V-angles, Cross Planes, and Flat Planes

I was looking at old Duesenberg stuff today when I noticed that the old straight eights seem to have a cross plane crank shaft. Judging by the sounds I've heard from them, I suspect a lot of the early Bugattis did as well. 

This made me realize I don't understand balance principles. I had thought that the crossplane design of most large displacement V8 engines was a design consideration to reduce vibrations primarily caused by the 90 V configuration. Ie, the second order harmonics of a flat plane V8 or I4 are disrupted and reduced in a 90 degree, crossplane V8. 

However, the second order harmonics of a straight eight would seemingly be even less than they are in an I6, what with more cylinders firing at closer intervals and more pistons accelerating at different rates more of the time. Thus I don't understand why anyone would go to the trouble, expense and complexity to create a crossplane design. (Yamaha crossplane I4s do this because uneven power pulses are better for tire grip in motorcycles, not because of smoothness.) I'm clearly missing something here.

I'm also missing something when it comes to understanding the role of V angle in even fire engines. Inline 6 engines have "perfect balance" at any angle, and I'm pretty sure all V12s are flat plane designs. If we basically have two even firing, perfectly balanced I6 engines, I struggle to understand why practically all V12s are at a 60 degree angle. Why not 53 degrees? Why not 107? Why not whatever packages best in your intended application?

They spared no expense designing the smoothest engines for the ultimate luxury cars of the time.

Duesenberg J crankshafts also had two cylinders filled with mercury located near the rear main journals to help dampen vibrations.

30s era V12s were 60 degrees, partly to make them fit in the same engine bay that housed a straight 8. They were sideflow engines with exhaust and intake in the vee which also kept them skinny.

A flat plane straight 8 would fire two cylinders at a time. You would save a tiny bit on crankshaft tooling but lose all the smoothness of 8 cylinders.

Cross plane/flat plane concepts don't apply to an inline engine.

The concept mainly applies to V8 engines specifically, because of a quirk of the mechanics where a smooth running V8 will have an uneven firing order on each bank.  I don't know if any other cylinder count has this effect.

V6 engines are another oddity but only because V6s engine built to the proper bank angle (120 degrees) for an even firing engine with shared crank pins is exceedingly rare - I think only one Italian company did that, in the 1950s.

In reply to Pete. (l33t FS) :

To add to that, I6s have crank throw angles of 120deg and they are balanced.

There used to be a website that had animations of all engine types including some real weirdos, along with explanations of their inherent balance. It was really good but I think it's been gone for years. 

alfadriver said:

In reply to Pete. (l33t FS) :

To add to that, I6s have crank throw angles of 120deg and they are balanced.

The more cylinders you have, I think the more options you have. I think.

Really, any V engine can have an even firing order at any bank angle if you are willing to accept split pins or separate crank throws.  Practically all 60 degree V6s have cranks that look like twisted 4-main straight six cranks.  (Even fire 90 degree cranks, by comparison, merely have 30 degree split pins and no crazy multi-counterweighted monstrosities.  This is why 90 degree V6s tend to be shorter in length than 60 degree V6s)

DaewooOfDeath said:

I'm also missing something when it comes to understanding the role of V angle in even fire engines. Inline 6 engines have "perfect balance" at any angle, and I'm pretty sure all V12s are flat plane designs. If we basically have two even firing, perfectly balanced I6 engines, I struggle to understand why practically all V12s are at a 60 degree angle. Why not 53 degrees? Why not 107? Why not whatever packages best in your intended application?

I had a longish reply for this and didn't have much time this morning due to some very important tasks I had to accomplish.

The key is if you want to share crankpins or not, and this will determine if you have an even firing order or not.  If you look at a V engine as two inline engines joined at the crank, you also need to slot the second bank in so its firing impulses are evenly slotted in with the other bank.  So there are 720 degrees per cycle in a four stroke engine, an inline six therefore has a cycle begin/end/however you want to see it every 120 degrees.  720/6=120.  So if you want to merge a second bank to it, and share crank pins, and have an even firing order, you will need to halve that and that's your bank angle.  60 degrees in this case.  You can make a different bank angle, but it will be either an uneven firing order or it will require split/separate rod throws.

V8s are a very special case because you can take this math, see that there are 180 degrees between cylinders in an even firing four, then divide that in half and get a 90 degree bank angle.  The problem is that you get the horrific shaking forces involved with what becomes essentially a very large four cylinder engine.  So a simple "hack" for this is to shift the center of the crank 90 degrees out of phase.  You still have an even firing order because the 90 degrees cancel each other out with the 90 degree bank angle, but each bank taken individually no longer has an even firing order.  This isn't as much a problem for the intake as you can make a dual plane intake that interlaces the outer pair on one side with the inner pair on the other side, but the exhaust can not have even pulses meeting at the collector unless you get crazy with crossover tubes under the engine.  Or a bundle of snakes over the transmission.  The more you rely on exhaust scavenging to make power the larger this effect is known.

An even firing straight eight has 90 degrees between cylinders, so an even firing straight eight will have throws pointing up, down, left, and right.  This is not a cross plane crank in the sense that the term is used, because there is no way to make an even firing straight eight with all the throws up and down only.

Motorcycle fours with a "cross plane" crank are not even firing.  This is intentional, as there are traction benefits to having an odd firing engine when you are trying to put power down through a very limited contact patch.  This was first noted when Harleys could walk away from much more powerful even firing engines on dirt tracks, and as motorcycle power increased, this advantage began to make itself known even on pavement.

DaewooOfDeath said:

I'm also missing something when it comes to understanding the role of V angle in even fire engines. Inline 6 engines have "perfect balance" at any angle, and I'm pretty sure all V12s are flat plane designs. If we basically have two even firing, perfectly balanced I6 engines, I struggle to understand why practically all V12s are at a 60 degree angle. Why not 53 degrees? Why not 107? Why not whatever packages best in your intended application?

Missed part of this- a V12 doesn't have a flat plane crank, since the bank specific firing order is 120 deg out, so at a min, they have 120deg throws on them.  The V angle can be 60, 120, or 180, and the pins can be shared for the two opposite pistons in the V.  Which is why the Ferrari 512BB isn't a boxer, it's actually a 180deg V12.  Boxers have the opposing pistons moving in the opposite directions from each other.

Fun fact- most V12's in WWII planes used forked rods, so when the two cylinders shared a pin, the pistons were directly across from each other, making the engine a little shorter.  I learned earlier this year that Ford "borrowed" a RR V12 on it's way to Packard, redesigned it, and made it non forked- so it was slightly longer.  It seems that the Ford V12 had some serious potential for aircraft engine, but wasn't asked for so didn't go far.  Where it did go was to shorten to a V8 and get dropped into the Sherman tank.  And where I *really* got shocked was that Ford specialized in casting HUGE engines in Aluminum early in WWII.  Let alone they had a fully developed DOHC system done and in production.  It was another ~30 years until Ford had a production Al block or DOHC 4V engine.  Why did it take so long??  I actually wondered many times if Ford could have copied some of the Euro aluminum engines and/or 4V heads around that time, and we really did.  Other than it making it into passenger hands.  Imagine a 427 Al block with DOHC 4v....  In a GT40.  

oops, sorry about the tangent.

In reply to alfadriver :

IIRC Ford wanted to sell the V12 as a plane engine, but the Navy was 100% committed to air cooled engines and there was no need for/contracts already existed for Army planes.  So they cut off four cylinders and made it into a tank engine.  Which is why it was a 60 degree V8.

At the time it was the largest single aluminum casting in the world.  Later they made some V12 examples for tanks and I think also some boats.

The GAA engine is fascinating for its design.  Its cams were SHAFT driven, with worm gears.  The valve guides were also the guides for the cam followers, so the head castings could be a lot simpler.  Simple machining is cheap machining and fast production.  (See also: the Thompson submachine gun with all of its intricate parts vs. the M3A1 that was a stamped steel clamshell with like two moving parts, three if you count the combination dust cover/safety)

This is the same company that made flathead V8s for passenger cars well into the 1950s!

In reply to Pete. (l33t FS) :

It's really too bad that it wasn't put into a plane that it could have fit into- like the P51 or something.  Or perhaps even a P40 with a big shoehorn.  Given the war was very much an economic war, having a cheaper to make V12 was a big deal, and with a single block casting vs how the RR engine was made- it would have been considerably less expensive to make.  Not sure how accurate the estimate was, but if +20k GAA V8's were made for tanks, then making a less expensive V12 would have been easily possible.

Alas, history didn't turn that way.  Nor did that engine influence anything until after I was born (and that's a pretty loose influence- I dount anyone who worked on the DOHC head nor the AL cast blocks had any real idea of the GAA).  I actually recall that the AL modular DOHC V8 that went into the MKVIII was a big deal for the basic design of a 4V Al V8 in 1994, even though Ford did that 50 years earlier. (just as a reminder, I had never heard of the Ford GAA until 2 years after I retired)

In reply to alfadriver :

GAA stuff

Also, while I'm here, Ford DOHC engine from the 60s

In reply to alfadriver :

Since you brought up the tangent...I saw one of these recently at the American Heritage Museum in Hudson MA.  Their collection of tanks and armored vehicles is pretty impressive, my Dad had been interested in seeing it as he did his service in the 3rd Armored Division in Germany in the late 50's.

In reply to Pete. (l33t FS) :

I knew about the racing motor, I'm talking production motors.  Back in the 40's, Ford made +20k Al v8s.  And it took almost 50 years to replicate those kinds of numbers.

And this is the video I saw https://www.youtube.com/watch?v=6p9eyUAIcX8&t=2s&ab_channel=BrianLohnes

Pete. (l33t FS) said:

DaewooOfDeath said:

I'm also missing something when it comes to understanding the role of V angle in even fire engines. Inline 6 engines have "perfect balance" at any angle, and I'm pretty sure all V12s are flat plane designs. If we basically have two even firing, perfectly balanced I6 engines, I struggle to understand why practically all V12s are at a 60 degree angle. Why not 53 degrees? Why not 107? Why not whatever packages best in your intended application?

I had a longish reply for this and didn't have much time this morning due to some very important tasks I had to accomplish.

The key is if you want to share crankpins or not, and this will determine if you have an even firing order or not.  If you look at a V engine as two inline engines joined at the crank, you also need to slot the second bank in so its firing impulses are evenly slotted in with the other bank.  So there are 720 degrees per cycle in a four stroke engine, an inline six therefore has a cycle begin/end/however you want to see it every 120 degrees.  720/6=120.  So if you want to merge a second bank to it, and share crank pins, and have an even firing order, you will need to halve that and that's your bank angle.  60 degrees in this case.  You can make a different bank angle, but it will be either an uneven firing order or it will require split/separate rod throws.

V8s are a very special case because you can take this math, see that there are 180 degrees between cylinders in an even firing four, then divide that in half and get a 90 degree bank angle.  The problem is that you get the horrific shaking forces involved with what becomes essentially a very large four cylinder engine.  So a simple "hack" for this is to shift the center of the crank 90 degrees out of phase.  You still have an even firing order because the 90 degrees cancel each other out with the 90 degree bank angle, but each bank taken individually no longer has an even firing order.  This isn't as much a problem for the intake as you can make a dual plane intake that interlaces the outer pair on one side with the inner pair on the other side, but the exhaust can not have even pulses meeting at the collector unless you get crazy with crossover tubes under the engine.  Or a bundle of snakes over the transmission.  The more you rely on exhaust scavenging to make power the larger this effect is known.

 

An even firing straight eight has 90 degrees between cylinders, so an even firing straight eight will have throws pointing up, down, left, and right.  This is not a cross plane crank in the sense that the term is used, because there is no way to make an even firing straight eight with all the throws up and down only.

 

Motorcycle fours with a "cross plane" crank are not even firing.  This is intentional, as there are traction benefits to having an odd firing engine when you are trying to put power down through a very limited contact patch.  This was first noted when Harleys could walk away from much more powerful even firing engines on dirt tracks, and as motorcycle power increased, this advantage began to make itself known even on pavement.

This is extremely interesting and informative. Thank you. 

I do have two more questions, though.

1. Taking into account what Alfa said about the difference between 180 degree V12s and boxer 12s, does a boxer layout simply make everything balanced? Ie, a flat four with natural balance, a flat six, a flat 8, etc. I think even the flat twin BMW motorcycle engines are supposed to be very smooth. If my math is correct, a twin and a four would be naturally even firing, but a six would not, and neither would an eight. 

2. How big a deal is even firing for smoothness compared to the balance of the mechanical assembly? The borderline case I'm thinking of here are the VR5/6 engines from VW. I don't think any of them is even firing, but they all use split pins for - I assume - mechanical balance.

In reply to DaewooOfDeath :

I'm no expert, but I'm fairly certain that boxers still have a rocking couple. Think about it when viewed from above...the rotating assemblies are offset because they can't both attach to the crank at exact same spot, so the masses don't offset each other in all three dimensions, resulting in some small vibrations.

In reply to DaewooOfDeath :

Boxer engines make sure that the pistons operate in opposite directions at the same time- which balances them out all by themselves.  With two I6's paired up, you don't need to do that, as the I6 is already balanced out to the 5th order (IIRC)- so you can put them at any angle and they will be balanced as long as each bank is even firing.  With bank angles of 60, 120, or 180deg, the engine is naturally engine can be naturally even firing with shared rod pins- which makes the crank the strongest.

WRT firing order, you can do the same bank angles for V6s, but none of them are perfectly balanced, so it's an acceptable compromise to make a flat 6 a boxer for the better balance.  But it should be pointed out that Porsche is a pretty premium motor- meaning they can put more into making sure their boxer crank is super strong.

For flat 4's, you pretty much have to have a boxer, as either the two alternatives for shared rod pins would be some really bad movement (rocking really bad or shaking really bad).

The Oakland V8 in the 30's was 90 degrees with a flat-plane crank.

To get rid of the shake, they added a "synchronizer" which was an engine mount that was attached to a lever driven by the camshaft.

Every second rotation, then engine would shove itself over a little bit to stop the shake.

In reply to alfadriver :

Aw man, a 180 degree V4 (as opposed to a boxer) would have some really wild corkscrew vibrations as the engine tried to wobble around a vertical axis.  I'm giggling just thinking about it.

In reply to DaewooOfDeath :

1) When not using a forked connecting rod, a flat twin has a small (vertical axis) rocking couple from the cylinder bank offset. However, once expanded to flat 4 and beyond, these individual cylinder pair rocking couples tend to balance each other out.

2) VR5/6 uses pin angle separation to create even firing intervals. While odd fire engines are uncommon in cars, there have been some over the years. Most commonly it seems to come up when manufacturers save money by (re)using a 90* bank angle on non-V8's with the stronger and better mechanically balanced shared crank pin configuration. Notable examples include the early Buick V6, Dodge V10, Audi V10, and BMW V10. The non-performance Ford V10 went with split pins and a balance shaft to be even fire. As far as effect on smoothness, the odd-fire V6 is known as a rough engine, but the V10's certainly aren't. So for a number of intertwined reasons, more cylinders generally helps mitigate the odd-fire roughness.

"Driving 4 Answers" on YouTube does some great videos on engine balance, among other things.

Fun V12 fact. While 60, 120, and 180 bank angles will all produce even firing intervals, 120 will fire 2x cylinders at a time resulting in the same firing interval as an inline 6. Only 60 and 180 will produce the full '12 cylinder' sound.

In reply to Driven5 :

Chevy made millions of 4.3 V6s in the 80s, 90s, and 2000s that were "semi even fire" (semi odd fire?).  They had split pins but not split enough to be completely even firing.  IIRC they were 108/132 firing intervals.  Not sure what the new LT based V6 is, if it is a true even fire engine or not.  It certainly has a balance shaft.

I am not sure about the various GM Busch Grand National engines, but the Ford NASCAR V6 had an interesting development.  Fairly early on they went to shared pins and, by recollection, 120 degrees between crank throws, so the cylinders had a funky 90/150/90/150/90/150 firing interval (as opposed to the oddfire Buick from 1977 and earlier, which was just a V8 with two cylinders missing)

I personally would posit that a theoretical 120 degree V12 with shared pins would be an odd fire engine.  Having two cylinder fire at the same time is the ultimate in uneven firing orders .  On that note, there was a motorcycle recently with an inline four that fired two cylinders at the same time, then 180 degrees later fired the other two, and then 540 degrees of dead time.  (May have been 90/630 but the 540 degrees sticks in my mind) Used an electric motor to smooth it out at low RPM on the street.

In reply to Pete. (l33t FS) :

The Chevy 3.3 and 3.8 were 'semi-even' (odd) fire, but the venerable 4.3 that followed was revised to be genuine even-fire as is the Gen V.

The early Buick V6 also used 120* throws with 90-150 intervals. The oft-parroted claims are that it used heavy flywheels as a means of smooths out the engine. The Jeeps got a much heavier (2x+) flywheel than the cars did. Sure that crazy heavy flywheel helped get the Jeeps rolling under more challenging conditions, but if anything I'd expect them to have wanted to reduce NVH more in the cars than the Jeeps. So I remain unconvinced that this isn't just another urban legend turned repeated until 'fact'. The laws of motion dictate that while a heavier flywheel will smooth the power output (tires) side, it will do so at the expense of added roughness on the power input (engine) side. Unfortunately though, I have not found any 1st hand reports on the felt effects from somebody who has actually back-to-back tested putting a Jeep flywheel in a car or a car flywheel in a Jeep.

In reply to Driven5 :

I would suspect that the vast majority of oddfire Buicks in cars were attached to automatic transmissions.  The ones that had manuals were probably in the cheapest low-end cars available and if they wanted better NVH then they can just buy a V8 car.

(The usual reason why small cars were garbage at GM - "if you want something nice then get a bigger car" - an attitude which helped destroy the US automotive industry and one that GM didn't really corporately understand probably ever.  They had an inkling when they found out that the "5er size at 3er price" Caddy CTS was not very popular because - get this - people who bought 3ers weren't wishing for a 5er but settling for less, they actually WANTED a 3er.  Which is why they came out with the ATS... which wasn't all that much smaller than a CTS, but anyway, I do digress)

In reply to Pete. (l33t FS) :

Can you help me see this?  "for this is to shift the center of the crank 90 degrees out of phase" From (v8): The problem is that you get the horrific shaking forces involved with what becomes essentially a very large four cylinder engine.  So a simple "hack" for this is to shift the center of the crank 90 degrees out of phase.

In reply to NermalSnert (Forum Supporter) :

Briefly, inline four cylinder engines have a certain characteristic shake because in a nutshell all four pistons are at 90TDC at the same time, but because of the connecting rod geometry, this is more than halfway down the bore.  The center of gravity of the reciprocating mass at TDC/BDC is, however, halfway down the bore.  So, the center of gravity of the reciprocating mass is moving up and down twice per revolution, which makes the four cylinder vertical shake that can be reduced/eliminated with counterrotating balance shafts.  The bigger the engine (longer stroke, heavier pistons) the worse it is.

If you made a V8 that used a regular four cylinder crank, you still get this shake, only much worse because now you have two banks doing it.  And very few people bother making V8s in the 2-3 liter range so the engine is necessarily going to be large.

By twisting the middle of the crank 90 degrees, you don't have this concentrated reciprocating mass CG moving up and down like that, so the shake is gone.  Or at least, it gets lost in all the other things.