Engine engineering - another question about displacement

Joined
Mar 12, 2014
Messages
635
Location
Bay Village, Ohio
Got another engine question about displacement (I guess I'm fixated on it?) - I had asked a month or so ago about cylinder size and got great answers. Here is another that just sort of popped into my head:

Say you're a manufacturer designing a V6 engine. Does the engine get designed around the displacement of the engine, or does it begin somewhere else and the displacement is more of an outcome? For example, do the designers/engineers say at the onset "We want to create a 3.5L V6 engine" or does it end up more like "We created an engine, and it turned out to be 3.5L"?

Like I said, just curious, and thanks in advance.
 
Joined
Feb 27, 2009
Messages
8,040
Location
down in the park
From the onset, but they usually design a range. for instance, the4 small 1.6, 1.8, 2.0 4 cylinder engines you find in smaller cars are often versions of the same engine.
 
Joined
Jun 4, 2003
Messages
1,631
Location
98245
Some engine criteria are more important than cylinder count, and come first:
  • Match the power and torque output to the size & mass of the vehicle.
  • Consider the engine duty cycle: is it a typical car rarely asked to produce more than 25% of its rated power? Or will be towing heavy loads continuously producing 75% of its rated power?
  • Based on intended vehicle type, prioritize different aspects of the engine: efficiency vs. power etc.
  • Consider engine size, shape and mass, physically fitting the car and keeping it properly balanced.
  • __ insert other criteria I'm leaving out ___
These more important constraints narrow the cylinder count options: it isn't anywhere from 2 to 12 anymore, but more like 4-6, or 6-8, or > 8. Within these narrower range of choices, other aspects of the engine's intended application (e.g. the car it's going into) will guide you toward the ideal cylinder count. And cylinder configuration: inline, horizontally opposed, V, angle of V, etc.
 
Joined
Mar 2, 2013
Messages
10,350
Location
The Midwest
Mazda accidently made a tiny 1.8L V6, so they put it in the small MX-3. Joking.

Short sound clip and this car was available in the USA despite this video being from Europe:

 

Kestas

Staff member
Joined
Jun 4, 2002
Messages
14,217
Location
The Motor City
Building an engine from scratch is a complicated undertaking. It doesn't happen as easily as updating other parts of the car. Most engines evolve over the decades. For example, the 265 V8 Chevy engine from 1955? evolved into the 283 cid, the 327, until it became the 350. This basic block may still be in use... I haven't kept up.

Basic engines further evolve using different induction designs, turbos, cams, and sometimes longer throw crankshafts (e.g., Ford 302/351).

I don't believe a certain niche market requirement spawns a new engine design. I believe a new engine is designed when the old one is simply too outdated. One thing I've learned in the business is that the core cast block is the last thing the manufacturer will mess with. It's much easier to come up with a new head design... and even then they don't want to mess with the casting shape once it's in production.
 
Joined
Dec 19, 2006
Messages
2,110
Location
Michigan
Talking about the modern V6, pretty much every automaker ended up with the same formula because it simply works well:

60 degree V6
Mid-80mm stroke
Mid-90mm bore

This gives a mildly-oversqaure configuration. It's adaptable and has good power potential without being too peaky. Modern 8+ speed transmissions make them even more adaptable.

Toyota has a torquier 4.0L version of their GR V6 with a longer stroke (longer stroke = more leverage = more torque) used in trucks and SUVs but they're an outlier. Most automakers use the same mid-3L V6 in cars and trucks/SUVs.

It's just a nice sweet spot in engine design.

Big bore means less shrouding of the valves and good flow from the heads. That'll produce good power in non-turbo applications. The short stroke means peak piston speed is low enough that you can throw revs at it to increase peak power. VVT and modern multi-speed transmission make the most of the powerband. Cylinder deactivation tech increases efficiency (along with those transmissions).
 
Joined
Jun 4, 2003
Messages
1,631
Location
98245
Talking about the modern V6, pretty much every automaker ended up with the same formula because it simply works well:

60 degree V6
Mid-80mm stroke
Mid-90mm bore

This gives a mildly-oversqaure configuration. It's adaptable and has good power potential without being too peaky. Modern 8+ speed transmissions make them even more adaptable.
...
We could answer the OP question by example. If the goal is a given total displacement and stroke/bore ratio, any number of cylinders can satisfy it. What constraints lead to picking a particular # of cylinders? For example suppose you want a 3.6 L with 0.9:1 stroke:bore ratio. Approximate cylinder sizes would be the following. I'll make some guesses why some configurations aren't practical.

2 cylinders: 123mm stroke, 136mm bore, 1787cc * 2 = 3.57 liters
These are big cylinders. The long stroke limits RPM range, limiting power output. Probably would also be a strange shape hard to fit into a car. Imagine a giant Harley V-twin. Not an ideal shape, especially in a balanced layout like horizontally opposed.

4 cylinders: 98mm stroke, 108mm bore, 898cc * 4 = 3.59 liters
These are still big cylinders, but getting closer to practical for a car. What's wrong with this configuration - size/shape packaging?

6 cylinders: 85mm stroke, 95mm bore, 602cc * 6 = 3.61 liters
Same as @MrHorspwer described above. Typical car engine.

8 cylinders: 77mm stroke, 86mm bore, 447cc * 8 = 3.58 liters
This seems like a doable engine, though more parts & expense than needed. Perhaps the shorter stroke could enable it to rev higher? But that would be offset by the higher overall mass, and lower efficiency. A tiny block V-8 like this sounds like high fun factor.

12 cylinders: 67mm stroke, 75mm bore, 296cc * 12 = 3.55 liters
This sounds like the 1960s Lamborghini 3.5 liter V-12. Small cylinders, high revving, high power low torque. Complex/expensive/unreliable.

PS: a real-world example of a high displacement engine with low cylinder count. The Lycoming O-360 in my airplane is a 360 ci / 6 liter / 4 cylinder engine with 111mm stroke, 130mm bore. Cylinder shape similar to the 2-cylinder above. Laid out in a horizontally opposed configuration for ideal balance, cooling and packing in an airplane cowl. It redlines at 2700 RPM and typical cruise operation is 2,100 - 2,600.

Why only 4 cylinders in such a big engine? It's lighter, simpler, fewer moving parts, more reliable. Engine speed is limited by the propellor (at 2700 RPM prop tips are near the speed of sound, so you can't spin it faster). So they need the most power they can get at low RPM, that means high torque, might as wall have a long stroke. Fewer cylinders meets all these goals.
 
Last edited:
Joined
Jun 4, 2003
Messages
1,631
Location
98245
Thinking the wrong way. Marketing says “we want an engine that makes X hp and X torque”, then design says “here is the space you have to work with”. Then other groups say “this is the target FE”. Then the power train engineers try and stuff all that in the space they are allowed.
That's the practical approach, similar to my thoughts in post #3.
 
Joined
Mar 4, 2017
Messages
24,677
Location
...
In the Automotive Topics subforum there is a interesting thread on the Subaru Sambar. In this particular version Subaru used a 658cc four cylinder motor, the EN07 or the Clover 4 as it was called. Quite the tiny engine but Japan regulations on engine size are very strict.

Here is a video I found about that Clover 4


 
Joined
Nov 19, 2020
Messages
836
Location
Eastern NC
Good video on this I saw the other day Apparently it turns out more or less 500cc per cylinder is ideal. This is also why undersquare engines are replacing oversquare engines that were more common in recent years.
 
Last edited:
Joined
Feb 27, 2009
Messages
8,040
Location
down in the park
Thinking the wrong way. Marketing says “we want an engine that makes X hp and X torque”, then design says “here is the space you have to work with”. Then other groups say “this is the target FE”. Then the power train engineers try and stuff all that in the space they are allowed.

add to that: "and the intended market penalises above xx CCM with taxation."
 
Joined
Dec 19, 2006
Messages
2,110
Location
Michigan
Good video on this I saw the other day Apparently it turns out more or less 500cc per cylinder is ideal. This is also why undersquare engines are replacing oversquare engines that were more common in recent years.


The video doesn't touch on two things: All those undersquare engines are turbocharged and the quirks of the V6 (it focuses on inline 3,4, and 6 and V 8, 10, and 12).

Why does turbocharging matter? Undersquare engines don't breath as well and it's because of the small bore diameter.

Put on your hot rod hat for a minute: small block Chevrolet 302 from 1969 and small Chevrolet 305 from 1991. Why does the 302 has such massive performance potential and why is the 305 such a dog despite being nearly the same displacement?

302 = 4.0 in bore x 3.0 in stroke
305 = 3.736 in bore x 3.48 in stroke

The small bore 305 shrouds the valves so much. That is, when a valve is open, it's so near the cylinder wall that it disrupts in the incoming air. It has a big metallic wall on one side of the valve, blocking air movement through the engine. That's why you never see a hot rod small block Chevrolet with a small bore: The inherent dimension of the cylinder and it's impediment to airflow limit the power potential of the engine.

I used the 302 and 305 Chevrolet small blocks as an example to keep the same engine family but it applies further. 1990s Camaro vs. Mustang. Camaro was available with 305 (5.0L) and 350 (5.7L) but Ford only had the 302 (5.0L). A Camaro with a 305 was no match for a Mustang, despite near-equal displacement. Why was Ford able to coax more power out of the same displacement? Ford's 302 also has a 4.00 in bore. Chevrolet's 350 has a 4.00 in bore. Not a coincidence on why their performance is more on par.

Bore diameter isn't an absolute and head and combustion chamber design do play a factor, i.e. all engines don't need a 4.00 in bore to make power and two small valves (as in a 4 valve engine) would have less shrouding than one large valve. The idea to take away is that bore diameter has an outsized effect on power, more than just increasing or decreasing displacement.

That is, until you add a turbocharger. Pressurize the intake and the engines (in)ability to ingest air on its own is mitigated. Turbocharging opens up more avenues for engine design because it neutralizes the airflow detriments inherent to some designs.

What about those V6 quirks?

First, understand that horsepower is a calculation of torque x RPM. If you want to increase HP you must increase torque or increase RPM. This is only math:
{\displaystyle P[{\text{hp}}]={\frac {T[{\text{ft}}{\cdot }{\text{lbf}}]\times N[{\text{rpm}}]}{5252}}.}

Now, on to the V6 quirks. All modern V6s use a 60 degree V for packaging purposes. It's narrow and easy to package in a tight engine compartment. Because of this angle and inherent imbalance issues with the V6 configuration, the crankshaft uses a split-pin configuration. That is, each connecting rod has it's own rod journal that's offset by a few degree from the cylinder next to it. This is not an issue for a 90 degree V6, which use a common pin for two cylinders (but still has imbalance issues and they're really wide) and it's obviously not an issue for an inline-6 engine.

To increase power, we can increase torque. We can increase torque by increasing stroke. Think of increasing stroke like having a longer lever on the crankshaft. The longer the breaker bar, the more torque we can apply to the lug nut. Same idea with a crankshaft. One problem: That wonky split-pin crankshaft design that makes a 60-degree V6 work also limits maximum stroke. Darn.

Let's look at the Toyota GR V6 as an example of stretching stroke (still not a long-stroke engine by any means):

1GR-FE - 4.0L = 94mm bore x 95 mm stroke (nearly square) - 270 [email protected],600 rpm, 278 lb [email protected],400 rpm
2GR-FE - 3.5L = 94mm bore x 83 mm stroke (slightly oversquare) - 314 [email protected],200 RPM, 260 lb [email protected],700 rpm

The longer-stroke 4.0L makes more torque but the shorter-stroke 3.5L makes more power. The key is at what RPM the power is made. Using the HP formula from above and a little math, you'll find the 4.0L makes 253 lb ft at peak power (5,600 RPM) and the 3.5 L makes 266 lb ft at peak power (6,200). Despite making a bunch more torque in the mid-4,000 RPM range, the 3.5L extends torque production further into the engine RPM range than a 4.0L.

Why not just design the 4.0L to rev like the 3.5L? You can't. The longer stroke limits maximum RPM. That big lever crankshaft weighs more and creates more force. You just can't swing it as fast.

So, to make power in a 60-degree V6 you need to rev it more. To rev it more you need a shorter stroke. To make it breath properly at higher RPM, you need a large bore to unshroud the valves.

This is how every auto manufacturer who makes a 60 degree V6 ended up with an over-square engine.

The last question is: Why? Why does all this matter?

Marketing. Every automaker needs to make a 300 HP V6 to compete with the next automaker. The 300+ HP 3.5[ish]L 60 degree V6 is like a blockbuster summer action movie: Every movie studio has to have one.
 
Joined
Jun 4, 2003
Messages
1,631
Location
98245
Here's a tradeoff I've often pondered: for a given displacement and stroke/bore ratio, more cylinders means smaller cylinders, which means shorter stroke, which means slower piston speeds at any given RPM, which means you can rev it higher, which means more power potential.

In @MrHorspwer 's example above, make that 4 liter a V8, now you have smaller cylinders with a shorter stroke, you get all that revvability back.

However, more cylinders means more moving parts, higher mass, longer crankshaft, which can limit the RPM potential.

I've seen both high and low revving engines anywhere from 2 to 12 cylinders. So I wonder whether cylinder count really makes much difference in revvability from a pragmatic perspective.
 
Joined
Nov 19, 2020
Messages
836
Location
Eastern NC
The video doesn't touch on two things: All those undersquare engines are turbocharged and the quirks of the V6 (it focuses on inline 3,4, and 6 and V 8, 10, and 12).

Why does turbocharging matter? Undersquare engines don't breath as well and it's because of the small bore diameter.

Put on your hot rod hat for a minute: small block Chevrolet 302 from 1969 and small Chevrolet 305 from 1991. Why does the 302 has such massive performance potential and why is the 305 such a dog despite being nearly the same displacement?

302 = 4.0 in bore x 3.0 in stroke
305 = 3.736 in bore x 3.48 in stroke

The small bore 305 shrouds the valves so much. That is, when a valve is open, it's so near the cylinder wall that it disrupts in the incoming air. It has a big metallic wall on one side of the valve, blocking air movement through the engine. That's why you never see a hot rod small block Chevrolet with a small bore: The inherent dimension of the cylinder and it's impediment to airflow limit the power potential of the engine.

I used the 302 and 305 Chevrolet small blocks as an example to keep the same engine family but it applies further. 1990s Camaro vs. Mustang. Camaro was available with 305 (5.0L) and 350 (5.7L) but Ford only had the 302 (5.0L). A Camaro with a 305 was no match for a Mustang, despite near-equal displacement. Why was Ford able to coax more power out of the same displacement? Ford's 302 also has a 4.00 in bore. Chevrolet's 350 has a 4.00 in bore. Not a coincidence on why their performance is more on par.

Bore diameter isn't an absolute and head and combustion chamber design do play a factor, i.e. all engines don't need a 4.00 in bore to make power and two small valves (as in a 4 valve engine) would have less shrouding than one large valve. The idea to take away is that bore diameter has an outsized effect on power, more than just increasing or decreasing displacement.

That is, until you add a turbocharger. Pressurize the intake and the engines (in)ability to ingest air on its own is mitigated. Turbocharging opens up more avenues for engine design because it neutralizes the airflow detriments inherent to some designs.

What about those V6 quirks?

First, understand that horsepower is a calculation of torque x RPM. If you want to increase HP you must increase torque or increase RPM. This is only math:
{\displaystyle P[{\text{hp}}]={\frac {T[{\text{ft}}{\cdot }{\text{lbf}}]\times N[{\text{rpm}}]}{5252}}.}

Now, on to the V6 quirks. All modern V6s use a 60 degree V for packaging purposes. It's narrow and easy to package in a tight engine compartment. Because of this angle and inherent imbalance issues with the V6 configuration, the crankshaft uses a split-pin configuration. That is, each connecting rod has it's own rod journal that's offset by a few degree from the cylinder next to it. This is not an issue for a 90 degree V6, which use a common pin for two cylinders (but still has imbalance issues and they're really wide) and it's obviously not an issue for an inline-6 engine.

To increase power, we can increase torque. We can increase torque by increasing stroke. Think of increasing stroke like having a longer lever on the crankshaft. The longer the breaker bar, the more torque we can apply to the lug nut. Same idea with a crankshaft. One problem: That wonky split-pin crankshaft design that makes a 60-degree V6 work also limits maximum stroke. Darn.

Let's look at the Toyota GR V6 as an example of stretching stroke (still not a long-stroke engine by any means):

1GR-FE - 4.0L = 94mm bore x 95 mm stroke (nearly square) - 270 [email protected],600 rpm, 278 lb [email protected],400 rpm
2GR-FE - 3.5L = 94mm bore x 83 mm stroke (slightly oversquare) - 314 [email protected],200 RPM, 260 lb [email protected],700 rpm

The longer-stroke 4.0L makes more torque but the shorter-stroke 3.5L makes more power. The key is at what RPM the power is made. Using the HP formula from above and a little math, you'll find the 4.0L makes 253 lb ft at peak power (5,600 RPM) and the 3.5 L makes 266 lb ft at peak power (6,200). Despite making a bunch more torque in the mid-4,000 RPM range, the 3.5L extends torque production further into the engine RPM range than a 4.0L.

Why not just design the 4.0L to rev like the 3.5L? You can't. The longer stroke limits maximum RPM. That big lever crankshaft weighs more and creates more force. You just can't swing it as fast.

So, to make power in a 60-degree V6 you need to rev it more. To rev it more you need a shorter stroke. To make it breath properly at higher RPM, you need a large bore to unshroud the valves.

This is how every auto manufacturer who makes a 60 degree V6 ended up with an over-square engine.

The last question is: Why? Why does all this matter?

Marketing. Every automaker needs to make a 300 HP V6 to compete with the next automaker. The 300+ HP 3.5[ish]L 60 degree V6 is like a blockbuster summer action movie: Every movie studio has to have one.
While I generally agree with you I think it's important to recognize that most engines today are designed with driveability in mind. For your average driver abundant low end torque is more important than high RPM power. Long stroke engines are good for that, even if naturally aspirated. That's why engines such as the Jeep 4.0 were so good at what they did. Low end torque makes cars feel fast even if they actually aren't, which is what will impress most drivers. Also not having to rev the nuts off it to get power is better for fuel economy. Turbos just take a good thing and make it even better.
 
Joined
Dec 19, 2006
Messages
2,110
Location
Michigan
While I generally agree with you I think it's important to recognize that most engines today are designed with driveability in mind. For your average driver abundant low end torque is more important than high RPM power. Long stroke engines are good for that, even if naturally aspirated. That's why engines such as the Jeep 4.0 were so good at what they did. Low end torque makes cars feel fast even if they actually aren't, which is what will impress most drivers. Also not having to rev the nuts off it to get power is better for fuel economy. Turbos just take a good thing and make it even better.

Totally, which is why I added that caveat about marketing and horsepower numbers at the end. Gotta have that HP number that starts with a "3" for your CUV.

My wife's current vehicle is a 2021 Chevrolet Blazer with the 3.6L. My preferred powertrain was the 2.0L turbo that's available because of it's friendly torque curve. At the point we were shopping though, we were lucky to find *any* vehicle, let alone one with the engine option I wanted.
 
Joined
Feb 7, 2013
Messages
761
Location
st louis, mo
Now, on to the V6 quirks. All modern V6s use a 60 degree V for packaging purposes. It's narrow and easy to package in a tight engine compartment. Because of this angle and inherent imbalance issues with the V6 configuration, the crankshaft uses a split-pin configuration. That is, each connecting rod has it's own rod journal that's offset by a few degree from the cylinder next to it. This is not an issue for a 90 degree V6, which use a common pin for two cylinders (but still has imbalance issues and they're really wide) and it's obviously not an issue for an inline-6 engine.

I don't think this is quite correct. Split pins are often used in 90° V6s to get even firing (i.e. Buick 3800). Common pins can be used in 90° V6s but it results in uneven power pulses.

60° V6s cannot use common pins but the offset is great enough that a split pin is not enough and a "flying arm" must be used.
 
Top