An even firing six cylinder must have 120° firing intervals. That again is not the predominant sound in most six cylinder engines, again because of the separate banks - some racers do have that nice 6 cylinder howl, including the aforementioned V12 F1s. You need a well tuned, equal-length 6-into-1 pipe for that.
The two banks are always 240-240-240, even firing. But any unevenness in the intake, exhaust pipes or fueling etc. highlights a particularly nice 240-480 interval that has the quality of a perfect fifth, or "power chord" in musicians' parlance. The ratio of the frequencies in said chord is 3:2 - notice the split in the interval is 1/3 and 2/3.
The easiest way to achieve 120° intervals in a V6 is is to make the bank angle 120°, but it's not very well balanced, and is very wide - about as wide as a boxer 6. There are nonetheless a few examples, including Ferrari.
The old Buick "Fireball" or "odd-fire" V6 is an exception already noted: 90-150-90-150-90-150. You can clearly see the influence of the common crank pin and the 90° V-angle, coupled with the 240 degree crank throw spacings (like a triple).
The uneven firing aspect brought extra rawness to the sound that is probably best exemplified by the Cosworth DFV-inspired V6, used in the Metro 6R4 Group B car and various TWR Jaguars, including the XJ220.
Because the first odd-fire V6s had a rough idle, like a triple, what with the mass-produced fueling and ignition tech of the time, they weren't initially very popular. But changing the bank angle was not going to be tolerated, since it was selected for the ability to share tooling with the V8s and because it balances nicely with crank counterbalances, also like a V8.
So to achieve even firing and smoothness, they sacrificed the balance aspect slightly and used "split-pin" cranks, where each crank throw had two separate journals machined into it, split at 30°. Here's Audi's version:
Many engines went with a 60° bank angle for even more compactness, and not terrible balance. The split needs to be 60°, so large that it needs a proper web between each throw - the crankshaft is noticeably longer as a result. Not the performance option, you might think, but Alfa Romeo and even Ferrari (60 or 65 degrees) used it for some time.
Talking about the Busso, the intake gives that engine its glorious gurgle, by having one long plenum along the entire length of the engine, with each cylinder connected to it in turn via runners. The NSX is similar. Some engines use twin plenums, isolating the triple growl, or one large one with the cylinders connected more or less in pairs, side by side - this sounds more warbly, i.e. Nissan, possibly best reflecting the firing order itself. Some had complicated cast plenums connecting even-firing (360-360) pairs for better torque, not unlike a straight six, giving a honking, buzzing tone.
And things have come full circle, as the V6 in the Giulia Quadrifoglio is an uneven firing, single-pin-cranked 90° V6 - which is why it sounds so sweet! Sometimes you can engineer in a nice sound via packaging and plumbing choices, like the Busso, but often it helps to start with a strong basis: harmonics dictated by the firing intervals. But those harmonics need to be treated carefully, as in the case of the rough idle of the original Buick V6, or crank torsion in something like the high-performance Giulia.
However, current F1 V6s are also single-pin 90° engines. The exhaust note is drier because they are closer to equal length, removing some of that "power chord" vibe that makes all three cylinders, and their derivative sixes and twelves etc., so appealing, without having the benefit of the even-firing scream.
Me & Griffith had a very, very long and technical conversation about how to make a Nissan VQ sound like a BMW M88 a few years ago. You can guess who was doing most of the technical talk.
For the record, the V6s I've enjoyed the most are the NSX's 3.2 (which needs no real explanation - incredible induction noise and exciting power delivery), and Mazda's 60-degree K-series V6 in 1.8-litre form in the MX-3. The Mazda V6 could do with a little more top end - it starts to run out of breath towards the red line - but it's one of the smoothest engines I've used this side of a rotary and has a lovely fizz (not unlike a Busso) through the rev range.
The current Giulia Quadrifoglio's V6 is lovely too, but that's more for its performance than the sound necessarily - it's slightly too "over-processed" for me (something true of virtually all modern production engines). Fairly faultless in terms of delivery though - great responses despite the turbocharging.
For the record, the V6s I've enjoyed the most are the NSX's 3.2 (which needs no real explanation - incredible induction noise and exciting power delivery), and Mazda's 60-degree K-series V6 in 1.8-litre form in the MX-3. The Mazda V6 could do with a little more top end - it starts to run out of breath towards the red line - but it's one of the smoothest engines I've used this side of a rotary and has a lovely fizz (not unlike a Busso) through the rev range.
The current Giulia Quadrifoglio's V6 is lovely too, but that's more for its performance than the sound necessarily - it's slightly too "over-processed" for me (something true of virtually all modern production engines). Fairly faultless in terms of delivery though - great responses despite the turbocharging.
Have to agree regarding the Honda & Mazda V6s. I actually tend to like 3, 6, & 12 cylinder engine sounds the best. For me they have so much more depth and character than 4s, 8s, & 16s. 5s and 10s scream, but they sound rather flat to me.
I still think the Pro-car spec M88 (inline 6) is the greatest sounding engine of all time. Absolutely thundering intake noise but also a lot of character from the exhaust which you can hear most clearly on part throttle (and of course exterior flybys). Part of this is due to the engine placement in the M1 and the 6 intake trumpets over your right ear.
With thoughtful plumbing work, you can make even basic V6 engines sound pretty exotic:
Have to agree regarding the Honda & Mazda V6s. I actually tend to like 3, 6, & 12 cylinder engine sounds the best. For me they have so much more depth and character than 4s, 8s, & 16s. 5s and 10s scream, but they sound rather flat to me.
I was a big fan of three-cylinder engines for a while but I've gone off them recently. In smaller cars over here they've become as ubiquitous as four-cylinders, and modern refinements have dulled most of the character out of them without being able to completely remove negative aspects like the vibrations. While I love the way the Caterham Seven 160 drives, that little 3cyl Suzuki engine is probably its worst aspect, not in terms of performance which is fine in a car that light, but in the way it sounds and the way it struggles to rev.
I'd agree on the Procar too, to which I'd add the inline six in the race-prepped Hakosuka GT-R in this video (the best sounds start around 1:15 through the video). Basically, race-prepped inline sixes running through carbs are pretty much perfection.
I've driven Audi's inline-fives and V10s and again I'd agree - they sound interesting, but if I never heard either again I'd not be too upset. I'm a fan of V8s though, if only because over here their rarity always makes them feel quite exotic - even a Mustang GT feels quite special compared to a lot of the stuff I get my hands on.
I was a big fan of three-cylinder engines for a while but I've gone off them recently. In smaller cars over here they've become as ubiquitous as four-cylinders, and modern refinements have dulled most of the character out of them without being able to completely remove negative aspects like the vibrations. While I love the way the Caterham Seven 160 drives, that little 3cyl Suzuki engine is probably its worst aspect, not in terms of performance which is fine in a car that light, but in the way it sounds and the way it struggles to rev.
I'd agree on the Procar too, to which I'd add the inline six in the race-prepped Hakosuka GT-R in this video (the best sounds start around 1:15 through the video). Basically, race-prepped inline sixes running through carbs are pretty much perfection.
I've driven Audi's inline-fives and V10s and again I'd agree - they sound interesting, but if I never heard either again I'd not be too upset. I'm a fan of V8s though, if only because over here their rarity always makes them feel quite exotic - even a Mustang GT feels quite special compared to a lot of the stuff I get my hands on.
Ha. For me V8s are totally ubiquitous (especially growing up in semi-rural Texas) and 3 cylinders are a complete novelty. Interesting how experience can shape perception.
I can can't see a thread on engine noise without wanting to describe how crying tears of joy is an involuntary response to Top Fuel dragsters... so yeah, that.
I can can't see a thread on engine noise without wanting to describe how crying tears of joy is an involuntary response to Top Fuel dragsters... so yeah, that.
They do sound unusual. It's mainly a result of the unequal port lengths, they alternate between long and short because of the staggering of the cylinders. The intake manifolds on newer models are designed to account for that, the effect of which can be heard in that video - it's got the typical 360-360 honk, not too dissimilar to a modern boxer 6 or many straight sixes (example). But the exhaust has the warbly / lumpy sound caused by the effective grouping of sequentially firing cylinders.
This comes about from the firing order: 1,5,3,6,2,4.
The cylinders having the same port lengths would be grouped: 1,3,5 and 2,4,6, due to the alternated staggering.
Hopefully you can see that the intervals in each group are 120-120-480 - very lumpy! Then you add on any extra unevenness in e.g. the exhaust manifold (there's not normally much room for it) and you get quite a characterful sound.
Somehow that lumpiness always worked better on the intake side, for me. Kinda reminds me a bit more of the old Ferraris, with their three carburettors (example), each receiving uneven pulse intervals and including at least one sequential (60°). It is, however, also remarkably close to something like a single plenum V6, i.e. a Nissan VQ - the intervals for the pairs of cylinders that would share a crankpin, if they weren't split, is 120-600: sequential.
Here's an equalised exhaust, where some runners are longer than others to account for the two different port lengths
And here's an ITB setup with each bell-mouth the same length, accentuating the inherent lumpiness, over and above a straight six with individual, open throttles (like an M1 Procar)
As for triples being a little dull, I think that's generally more the fault of the cars they're installed in, in the sense that it dictates the kind of powertrain you get. Very few "economy" cars have genuinely interesting, fizzy engines these days - that said the only three cylinder car I've driven is a BMW Diesel, in a "Mini". In motorbikes, on the other hand...
I've ridden a few of those, and it's my second-favourite engine configuration - after the 90° V-Twin (dat pulsing). The induction growl, best heard in the second video, is intoxicating and possibly even dangerously, um, "motivating".
I was a big fan of three-cylinder engines for a while but I've gone off them recently. In smaller cars over here they've become as ubiquitous as four-cylinders, and modern refinements have dulled most of the character out of them without being able to completely remove negative aspects like the vibrations. While I love the way the Caterham Seven 160 drives, that little 3cyl Suzuki engine is probably its worst aspect, not in terms of performance which is fine in a car that light, but in the way it sounds and the way it struggles to rev.
I'd agree on the Procar too, to which I'd add the inline six in the race-prepped Hakosuka GT-R in this video (the best sounds start around 1:15 through the video). Basically, race-prepped inline sixes running through carbs are pretty much perfection.
I've driven Audi's inline-fives and V10s and again I'd agree - they sound interesting, but if I never heard either again I'd not be too upset. I'm a fan of V8s though, if only because over here their rarity always makes them feel quite exotic - even a Mustang GT feels quite special compared to a lot of the stuff I get my hands on.
As for triples being a little dull, I think that's generally more the fault of the cars they're installed in, in the sense that it dictates the kind of powertrain you get. Very few "economy" cars have genuinely interesting, fizzy engines these days
I think that's something to do with it, though having driven a few older three-cylinder cars (not classics as such, but stuff from 10-15 years ago) there seems to have been a big push to make modern triples as smooth and quiet and personality-free as possible, probably because most buyers are downsizing from smoother four-cylinders for them.
I quite liked the 3cyl in my old Honda Insight, though that was a long way from being a performance engine. The old Smart Fortwo CDI's diesel 3cyl was interesting too - diesel four-pots are nasty things so the slightly offbeat feel of the 3cyl just made it a more pleasant engine to use. And Suzuki's current turbocharged "Boosterjet" 3cyl is an excellent engine, but not inherently because it's a triple. Sound generators help - both the VW Up GTI and the Mini-derived three in the BMW i8 have quite a nice note to them, but ultimately they don't quite feel right as performance engines in cars.
As for other interesting engines in economy cars I have to mention the Fiat TwinAir. That one really does feel different from everything else. I quite like the really unusual (for a car) parallel twin sound, and Fiat's done a good job of reducing vibrations, but I suspect you couldn't use an engine like that too widely as it might be a step too far for many customers.
I think its economy-car roots are just a little too apparent. Although Suzuki's used it in everything forever - including the Cappuccino - it was never really designed to be a performance car engine. In the Seven it feels rough and loses interest in being revved after the mid-range. The one Caterham has tuned up to 95bhp in the SuperSprint is better, but they're weirdly characterless otherwise and generally overwhelmed by a farty and tuneless exhaust note.
Ha, I was thinking of the Fiat's little twin cylinder when I wrote that! It is unusual, for sure, and I'm not really a fan of 360-twins generally, including boxers and the like. As far as flatulent exhaust notes go, they bear the crown in my opinion (well, perhaps ignoring high-revving four-stroke singles). But I'm pleased we have that variety, at least. And sometimes I find one I like the sound of.
Anyway, in terms of vibrations, once you pop a balance shaft on 'er, all that's left to worry about is the half-as-frequent combustion intervals. Admittedly, that is a pretty big deal with a car engine (it's why there were multi-cylinders fairly early on), but with clever ignition tricks and precise fueling with a good cylinder head (not necessarily performance oriented) you can mask it at low load. Get it spinning and it's no buzzier than a four, which have to have their own balance shafts these days. The drivetrain will take a beating if you lug it, though, which the target market are highly likely to do - that will necessarily dull its responses once you've applied the requisite beefing up. That is also the main issue with a three cylinder, it just doesn't fire frequently enough - bring back the two strokes, I say (1) (2)
It's a real mistake for engineers to assume that dulling vibrations and dulling sound are related targets, even though they are clearly related physical phenomena and are both covered by NVH standards and legislation. It's not enough to just make it quieter, as you say, it must have character. Manufacturers have gone to a lot of trouble with these "sound tubes" and the like, but honestly the best way is not to silence them in the first place. At least not the frequencies and, more importantly, the harmonics we want to hear. That, of course, is the big trick.
I have fond memories of drawing stares in the Pyrenées one summer as we growled our way up every climb in our 3 cylinder VW Diesel hire car!
I like a V10, that harmonic major third it inherits from the 5 cylinder (5:4 ratio) is sweet sounding on its own. Add in the uneven-firing common-pin 90° variety (sensing a theme?) and you get a gloriously crunchy gargle of an intake sound that practically makes a Gallardo - of course, it's easier to find a video of an S85 instead.
Ha, I was thinking of the Fiat's little twin cylinder when I wrote that! It is unusual, for sure, and I'm not really a fan of 360-twins generally, including boxers and the like. As far as flatulent exhaust notes go, they bear the crown in my opinion (well, perhaps ignoring high-revving four-stroke singles). But I'm pleased we have that variety, at least. And sometimes I find one I like the sound of.
Missed your reply to my comment before, but just rediscovered this thread and I found this line interesting. I know harmonics are something a lot of manufacturers are trying to tune out - the undesirable ones at least. I believe the technology around direct injection and the way engines behave running on that can introduce some unwanted harmonics, and today's high gearing means they spend a lot more time pulling from lower revs, at engine speeds that used to mean labouring the engine. I've certainly heard of manufacturers aiming to tune out resonant frequencies with noise-cancelling tech.
Being a bit spoiled like that, I've spent the last couple of weeks cruising around in V6s and V8s and it's really, really illustrated to me what we're missing from everything being turbocharged and four-cylinder. You can do what you want with an inline four but the way a six or eight idles and spins instantly makes it feel more expensive and cultured, even if you never use its performance potential. I think my next car may very well have a V8 just so I can experience that day-to-day for a while.
We definitely need both, engines with character and those that are as transparent as possible.
I've been kind of avoiding going into too much detail on the whole harmonics thing, but I've done it now. And it's probably just going to raise many more questions than it can answer, but I've tried to make it as clear as I can! Lots of text, unfortunately.
So, in terms of the harmonics, I sometimes think of it as being like a musical instrument, say a subtractive synthesiser. The engine provides the initial waveform (and hence its harmonic structure) and then the whole of the vehicle acts as a filter. Some of those parts have resonances, and the problems arise when the resonance coincides with one of the harmonics, especially at a favoured / more likely engine speed, e.g. corresponding to certain speed limits.
If a) is the harmonic signature of the engine, and b) is the "filter", e.g. the frequency response of a part or assembly, then c) shows the potentially troublesome result.
Now, clearly, the engine speed changes, so the "pitch" of that input to the car-as-filter changes, and so do the harmonics in proportion. So parts have to be designed to not vibrate at all of the harmonics of those engine speeds. And not just in isolation, but also as a complete assembly - it's a really monumental task.
I say all the harmonics, but this is rather the point: not all are as strong as others. It's often stated that the 3rd and 5th harmonic are particularly strong / troublesome in crossplane V8s (that's 1.5 and 2.5 times crank speed, because four strokes), whereas the 4th harmonic is absent entirely. In a flatplane, you have a regular progression of harmonics equally spaced and contributing equally (more or less), with very little to ideally no contribution from the "odd" harmonics. The crossplane's vibration (and sound), then, owing to its uneven firing, has a much more complex harmonic structure.
I've included several examples of combustion harmonics for different firing intervals, attached as pictures - these would be felt, but also represent the sound to an extent. The Boxer Rumble is a special case I've highlighted, in that it is an even firing engine, but sounds richer because of unevenness in the exhaust (a bit like a cross-plane V8, but for a subtly different reason). The same harmonic enrichment occurs from other sources of unevenness: e.g. in cam, fuel and ignition timing etc. including misfires or differences in volumetric efficiency, compression etc. in each cylinder; vibration can also be caused / worsened by large variances in the masses of the reciprocating parts as well.
On the one hand, the flatplane is advantageous in vibrational (NVH) terms because if a component on the vehicle vibrates for one harmonic, it will for another at the corresponding engine speed. That means tuning an object out for one harmonic can do so for most others as they occur. However, if there is a troublesome vibration somewhere, it will be constantly excited as the engine speed changes and each harmonic picks it up in turn.
The cross-plane can excite far more frequencies more of the time, and you can feel this from the engine, but any interactions between troublesome resonances and engine harmonics become seemingly random and usually much smaller, although it will affect more components. Also, since the driving frequencies aren't all evenly spaced, it has a more complex, rumbly and textured feel, rather than a simple buzz, which means you might not have to damp it out so much anyway as it's not so intrusive or unpleasant.
These harmonics all have a phase as well as a magnitude, but I've only plotted the magnitude - sometimes the phases can cause interference and reduce the impact of resonances, sometimes they will make them worse, depends on the situation (hence "random"). It can be hard to predict what might be excited by which harmonic and when, but once something is vibrating it will be pretty clear from measurements which harmonic is at fault. Then it's a matter of shifting the object's resonances by adding material, changing shape etc. or reducing the excitation directly by changing the engine speed, e.g. through gearing.
I have been fortunate to ride bikes with cross-plane and flat-plane cranked four pots, and in those cases the engine vibration is naturally going to be more easily felt as compared with a typical car, so I think it makes for an excellent illustration. And the issue is always that you will never isolate all vibrations in all situations, for structural reasons (rigid coupling) or for reasons of cost / complexity, weight etc. So the natural character of the engine is going to rear its head at some point in any vehicle (which says a lot about electric cars), but it more or less defines a bike.
The flatplane has a really strong, buzzy character at particular engine speeds, and then constantly above a certain speed (which can cause fatigue). This isn't helped because the ignition intervals are twice per crank revolution, which is the same as the start-stop motion of the pistons, and that gives rise to two linked phenomena. The first is an oscillation in the output torque, which can be felt as a vibration at certain times, especially at very high engine speeds - you can also often hear it in the final chain drive of a motorcycle. Secondly, it creates the characteristic second-order imbalance of your typical inline four, where it generally requires a balance shaft for modern NVH standards. That said, inline fours were originally lauded for their "turbine-like" power delivery in a bike, although I think you still need a six even then.
The crossplane is not as smooth at idle, it murmurs and tingles pleasantly (I think), and a blip of the throttle gives a soft, drumming sensation, as opposed to the sharp, buzzing sensation of the flat plane. It really does feel and sound like a smooth / tiny V-twin. Because the combustion intervals are scattered somewhat, there's no real buzzing from that, although you do feel a slight pulsing from the strong first harmonic which is mostly absent in the flat plane in the real world. That is less noticeable at high engine speeds and, because the engine is balanced up to fourth order (equivalent to 8th combustion harmonic), it is much smoother up top, save for a very slight, purposeful thumping. It all adds up to a surprisingly refined yet characterful experience for such a rampant motor.
Here's a bevy of on-boards that are deliberately poor quality audio-wise, but they capture the mechanical vibrations within the machine very nicely. The Honda (a flat-plane) has a constant wobble to its sound, the same kind you can hear in the chain drive as an observer - it's not that second-order vibration directly, but some kind of sub-harmonic of it - most parts assemblies have many, many, many "modes" of vibration, some harder or easier to excite by any particular frequency / motion. Notice it doesn't stop vibrating at any point over 7000 rpm.
[LINK]
The Yamaha (crossplane) still has a bit of final drive flutter, but it's more subdued compared with the rest of the sound and doesn't really seem "in tune" with it. Yamaha of course claimed that they chose the cross-plane crank because it minimises the rotational speed fluctuations caused by the pistons' start-stop motion, and this allegedly helps the tyre somehow (it's all still shrouded in mis-information, e.g. the "big-bang effect".) There is a slight, but definite sense of a single-cylinder / thumping at 14 000 rpm, although it's almost two-stroke-like in its sound at that speed.
[LINK]
For completeness, a triple. The rotational speed fluctuation is surprisingly still there, despite the three equally separated piston phases, similar in concept to the cross-plane. In three-phase electric power, this would result in a perfectly neutral average, but the non-sinusoidal piston motion has other ideas. What occurs is a 3rd order (6th harmonic!) fluctuation on top of the 1.5 order (3rd harmonic) of the combustion intervals. The separation of frequencies means less perceptible vibration compared to a flat-plane and, although you do still get the even firing buzz, it's mainly very noticeable only when tractoring around at low speed. Interestingly, you can clearly hear a drivetrain flutter which indicates the oscillation is significant even though the driving force is at a different, higher frequency (higher frequencies are generally lower amplitude and more likely to be self-damping). An even-firing six cylinder simply doubles the combustion frequency so it matches the inertial variation, just as in the flat-plane four, only at different harmonics.
[LINK]
As I said before, an even firing twin would have the same buzzy inertial characteristics of a flat-plane four, just with half as frequent combustion, making it almost the worst combination of the flat-plane and the triple!
[LINK]
A 90° V-twin has the minimum torque oscillation (from piston motion) for any twin, assuming long connecting rods as in a typical motorcycle engine. They're well balanced as well, so they should be silky smooth if it weren't for the lumpy power delivery. Indeed, although the rotational inertia is more constant than even an equivalent triple, it still upsets the drivetrain due to the pair of torque hammers attached to the crankshaft!
[LINK]
Compare a two-stroke V-Twin (intervals 90-270 instead of 270-450; the former is the equivalent of 180-540 in a four stroke, but twice as frequent.)
[LINK]
Some more graphs to illustrate the point:
~100 Nm / litre & 200 bhp / litre @ 14 000 rpm, 400 g reciprocating mass, 81 x 48.5 mm bore / stroke, 2:1 rod.
And Yamaha's crossplane video to uniquely explain the situation as it stands for motorcycles.
The inertial fluctuation is still important for cars, as it partly determines flywheel sizing. Due to some unknown (to me) reason, "rephasing" old parallel twins to be closer to 90 / 270 degrees (from 0 / 360) makes for a more responsive engine, so the concept even has performance implications in cars. The "inertial torque" has its own harmonic structure, too, and has serious implications in very large engines, such as those used in container ships.
There was at least one instance of an engine being constructed and more or less immediately destroying itself upon running due to crank resonances being excited by the piston motion itself, rather than the cylinder pressure. Such things are (were?) not normally accounted for in the design step (it is simpler to assume constant inertia), but clearly could be taken advantage of if you want a really smooth engine.
I haven't properly gone into crankshaft torsional vibration, but as I just hinted, it's much the same situation: there are excitations from combustion forces, and those by the inertial forces. As I'm sure I've said before, it was thought in the late '50s that a cross-plane V8 was the ideal choice for an F1 engine, because it should be smoother thanks to its ability to be easily balanced. However, the torsional vibration of the uneven firing along the length of the cross-plane crank really becomes a problem at higher rpm, so in fact the flat-plane feels smoother!
The inertial harmonics are all the even numbered ones in the graphs above - if you look at the inertial torque plot for the 270 twin, you can see the influence of the 2nd and 6th harmonics. For engine balance ("free force", inline engine), 2 and 4 are the most important harmonics to consider, being first / primary / crank-speed and second order imbalance respectively. Higher inertial harmonic numbers have a very small influence for most engines as their magnitude drops off rapidly, but combustion harmonics stay strong over a wider range because of the sharp nature of the pressure rise in the cylinder.
) to explain exactly why different V6s sound different.
