# Why are engine blocks so robust apart from containing high pressure?

Lately, I've been pondering why some engine blocks are so bulky, I always thought intuitively it was because they had to last a long time whilst containing thousands of combustion cycles but the more, I look into the reality of engine design that understanding doesn't always hold up to the design of an engine.

It seems cylinder walls have a set thickness which makes pretty good sense, then there is a water-jacket chamber for coolant which again, makes sense but then there is an outer shell of material that is sometimes even thicker than the cylinder walls themselves. This is where I become confused (An example image below)

Surely, this outer shell doesn't need to be as thick as it is. In my mind all it's doing is containing a pressurized coolant which can't amount to more pressure than the combustion of the engine itself. So why is it sometimes as thick as the cylinder wall?

• One approach to answer this uestion would be to look at a FE analysis of the stresses in an engine block and where they come from. I'm sure there are research papers out there. – mart Dec 8 '20 at 13:25
• Not an answer since it's just a guess: if a cylinder does fail in such a way that it explodes, you probably want something pretty bulky to contain it – llama Dec 8 '20 at 18:15
• How much of a factor is the requirement for a large area for the gasket to work over? I.e. is the wall thinner again once you move away from the plane where the block splits? – Lamar Latrell Dec 8 '20 at 19:10
• Are you considering the water jacket/passages through the block or not ? – Criggie Dec 8 '20 at 19:34
• @llama My vote is for mitigating engine shrapnel to the face/body in the event of failure. – MonkeyZeus Dec 8 '20 at 19:35

If you consider only the static forces then indeed the thickness might seem over-engineered. However, engine blocks are not statically loaded. They operate in the range of a few hundred to a few thousand rpm (Revolution Per Minute), so there are dynamic considerations here.

# Fatigue

When materials are subjected to cyclic loading they exhibit a reduction in the allowable stresses. See below for an example:

In general, BCC (body-centered cubic) materials (like steel) show a marked drop in strength (close to 50% or more depending on the steel). For these material there is stress (called the endurance limit), for which the material can be loaded indefinitely.

Also, FCC (face-centered cubic) materials (like aluminium) exhibit a very bad fatigue behaviour.

In any case, when there is high number of cyclic loads (like in the case of motor engine), the allowable stress tends to fall quite a lot. Therefore, there is a need to increase the cross-sections/thicknesses in order to reduce operating loads.

Just for giving an order of magnitude: an engine operating at 4000rpm, performs 1 million revolutions in little over 4 hours of continuous operation.

# Vibrational Considerations

A second reason, has to do with vibrations. More specifically how to avoid having the engine operating in a frequency region that magnifies vibrations.

I'm sure everybody must have noticed at some point that, when an engine drops below certain rpms it starts to vibrate quite considerably. This is because it is in a frequency region near resonance.

In the following image you can see the magnification factor for different damping ratios ($$\zeta$$) and frequency ratios ($$r=\frac{\omega}{\omega_n}$$). $$\omega_n$$ is the natural frequency. If the excitation frequency (i.e. rpm) is close to $$\omega_n$$ then vibrations tend to magnify (see Tacoma Narrows bridge).

Although, it would be best if engines worked at a frequency ratio $$r= {\omega \over \omega_n}$$ which is close to zero, for many reasons its not feasible. So most engines opt to operate in a region close to $$r>1.5$$, so that the magnification factor is closer to 1. (This means that the amplitude of the vibration is approximately equal to the static deflection).

In order to do so they have to make sure that the operation is at $$r>1.5$$, they need to make sure that at the lowest operating rpm of a car (lets assume 750rpm$$\approx 78.5 \frac{rad}{s}$$) is greater than $$\omega_n$$ by at least 50%. (i.e. $$\omega_n$$ should be in the order of 50-55 rad/s).

The (simplified to a SDOF system) equation to calculate the $$\omega_n$$ is $$\omega_n=\sqrt{\frac{k}{m}}$$

where:

• k is the stiffness of the supports (not that this is not the stiffness of the structure)
• m is the mass of the engine block.

So in order to reduce the natural frequency, they need to increase the mass. By decreasing the natural frequency, the frequency ratio becomes higher. Therefore, added mass that can be used for increasing the stiffness of the structure (not the supports), also has the added benefit that it moves the engine operation away from the resonance range.

• Nice write-up. One point: In the very last paragraph, I think you mean 'By decreasing the natural frequency'? – Aganju Dec 8 '20 at 19:13
• Thanks for that. I was in a hurry to reply three looong answers this afternoon, and a few typos creeped in :-) – NMech Dec 8 '20 at 19:15
• what is the BCC, ACC acronyms? – Trevor Boyd Smith Dec 8 '20 at 20:39
• Just a minor nitpick: The Tacoma Narrows bridge collapse was most likely caused by aeroelastic effects (known as flutter in the aviation world) and not by resonance (of the structure itself. – TooTea Dec 9 '20 at 13:21
• I am aware that there is dispute on whether it fell due to resonance, or torsional flutter. I have a very broad definition of resonance. I consider it as a state where the energy absorbed in the structural deformation is greater than the dissipated energy within a circle. In my view, the argument is whether the bridge collapsed due to vertical motion (which is the resonance of an army on a bridge) or a twisting oscillation. My view is that both cases fall under the umbrella of vibrational resonance. – NMech Dec 9 '20 at 14:13

You need to consider that the complete engine block has to withstand the reciprocating forces generated by each of the pistons and con rods moving as well as the rotational forces from the rotating crankshaft.

The 1 litre 4 cylinder engine used in the Hillman imp was known for twisting under load especially once it was tuned as it was an aluminium block. One solution was a 1cm thick duralumin plate machined to fit the bottom of the block between the block and sump, this helped give the required stiffness.

• @RussellMcMahon Some of the tuned ones used Wills rings to be the head gasket around the combustion chamber. That was a cure... I had the Sunbeam Stilleto - the fastback with twin carbs racy cam etc – Solar Mike Dec 11 '20 at 6:52
• now that you mention the Wills rings I recall I did something "clever" to the head or block. Memory so far refuses to provide details but it was based on an idea from elsewhere. It may have been based on relieving areas away from the cylinder edge slightly or ... ? I do recall it seemed like a bit of a desperate measure but the friend who advised me was very competent so I took his advice., At one stage he built a Hillman Imp motorcycle :-) - Engine, wheels, gearbox internals, horn, headlight, shocks, .. more. It looked and rode like a big olde bike. – Russell McMahon Dec 11 '20 at 11:03

A typical consumer engine can be deployed in anything from say -50F to 120F (-45C to 50C), and some blocks will crack when operated or even stored at those extremes. An engine operating at those extremes will experience uneven heating/cooling of the block until it reaches equilibrium, which likely requires adding additional strength in some of the transition areas that would otherwise be fine at 'room' temperature.

The example picture does not show a very solid outter wall; what you see a gasket surface.

A closer look will show a lot of variations in thickness, with a lot of ribbs, bulky parts (for mounting, and vibration damping) but also very thin sections, that are just strong enough to contain the water pressure.

Vibration and fatique are indeed important factors. (even more, since the housing is a cast, so not that strong)

mechanical engineer & diy car mechanic

• Plus, it's 4 cylinders, which are a lot more prone to vibration than 6 cylinder (in-line) or 12 cylinder (v). – Haukinger Dec 9 '20 at 19:14
• Add too that this is a cast block that is then machined. While casting is amazing (and investment casting can do much), if this is a sand-cast part it is sometimes easier to keep it larger- cooling rates, shrinkage, grain structure and stress all go into the casting process and produce variation. – J.Hirsch Dec 14 '20 at 22:15

No one has mentioned yet that there are structural loads placed on the block by other components. The first that comes to mind are the cylinder heads, and as it (they) are torqued, the block needs to resist distorting due to those loads.