Can an old diesel run on anything? What happens when you visit the grocery store instead of the filling station

The most interesting part is that both sides may be partly right.

A diesel engine can indeed run on ordinary rapeseed, sunflower or other vegetable oil. The question is not so much whether the engine will start after oil is poured into the tank, but what happens over several thousand or several tens of thousands of kilometres.

Why does cooking oil burn in an engine at all?

In a diesel engine, the fuel mixture is not ignited by a spark plug. Air compressed heavily in the cylinder heats up, and the fuel injected into it ignites. Vegetable oil is also an energy-dense organic substance and can be burned in a compression-ignition engine.

There is therefore nothing extraordinary about what is seen in the videos. The use of vegetable oils in diesel engines has been studied for decades, and they can certainly be used to get an engine running.

The problem is that the physical properties of vegetable oil differ significantly from those of automotive diesel fuel. According to an overview by the US Department of Energy’s NREL, the main issues with unprocessed vegetable oil are its much higher viscosity and higher boiling point than diesel fuel. Long-term use has been associated in studies with carbon deposits in the engine and vegetable oil accumulating in the engine lubricating oil, potentially reducing engine life.

The engine runs. That does not mean it likes it

Fuel must pass through the filter, pump and high-pressure section, while the injector must form the finest possible fuel mist in the cylinder. Vegetable oil is much more viscous than diesel fuel, making it harder to atomise properly.

Poorer atomisation means larger fuel droplets and less complete combustion. In long-term tests, vegetable oil use has been linked to injector coking, combustion-chamber deposits, sticking piston rings and lubricating-oil contamination. It is precisely because of the long-term effects that tests lasting a few hours or a few tankfuls can produce deceptively good results.

Facebook comments, however, include many people who say they have tried it themselves.

One commenter, for example, claims to have driven a 2007 Audi A6 2.0 TDI on vegetable oil for two years. The comment does not reveal the precise composition of the blend used or the engine’s subsequent technical condition. Another writes that cooking oil mixed 50:50 with diesel fuel works excellently in an old engine.

These experiences may be entirely genuine. However, one car running is not an endurance test and does not prove that wear remained the same compared with conventional diesel fuel.

There is a major difference between an old diesel and common rail

Comments repeatedly distinguish between old diesels and common rail and pump-injector engines. An old diesel tolerates cooking oil better, while it should not be poured into a common rail or PD engine. This is a sensible warning, but not an absolute technical rule.

Older diesels with mechanical injection pumps, especially indirect-injection units, may tolerate changes in fuel properties better. Long-term tests of newer direct-injection engines have found more injector coking, piston problems and lubrication issues with vegetable oil. Reviews of scientific literature have found that durability issues tend to be more serious in direct-injection engines than in older indirect-injection designs.

At the same time, it is not true that a common rail engine cannot run on vegetable oil at all. It has been done experimentally.

In one rapeseed-oil test carried out with a common rail engine, the oil was heated to 55 degrees and the injection pressure and timing were altered. The study’s authors stressed that rapeseed oil’s high viscosity and surface tension can cause incomplete combustion and deposits on injectors, valves and in the combustion chamber in an unmodified engine. Possible long-term problems cited also included sticking piston rings and dilution of the lubricating oil.

So the question is not whether a modern diesel will start on vegetable oil. The question is what its fuel system was actually designed for and what happens during long-term use.

Store-bought rapeseed oil does not meet the automotive diesel standard

Another important distinction should be made here. The standard in force locally sets the requirements for automotive diesel fuel. It covers diesel fuel containing up to seven per cent by volume of FAME biodiesel component. Rapeseed oil sold in shops is not FAME and therefore does not meet the automotive diesel standard.

Nor does this seven per cent limit mean diesel fuel may contain seven per cent raw rapeseed oil. FAME is biodiesel produced from vegetable oil or other fat through chemical processing.

But what if cooking oil is mixed 50:50 with diesel fuel?

This is one of the most common solutions in the comments: for example, 10, 20 or 50 per cent cooking oil, with the remainder diesel fuel.

Blending does reduce the mixture’s viscosity, and engine tests with small quantities of vegetable oil have produced fairly good results. However, there is no universal percentage below which cooking oil can be declared safe in every diesel engine.

Study results depend on the type of oil, engine, injection system, temperature and load. One larger review of scientific literature considered blends of up to 20 per cent promising in several tests, but exhaust-emissions and smoke-opacity results varied between studies.

NREL also points out that carbon-deposit formation increases as the share of unprocessed vegetable oil in the fuel rises. Therefore, even a small blending percentage does not automatically make cooking oil a fuel that meets the vehicle manufacturer’s requirements.

Try the same thing in winter

What happens at −20 degrees? In cold weather, vegetable oil’s already high viscosity rises further, making it harder for fuel to pass through the filter and pump. Studies have used oil preheating specifically to reduce viscosity and improve atomisation. In one rapeseed-oil test, heating the oil to 60 degrees reduced its viscosity enough for fuel flow through the filter to become more normal.

This also explains why a warm engine running in a summer video gives little indication of how the same solution would behave during a January cold start.

Free French-fry oil from a restaurant?

Social media offers an even more tempting option: take used frying oil from a restaurant, run it through a 20-micron filter and pour it into an old diesel.

Such solutions have genuinely been used. Filtering removes solid particles from the oil but does not change its fundamental fuel properties. The oil remains vegetable oil, its viscosity does not become that of diesel fuel, and filtering does not turn it into biodiesel.

The US Department of Energy’s Alternative Fuels Data Center states directly that unprocessed or refined vegetable oil and unprocessed used cooking oil are not biodiesel and are not recommended for use as vehicle fuel in that form. Among the reasons cited are long-term deposits, sticking piston rings and lubricating-oil-related problems.

One Facebook commenter, meanwhile, describes how what he calls “fungus” developed in his car’s tank, requiring the tank to be washed out and four or five fuel filters to be replaced.

Microbial contamination of fuel tanks is a real phenomenon and can indeed clog filters, but it usually requires free water in the tank. Microbial contamination can also occur with conventional diesel fuel. It is therefore impossible to say from a single Facebook comment that cooking oil specifically caused microbial growth in that car.

Cooking oil is not biodiesel

The two are constantly confused in online arguments. Ordinary cooking oil consists mainly of triglycerides. In biodiesel production, oil or fat is processed by transesterification, usually using alcohol and a catalyst. The result is fatty acid esters, usually FAME, with glycerol as a by-product.

Chemical processing changes the fuel’s properties considerably. Store-bought rapeseed oil, used frying oil and specification-compliant biodiesel fuel therefore cannot be treated as equivalent.

The advice also seen in comments that “oil plus alcohol and biodiesel is ready” greatly oversimplifies the process.

Can cooking oil improve the smoke reading at an inspection?

Another old piece of garage wisdom appears in the comments: pour vegetable oil into the tank before the vehicle inspection and the diesel will smoke less.

One Transit owner claims his car’s smoke-opacity reading fell from 4 to 1.5 with vegetable oil. This specific result cannot be verified, and the comment does not disclose the measurement conditions or whether the figures are comparable.

The idea itself is not made up. In one direct-injection diesel-engine test, smoke opacity at full load was 27–35 per cent lower with pure rapeseed oil than with diesel fuel. At low load, however, whitish smoke occurred, while other studies have produced differing results for soot and particulate matter.

This means vegetable oil may indeed reduce the smoke-opacity reading in some engines and under certain operating conditions. It cannot, however, be concluded from this that the exhaust emissions as a whole become cleaner.

Diesel smoke testing measures the radiation absorption coefficient K, in units of m⁻¹. This measurement indicates the exhaust gas’s optical smoke opacity, not its entire chemical composition.

And then excise duty comes into play

There is another nuance to driving on cooking oil that has nothing to do with engine durability.

The Alcohol, Tobacco, Fuel and Electricity Excise Duty Act treats, among other things, products used as fuel whose first four digits of the CN code fall between 1507 and 1518 as biofuel. Vegetable oils belong to this commodity group.

Section 66(18) of the same Act states that the excise-duty rate for biofuel is the same as for diesel fuel when used for the same purpose as diesel fuel. From 1 May 2026, the local diesel-fuel excise-duty rate is €459 per 1,000 litres.

In other words, rapeseed oil bought as food from a supermarket does not become tax-free vehicle fuel simply because it was sold in a bottle rather than from a pump.

The most deceptive thing is that the car drives

A person can pour cooking oil into the tank, start the engine and drive away. Some old diesels may tolerate it surprisingly well for a long time. That is exactly what makes the experiment deceptive.

A fuel does not have to be suitable for an engine just because the engine runs on it today. Injector coking, combustion-chamber deposits, piston-ring problems and lubricating-oil contamination may not appear within the first 10, 100 or even 1,000 kilometres. Technical literature makes a fairly clear distinction here between short-term operation and long-term durability.

The most interesting video would therefore not be one in which a man pours 10 litres of cooking oil into the tank and starts the engine.

It would be far more interesting to see the same car 50,000 kilometres later. The injectors could then be removed, compression checked, engine oil analysed and the combustion chamber inspected.

Only then could you say how cheap that stop at the grocery store really was.