Devices of this type include: Ecotek, Tornado, Hiclone, Powerjet USA, SpiralMax, Turbonator, Vortex Valve

Turbulence of the air in the combustion chamber is vital to the operation of all modern petrol and diesel engines. This is not in any way a new discovery; the legendary pioneer Sir Harry Ricardo produced the first "turbulent" cylinder head design in 1919. The main effect of the turbulence is to speed up the burning of the fuel / air mixture (or to promote mixing of fuel and air in the case of diesel engines). A petrol vapour / air mixture actually burns quite slowly, in the order of a few metres per second. So a flame starting at a central spark plug would take about 10 milliseconds to reach the edge of the combustion chamber. That doesn't sound long, but at 6000 rpm the engine would have completed an entire revolution in that time! Burning needs to occur within about a tenth of a revolution to achieve reasonable efficiency. What Sir Harry realised is that, if the air in the combustion chamber is moving rapidly, it will "stir up" the flame and help it to propagate much faster. This page concentrates on the effect of swirl on combustion in petrol engines.

So, how is this turbulence achieved? Unfortunately turbulence is a very short-lived phenomenon; turbulence is generated whenever air flows quickly past a stationary surface, but rapidly decays away through viscosity once the bulk air speed reduces. So modern thinking is to use careful design of the engine's inlet ports. By aiming the intake flow correctly, rapid air motion is set up during the induction stroke. This rapid motion breaks down into turbulence as the piston rises on the compression stroke, and if the engine is correctly designed, hits just the right level at the point of ignition. This "swirl" technology has been standard on engines for decades and is well understood. During my career I have worked on many such engines, and have directly measured both the in-cylinder air motion and the fuel economy effects.

Low swirl engine: Inlet port designed for high flow, no strong direction to air jet. Very little air motion in the cylinder at the point of ignition

High swirl engine: Inlet port designed to produce very directional air jet. Strong swirling motion set up in the cylinder

The amount of turbulence an engine should have is (as always!) a compromise. The number swirl ratio is used to characterise the level of swirl, where 0.3 would represent quite low swirl and 1.5 pretty high, for a petrol engine. Since generating more swirl requires more restrictive inlet ports, values around 0.5 to 1.0 are usually found in production engines. Adding more swirl speeds up the burn, less swirl slows it down.

Most "fuel saving devices" that claim to speed up the burn say that this improves fuel economy. To some extent that's right, but only at levels lower than found in most production engines. A very slow burn gives bad economy because the fuel is still burning when the exhaust valve opens! Theoretically the best efficiency would be obtained by an instananeous burn, but this would produce an extremely high in-cylinder temperature and so the heat loss to the cylinder walls would be much higher. The overall effect is something like this:
The optimum is typically around 1.5 to 2.0, but the fuel economy loss from dropping to around 0.5 is very small and gives much better "breathing", so improving power. This is the kind of swirl level that most modern engines operate with. I spent many years working on engines with various swirl ratios, and indeed where the swirl ratio could be altered while the engine was running (high at part load, low at full load), but in the end the higher turbulence levels simply did not give enough benefit to justify the cost and/or reduced performance.

Some engines do employ variable-swirl technology, such as Vauxhall (Opel)'s new Twinport engine, and some Fords. Partly this is because there is a slight economy benefit, but mainly because it allows use of high levels of valve overlap or exhaust gas recirculation while still giving a stable burn. Normally high overlap or EGR leads to rough engine running; adding turbulence increases the engine's tolerance to overlap or EGR, which bring their own benefits.

How does this relate to bolt-on "fuel saving devices"? The first and most obvious point is that even an engine designed from scratch to have high turbulence is only 5%, at the very most, more economical than an engine with "normal" turbulence.

As an example, the new 3.5-litre Mercedes V6 (in the SLK 350) contains a variable turbulence system. "The intake ducts are equipped with newly developed tumble flaps which improve the intake process and combustion of the fuel-air mixture. These pivot open under partial load, increasing the turbulence of the gas flow in the combustion chambers. Under higher engine loads the tumble flaps are completely recessed into the intake manifold. Thanks to the use of these innovative tumble flaps, the fuel consumption of the V6 engine is reduced by up to 0.2 litres per 100 kilometres depending on engine speed." Obviously any fuel saving is welcome, and in conjunction with other technologies makes this a relatively "green" engine. But 0.2 litres per 100 km is less than a 2% gain - and this on a system specifically designed and optimised for this particular engine, where the fuel and ignition can be adjusted to suit, and where the turbulence can be switched on or off as required. Equally, this is a system designed by Mercedes, a company with vast resources and a reputation for thorough engineering. If the very best fuel economy improvement Mercedes can get through turbulence is less than 2%, then you should be suspicious when a small company claims 5 - 10 times that benefit on pretty much any engine through their "bolt-on" device.

The second and more fundamental point is this. Turbulence speeds up the burn, which is generally a good thing. But the ignition must be retarded by a corresponding amount, or the burn will occur too early. Ignition advance is carefully mapped, based on the engine's burning speed, to give the burn peak at exactly the right point in the piston's stroke (generally just after top dead centre). Speeding up the burn requires the ignition to be retarded by anything up to 20 degrees to maintain optimum timing. Yet no "fuel saving devices" based on turbulence suggest making changes to the ignition timing.

Some devices even claim an increase in power, though it is well understood that the optimum swirl level for best power is less than found in most modern engines. Speeding up the burn at high load without retarding the ignition leads to much higher in-cylinder pressure, because the burn is too early. Knock, and indeed engine damage, is the most likely result.

But let's ignore all that for a moment and imagine that adding more turbulence does help fuel economy. Can a "bolt-on" device really increase the turbulence in the cylinder? The simple answer is no - the flow in the intake system is already highly turbulent at part load, with air speed past the throttle in the region of 200-300 metres/second. A device upstream of the throttle may make the air swirl there, but the swirl pattern will be destroyed as the flow squeezes past the throttle blade. Injection of "a small amount of turbulent air" into the inlet manifold, as some devices claim, will have even less effect. It is interesting that the makers of such devices state with confidence that the device increases turbulence, but do not have any direct measurements of turbulence to show that this actually is the case. How, then, do they know?

But don't just take my word for it - there have been many studies, both direct flow visualisation and computer simulations, showing turbulence in the intake system. For example, Rai Alsemgeest at Warwick University did a study of intake manifold flow for Jaguar; see his presentation for full details. Here's his animation showing a section through the intake manifold, where different colours represent different air speeds - red being the fastest and blue the slowest. Air enters from the left of the picture; the white shape in the middle of the flow is the partially open throttle blade:

It's very obvious how incredibly turbulent the air flow is. There's no way that a swirling air flow pattern set up before the throttle could survive and affect the burn within the cylinder, nor that a small additional air flow into the mainfold from a "bolt-on device" could make any significant difference. At wide-open throttle the air flow is much smoother, but from a fuel economy perspective this is unimportant as the engine spends 99% of its time at part throttle.

For more information on one of these devices, see the Ecotek CB-26P Case Study.

So in summary:

• Engines already have high levels of turbulence, and the physics is well understood
• Adding more turbulence can give only a tiny fuel economy benefit - this is proved by experiment
• Ignition must be adjusted to suit the faster burn, or the effect will be worse economy
• Increased turbulence at full load will most likely damage the engine unless the ignition is retarded
• Anything in the inlet manifold is extremely unlikely to affect in-cylinder air motion anyway

Also note that a turbulence-increasing device fitted upstream of the air meter may cause the meter to mis-read (they are intended to measure "smooth" flow) and so produce effects similar to manifold air bleed devices.

Please also read the general comments on fuel "saving" devices, if you have not done so already.

If you found this page helpful, you may like to support my work. If you think I have made a mistake, or am talking complete nonsense, please take a moment to read the Response to Critics before saying so in public (on discussion Forums and the like).