dabble journal

Dirty machine

Feb 20, 2020

It is 2020 and I thought, may be this is the year I’ll buy my first car. Living in the United States has been at least partially challenging without a car. I’ve managed to go without one for over 5 years now and I’d go without one if the benefits outweigh the hassle and of course if I manage to find pooled rides to the beautiful outdoors of the west coast without owning a car.

Like any sane person would, I’ve been wondering about electric cars and I started exploring. And all points considered, everyone I’ve spoken to (expect one of my best friends) thinks an electric car is the wiser, better and a fun choice.

However, what I could not fathom is the counter-force I see against electrics by the same auto makers who stand to make money off of it. Buying a combustion engine car today is a wasteful affair and it doesn’t take calculative brain to figure that out. It is a massively depreciating asset and will be a large sunk cost very quickly. Or so I thought.

The case for electric vs gas cars is much debated and weighing pros and cons has been written widely about - obviously each side conveniently ignoring aspects that need consideration.

Let’s drop the ecosystem altogether. Not to ignore that aspect, but rather to just view one aspect with the intent to differentiate machines qualitatively. Engineers always strive to (or in my opinion, always should) strike a balance between complexity and utility. As utility increases over time, complexity is appended to existing designs. This kind of overloading does cause a machine to deviate from it’s initial simplistic design.

I started looking into something that was staring me in the face for most of my childhoood. Why are petrol engines so complicated and need that much maintenance?. Design plays such an important role in engineering. The most important one in fact. Design problems cannot be retrofixed by building complex systems around a core that is inherently flawed. This does not improve anything, it just adds to the complexity. As an engineer, I wouldn’t start implementing something unless I have a solid design. The time traded for design is always worth it.

And so, I began looking into the design aspects of the internal combustion engine.

The complex dirty machine

Internal combustion engines are dirty and complex. As with any machine that supports “combustion”, the process of combusting something is inherently residual. This makes the combusting machine deal with the by products of combustion, which cannot be go well in most cases. Especially, when what is burnt is carbon based fuel. Carbon by products are an essential residue.

With the differentiation of very high compression ratio vs having a spark plug, the operating mechanism apply to both petrol and diesel engines.

Burning gasoline involves vaporizing it. This involves a carburator or in the modern engines, a fuel injector (only 50% of american cars use fuel injectors). An electromechanically timed value pressurizes small amounts of fuel pulled through a tube, filters it from gasoline debris. Simultaneously, a gasp of air is also filtered. Filtered petrol is jetted and atomized into a narrow venturi tube where air meets it half way. Now, we hope they have air and petrol have mixed. The amount of air to petrol, the mixture ratio, as one can imagine is pretty complicated to control. For a good mixture, the engine needs to be hot and running.

As one can imagine, when engines are cold, starting will suffer due to this. A choke is used to make the mixture fuel rich, as it makes it more combustible, also making it more dirty because of higher amounts of combustion by products.

In a fuel injector, the concept is similar, but air mixing happens in the combustion cylinder rather than outside. In a fuel injector, a timed solenoid piston pushes variable amounts of fuel (varied by load, speed etc.) on to the top of the piston or guided towards the spark plug by air flow. This needs such precise timing that ideal design conditions are almost never reached. Fuel injectors also suffer with reactance to rapidly changing loads on the engine. The lag causes excess fuel in cylinder, resulting in partial combustion. In all form of internal combustion engines, partial combustion is never fully solved. It exists and so does the resulting high emissions.

Sparks fly inside the combustion cylinder, timed and orderly. Except when they don’t. Spark plugs are consumables and are expected to be replaced periodically. Part of why this is so, is because of partial combustion that deposits soot on the spark plug. Oil debris deposits are yet another leading problem, we’ll get to that in a bit. Overheating also causes spark plugs to not operate as expected. They can develop gap deformities or body cracks. Eletrical sparks are generated by maintaining a very high voltage between two conductors until the medium between them breaks down. In this case, the medium is air, and it needs upto 25kV to breakdown. This is why spark plugs use ceramic insulation between conductors. High temperatures caused by prolonged initial high fuel mixtures result in spark plug damage.

The worst part is, it takes a while for spark plug problems to be obvious enough. Observable symtoms like tardy acceleration, bad fuel mileage and bad idling are already end point signs. At that point you’ve probably generated enough soot, high temperature and heavy emissions to overload other parts of the engine and exhaust system.

Pistons need to seal the engine cylinder while translating back and forth. The piston rings scrape against the walls of the cylinder. This is by design, and that is simply crazy! By design, the piston rings stack is expected to leave a super thin film of lube oil on the cylinder wall to reduce that friction. This thin film is combusted. Obviously, this is side-effect combustion and is unclean and is one of the main sources of dirty exhaust.

Since friction is expected by design, despite lubrication, wear of piston rings slowly causes combustion chamber seals to leak. Engine degradation starts from day one. Some might argue that the first few hundred miles of usage actually polishes the cylinder walls thus reducing friction. But, one or the other material is wearing. Where does that debris go? Piston rings possibly trap that. Combustion chamber is leaky long before an engine shows observable and easily detectable signs of bad performance or running smoky.

Lubrication is an absolute essential for an internal combustion engine due to the expected friction of pistons. They all normally carry a sump of oil which is sprayed on to the walls of the cylinder on the dead side. The crankshaft used to translate linear to rotational motion is swimming in oil. Since a thin film is continuously burnt and residue laden walls scraped down, the oil gets dirtier by mile. Then comes your periodic oil change. So does filter replacement and exhaust clearing.

Heating gets no points when the objective of an engine is to transform energy in fuel to rotational motion translated to wheels. It is infact one of the biggest inefficiencies of internal combustion engines. Most of the heat is spewed out of exhaust as hot residual gases, heating the exhaust system on its way out. Heat is dissipated into the engine walls and all mounts that hold it. Passenger cabin needs heat shielding from engines. Sure, some heat is tapped in most cars to heat the cabin, this is minimal and not a side effect to seek. All heating can be categorized as gross loss of efficiency.

Let’s see, are we done yet? Nope, now you have to implement a radiator and a coolant system to actively cool the engine and subsystems down. In fact the ideal temperature for the exhaust to be at is 0 Kelvin. Thermodynamic efficiency is inversely proportional to the sink temperature. This grossly puts maximum theoretical efficiency at arount 50%.

Given so many design constraints and flaws, we can expect inefficiencies in internal combustion engines. How much? Petrol engines are on an average about 20% efficient, i.e., they only convert that 20% of energy store in fuel to usable mechanical energy. Diesel engines due to higher compression ratios, carry 30% efficiency - but they come with their own caveats of complex designs, vibrations and being terrible with exhausts.

Moreover, functional testing of ICE demonstrates more issues. Delay in acceleration response due to the number of mechanical components a control signal needs to propagte through. Fuel injector increases the quantity of fuel injected, to which the stokes respond by increasing burnt volume. This increases strokes rate over many subsequent strokes to finally reach the target speed. The other issue is changing loads. Load change causes the engine to respond by altering combustion properties - dropping load results in incomplete combustion.

How did the auto industry choose something so grossly mal-designed and hailed it as the invention of the 21st century? It just blows my mind. Yes, I will give in some lenience for the fact that they were more portable for the distances they took you for the amount of gas you could carry in a car. But, design and engineering wise, they are a stop-gap solution at best. Not an invention on which we should’ve run a century.

During the same time petrol engine investment was the way of the world, the electric motors were ignored and overrun. The vastly underwhelming usage in single user transportation and underappreciated awesomeness of electric motors is incomprehensible. A generation of engineers sat on them, while finding no investment in making their use viable in portable single user transportation. This might be one of humanity’s biggest slip given where we stand on environmental impacts and sunk costs of internal combustion engines.