Fuel Economy Tip – Avoid Engine Braking

Today’s tip is yet another easy way to save money and wear and tear on your transmission.

Avoid engine breaking.

Essentially, engine braking is when you down shift the transmission in order to slow down your vehicle. For example, you’re going down a road in fifth gear, you see a red light ahead and instead of slowing your car by using the breaks, you down shift into third gear.

While down shifting is an easy way to take it easy on your brakes, it’s also a great way to needlessly waste gas.

Anytime your engine revs and the RPMs increase (which is what happens when you down shift), you will use more gas than you would at the same speed but in a higher gear (lower RPMs).

So, instead of using your engine to brake go ahead and ease on to your brake pedal and smoothly come to a stop.


  1. Utter nonsense. All modern fuel injected, drive by wire throttle cars have such a thing as throttle position sensor. This tells the Electronic Control Unit whether to and how much fuel to inject. Ignition probably carries on anyway.

    Even early fuel injected (single injector, in the intake manifold/plenum) engines that had the throttle actuated directly by the driver, through a cable, didn’t use any fuel if you downshifted without using the throttle. As they had, wait for it, a throttle position sensor as well.

    Then again, I always blip the throttle to help the clutch and smooth the downshift. Even so, the amount of fuel burnt is less than dropping to idle and using the brakes. And there’s the added bonus that you can start accelerating right away, should the light turn green, due to being in the appropriate gear from the downshift.

    Your advice is at most valid for carburetted cars.

    In the future kindly don’t presume to dispense advice on what you don’t have a good enough grasp of yourself.

    • Glad to see someone has their head screwed on right.

    • hey ive heard that elsewhere too. so in a modern car, does the computer tell it to give it NO gas, even though it’s in a high rpm? or does it tell it to give only the small default amount per stroke of the engine?

      also, you said it burns less fuel when blipping than it does to drop to idle. how does the engine use extra gas when shifting to neutral?

      • When engine braking the car’s motion gets transferred through the wheels, drive shaft, transmission, and ultimately drives the engine. The car’s forward momentum keeps the engine’s RPMs high enough that it doesn’t need gas to keep the car’s systems alive and running. Once the RPMs get low enough, the engine starts injecting fuel back into the cylinders.

        When shifted into neutral, the transmission and engine are disengaged. Thus, the engine HAS to put gas into the cylinders to keep running.

    • There is also a mass air flow sensor. Higher RPMs = more air flow = more fuel injected to keep the right air fuel mixture.

      • Most ECUs now run a system called Deceleration Fuel Cut Off. This does what it says on the tin, when you are off throttle and the TPS (Throttle Position Sensor) is below a pre-defined value, the fueling events are cut until i) the throttle is reapplied or the vehicle enteres it’s Idle Strategy.

        Great way of aiding the retardation process and saving fuel but needs to be setup/configured correctly.

    • Theodore Moody says:

      Thumbs up to the clear and concise rebutle

    • franz doreza says:

      I thought the same as I downshift on my BMW in sport mode to slow down before braking to stop completely, it doesn’t make sense that it would use up any more fuel than the usual method..thanks for that..I hope this is true for a 2008 335XI.

  2. Also, the people who liken an engine to an air compressor likewise don’t know what they’re talking about.

    A car engine is actually a self-powered air pump. It’s only somewhat a compressor if you restrict its exhaust. It’s the pumping losses (energy lost through moving air from one place to another – through the intake, engine and exhaust) and various kinds of internal friction that provide the retarding force.

    And engine braking is easier on the conrods, rod bearings, crank, etc. Because there is no combustion, the most demanding stroke of the cycle.

    Also, the main bearings, conrod bearings, etc. don’t care that the momentum is provided by the starter, the flywheel, the road or some other cylinder firing. Except for cylinders which ought to be firing, the forces occur practically the same, due to the flywheel inertia. However, the forces developed on the cylinder head, block, piston, rings, wrist pin, bearings, crank, main bearings, etc. by a cylinder firing are higher than those due to the intake, compression and exhaust strokes.

    Which ought to be obvious, since the energy for ther othe strokes comes from the power stroke. Which still has some mechanical energy left over to move the vehicle.

    In the meantime, oil is still being pumped through the engine. Because the oil pump likewise doesn’t care whether its internal combustion, flywheel inertia or some external force that is driving the engine.

    So, basically, except for rings and maybe piston skirts, the wear arguments are also bollocks.

    One more thing, in throttled engines throttling losses also contribute to engine braking. Because you’re basically making the engine pump air against a pressure gradient by restricting it’s air intake.

    Please, educate yourselves before presuming to dispense advice or technical opinions on an advanced topic.

  3. Yet another addendum. When the rpm nears idle, even as you’re engine braking, the idle computer or idle control function of the ECU takes over and starts injecting fuel again, regardless of the throttle position sensor.

  4. @SonyAD – I couldn’t agree more. Your arguments about the forces being less during engine braking (due to lack of a power stroke) make sense. I actually never thought about it before today, but I’m convinced you’re correct.

    As for fuel flow, your description pretty much sums up how my car behaves in my experience.

    I’m actually a mechanical engineering student at the University of Malta, and we had put an engine (1.4 litre petrol, 60kW / 80 bhp) on a dynamometer capable of both driving the engine and being driven by it. We basically made a few tests and through the measurements of rpm, torque, power input, MAP etc. we found out that the main contributing factor for engine braking was friction (about 7kW) followed by pumping losses (about 5kW).

    Thanks for posting!

  5. Thank you.

    As an automotive engineering student I think you’ll also enjoy reading about engines with offset cylinder bores. These are engines which have their cylinder bores offset laterally relative to the crank.

    This is done in an attempt to get the moment of peak cylinder pressure nearer to when the conrod describes a tangent to the circle described by the crankpin journal, the best time to transform linear force from the conrod to torque of the crankshaft. So that more of the force from combustion is converted into useful work instead of friction and heat loss into the walls of the combustion chamber and the piston head (because allowing the combustion gasses to expand sooner after peak cylinder pressure also reduces heat loss into the engine).

    There is also the added benefit of lessening piston side thrust into the cylinder wall during the power stroke, which is greater than the opposite side thrust during the exhaust and even compression stroke when the piston receives energy from the crank instead of viceversa. It’s the power stroke thrust side that wears fastest, in both the cylinder sleeve or block and the piston.

    Laterally offsetting the cylinder bores relative to the crank makes it such that the angle of incidence, even at its maximum, between the line described by the conrod and the surface of the cylinder bore is less, thus reducing side thrust (addition of forces).

    Another consequence is that the induction and combustion strokes last for more than 180° of crank rotation while the compression and exhaust strokes last for less than 180° of crank rotation.

    This is both beneficial and detrimental. It is beneficial because it allows more time to fill the cylinder with air-fuel mixture (or air) and more time for the power stroke to extract useful energy from the expansion of combustion gasses inside the cylinder. Also, there is less time for compression to leak past the rings during the compression stroke though there is more time for blowby to occur during the power stroke.

    The big problem with all this is engine balance is even poorer than in a normal engine and the efficiency gain is not really significant. So it’s rarely used, only in small displacement engines such as the Honda Insight 1 litre and the Toyota Aygo 1 litre, though I remember reading about this technology being used in some aircraft engines during the 2nd world war.

    Here’s a piston position and speed graph I drew up when I came up with the idea before having researched existing developments.


  6. Here’s a piece on the Honda Insight 1 litre:

  7. @SonyAD – that was a pretty interesting piece of engineering! Sounds like someone’s doing a dissertation on offset cylinder bores… 🙂

    Another type of engine design that you might find interesting is the Atkinson cycle which replaces the connecting rod and crank with a somewhat more complicated mechanism which makes the power stroke longer and thus increases the efficiency. It’s mainly used on the toyota prius hybrid because the engine has lower power-per-displacement than comparable Otto cycle engines.

    The wikipedia article has a pretty good description: http://en.wikipedia.org/wiki/Atkinson_cycle

    Here’s an animated diagram of the physical layout: http://www.animatedengines.com/atkinson.shtml

  8. Censoring comments, nice. I’ll be sure not to visit here again.

  9. Sorry, Carr. My fault. Nevermind.

    LMF, there are a plethora of problems innate to wankel engines, foremost of which are sealing and high combustion chamber surface to volume ratio, which is conducive to heat loss into the engine. Low compression, or rather expansion – in the case of the Atkinson cycle, ratio has traditionally also been a problem with the wankel. All this adds up to typically higher specific fuel consumption figures for the wankel.

    Other problems include uneven and higher heat loss troughout the stator shell as only one side of it is exposed to combustion, and exposed continuously. Difficult lubrication of rotor seals. Difficult maintenance.

    An Atkinson cycle engine based on the rotary concept would share problems with the Wankel. Designs that seem great on paper usually don’t work as great irl. I imagine sealing and lubrication would be even more of a problem for an Atkinson rotary than it is for the Wankel. The compound, intermeshed rotary adds a lot of complexity and great opportunities for wear, with its obvious lubrication difficulties. Possible balancing issues.

    The reciprocating piston based Atkinson engine also doesn’t seem very promising, imho. Packaging is the obvious concern. But I also see problems with its conrod and levers gear as well. There’s more to bend and flex at rpm and more reciprocating mass to suck up useful work. Lubrication issues and small journals. And on top of all, engine balance issues, the same as with the offset cylinders engines.

    If only there were some easy way to delay the moment of peak cylinder pressure closer to when the conrod is in the most favourable position to imprint torque to the crank without sacrificing compression ratio, engine balance or increasing reciprocating mass.

  10. @SonyAD – I was thinking of the reciprocating Atkinson cycle; I didn’t know there was a rotary one. There is a way to get all the advantages of an Atkinson cycle with a simple modification of the Otto cycle. All you do is keep the intake valve open for the first part of the compression stroke. Thus during the first couple of degrees of compression, the fuel and air are simply blown out of the intake manifold instead of being compressed. The intake valve closes for the rest of the compression, and the power stroke uses the same kind of valve timing you’d find in an otto cycle engine. This system results in a longer power stroke compared to the compression stroke (or, another way of looking at it is that the peak pressure of compression is lowered and hence the pressure everywhere is lowered, and hence final pressure after the power stroke is closer to atmospheric pressure, hence more efficiency).

  11. “All you do is keep the intake valve open for the first part of the compression stroke. Thus during the first couple of degrees of compression, the fuel and air are simply blown out of the intake manifold instead of being compressed. The intake valve closes for the rest of the compression, and the power stroke uses the same kind of valve timing you’d find in an otto cycle engine.”

    Why don’t you just not open the throttle plate as much?

    I would try a petrol engine with high compression ratio, something like 18:1, that would run choked (though still stoichiometric) through the throttle plate most of the time so as to avoid preignition or detonation. As rpms increased the throttle plate would be allowed to open to a greater extent so as to compensate for poorer cylinder filling.

    “This system results in a longer power stroke compared to the compression stroke (or, another way of looking at it is that the peak pressure of compression is lowered and hence the pressure everywhere is lowered, and hence final pressure after the power stroke is closer to atmospheric pressure, hence more efficiency).”

    As I see it, the power stroke isn’t any longer in terms of crank degrees.

    Low compression pressure is not a good thing. Besides not homogenising the air-fuel mixture as well and making combustion slower (which promotes heat loss into the engine) it means you’re working against a steeper pressure gradient in a naturally aspirated engine as the reciprocating piston engine makes power from the pressure difference between the combustion chamber and the crankcase during the power stroke.

    Restricting air flow into the cylinder during the induction stroke means the piston descends inside the cylinder against a pressure gradient. Part of this energy is probably recovered through less resistance during the compression stroke but energy is lost again during the power stroke, when the piston again descends. Some of the energy of combustion is expended first overcoming the negative pressure difference, before positive pressure difference (between the combustion chamber and crankcase) can start to push the piston down. This energy is not recovered during the exhaust stroke. This is what throttling losses are. And it’s one of the reasons why diesel engines usually have better specific fuel consumption, particularly at part throttle. Diesel engines are freebreathing. If they use a throttle it’s probably to promote and control exhaust gas recirculation into the intake manifold. Of course, throttling somewhat reduces pumping losses as well, but the net effect is one of decrease in efficiency.

    I see no difference between not opening the throttle plate as much and closing the intake valves after compression commences. In fact, I think closing the intake valves later might be less efficient than not letting as much mixture inside the cylinder in the first place. But I may be wrong, or else Toyota would have opted for this approach instead?

    In any case, in a naturally aspirated engine you’re always pumping against a pressure gradient. Even with no throttle at all, you’re still pumping against a pressure gradient because there is simply not enough time to completely fill the cylinder with air to atmospheric pressure. And this gets worse with rpm, which is why engines (naturally aspirated ones especially) benefit from variable intake valve timing and lift. But throttling makes things worse.

    Also, I don’t think exhaust pressure is a good indicator of engine efficiency. A good indicator of engine thermal efficiency is exhaust gas temperature. I presume measured as close to the exhaust valve as possible. Though this does not account for thermal energy lost to the engine and dissipated through its cooling system.

    Generally speaking, an engine with a higher compression ratio will be more thermally and mechanically efficient, because combustion gasses may expand more before being allowed out of the cylinder in addition to the advantages already espoused regarding . So there is more opportunity to extract useful work.

    This is another important reason why diesel engines are more efficient, as they generally have higher compression ratios.

  12. @SonyAD: Sorry for not explaining the intake valve thing very well. Here’s what I meant (this is copied from Wikipedia): Recently Atkinson cycle has been used to describe a modified Otto cycle engine in which the intake valve is held open longer than normal to allow a reverse flow of intake air into the intake manifold. This is more like a Miller cycle engine than an actual Atkinson cycle engine. The effective compression ratio is reduced (for a time the air is escaping the cylinder freely rather than being compressed) but the expansion ratio is unchanged. This means the compression ratio is smaller than the expansion ratio. Heat gained from burning fuel increases the pressure, thereby forcing the piston to move, expanding the air volume beyond the volume when compression began. The goal of the modern Atkinson cycle is to allow the pressure in the combustion chamber at the end of the power stroke to be equal to atmospheric pressure; when this occurs, all the available energy has been obtained from the combustion process. For any given portion of air, the greater expansion ratio allows more energy to be converted from heat to useful mechanical energy meaning the engine is more efficient.

    The disadvantage of the four-stroke Atkinson-cycle engine versus the more common Otto-cycle engine is reduced power density. Because a smaller portion of the compression stroke is devoted to compressing the intake air, an Atkinson-cycle engine does not take in as much air as would a similarly designed and sized Otto-cycle engine.

    Four-stroke engines of this type with this same type of intake valve motion but with a supercharger to make up for the loss of power density are known as Miller cycle engines.

    If you were to close the throttle plate, wouldn’t that increase pumping losses and reduce overall efficiency?

  13. BTW, have a look at the MCE-5. It’s a (constantly and continuously) variable compression engine which also eliminates piston side thrust. However, there’s also a lot of reciprocating mass. The concept has been in development since ’97 and still has some way to mass production. Apparently the concept has been validated by PSA, which is the first manufacturer licensed to produce such an engine, under the VCRi badge.

    Peugeot are one of my favourite manufacturers but I don’t think this design is all that good.

  14. LMF, the effect seems to me basically the same between not opening the throttle plate as much and closing the intake valves after compression begins. You still have to pump some of the mixture back out, which expends energy, and you still have throttling losses. That is to say, if ignition were not to occur, the pressure inside the cylinder when the piston reaches bottom dead centre on what should have been the power stroke (assuming the exhaust valves don’t open before BDC on the power stroke) is lower than atmospheric. This means that during the power stroke there is a negative pressure difference between the combustion chamber and the crankcase to overcome in addition to driving the other pistons in their strokes and propelling the vehicle. This is true of any naturally aspirated engine but is even more so for a throttled naturally aspirated engine running under part throttle.

    I think what you mean by ‘compression ratio’ is actually the overall pressure ratio. What I mean by ‘compression ratio’ is the ratio between the combustion chamber volume when the piston is at bottom dead centre and when it is at top dead centre.

    The pressure ratio is highly dependent on more than just compression ratio, as you point out. It also depends, for example, on atmospheric pressure or pressure inside the intake manifold, if the engine uses forced induction. And a host of other factors.

    In my comprehension, one can speak of an engine having distinct and different compression and expansion ratios if that engine’s piston travel during the compression stroke (from piston BDC to TDC) is longer or shorter than the piston travel during its power stroke (from piston TDC to BDC) irrespective of the moment ignition or injection occurs or the intake or exhaust valves open.

    Which, I think, is not the case with any automotive engine currently in production, though I may be wrong.

    “The goal of the modern Atkinson cycle is to allow the pressure in the combustion chamber at the end of the power stroke to be equal to atmospheric pressure; when this occurs, all the available energy has been obtained from the combustion process.”

    This is a bad goal. Nevertheless, if it’s attainable by closing the intake valves during the compression stroke then it’s surely attainable by not opening the throttle plate as much.

    As I said before, I think a much more relevant indicator of engine efficiency are specific fuel consumption under part load as well as full load and exhaust gas temperature under various loads and rpm.

  15. Wow… complicated stuff. I kinda feel glad that the future is in electric cars… thankfully the motor is a lot simpler than an engine and also a lot more efficient!

  16. sonyAD, i think i’m in love with you!

    Your amazing knowledge of the ins and outs of a modern motor vehicle engine shocks me, and i would love to know, where you have learnt this all from?

    I’m currently studying motorsport engineering in college, in my third year of HND level 3, which i want to take to uni.

    Could you possibly let me know where you’re studying and how its going etc?

    Thanks very much, and once again, amazing!

  17. Brian Carr, Your article is totally misleading. If not for the good folks on here that have helped straighten the path, I think I may have been fooled into thinking engine braking was a waste of fuel.

    Seriously, at least get your facts straight before you mislead people. There is no fun in that. Or change the topic. This is false, one of the reasons people don’t trust the internet. You came on here sounding like a “trusted source” but by all of your arguments and so-called ”logic” it is obvious you know jack-squat about engines and the effect of engine braking. Please stick to your day job (hopefully this is not it).


  18. Who wrote this? says:

    This is a load of nonsense, as the only time fuel enters the engine is when you push the gas pedal. Rev matching will push some fuel in there to match RPM’s for a down shift but while the engine is being used for breaking you are not touching the gas pedal and throttle is released therefore no fuel is being used. In fact the intake and exhaust valves are closed essentially creating a vacuum effect that slows the car down more than if you were to stay in neutral. Also, engine breaking is essential for optimal engine break-in results. Sure you might use a tad more gas by blipping the rpm’s up for a rev match but that should be negligible.

  19. I drive my car on cruise control. When I go down a steep hill, it automatically downshifts to keep the pre-set speed. The RPM increases relative to the steepness of the road.

    Question: Which is least expensive? Using brakes or letting the engine do the job??

  20. M. Ganesh babu says:

    Came across this wonderful discussion on engine braking and had to add my 10 cents 😉
    I’m a big fan of engine braking. To feel the braking torsion from the rear wheels, the gearbox hum while engaging it, controlling the braking linearity with the clutch, clutch down gliding and at the end, possible reward of a drift have always fascinated me. It now looks like the best might be yet to come. Engine braking it turns out is an essential part of brake force regeneration. During engine braking the crank pulley of the engine is revved up. Dynamo alternator and AC compressor are driven by the crank pulley and thus benefit from these extra revs. Hence those of us who have cultivated the habit of engine braking have also practiced to drive the hybrid cars of the future. There is however a chick in the armor. In-order to maximize the efficiency of brake force regenerated, manufacturers employ a technique known as valve idling. This is a technique used to eliminate engine drag during engine braking and make available the entire returning force from the wheels available at the crank pulley for these accessories to harvest. Valve idling is achieved by leaving the inlet valves open during engine braking. So my dear friends this could be the end of engine braking as we know it. Ofcourse dynamo alternators right from our bicycling days we know can provide drag. However no standalone alternator would be able to produce the kind of drag provided by a conventional IC engine while engine braking. Yes, if the alternator is part of the flywheel integrated DC motor reversing as an alternator while engine braking, then it could generate meaningful drag. And so my friends, it will be a short goodbye for engine braking before we can say hello to it once again in our hybrid cars 😉

  21. Actual, engine braking MIGHT be harder on the conrods especially at higher RPMs. THe inertia from the downstroke is harder because it is dragging down the piston that was already in an upward motion. The higher the speed the more this inertia.

    • I think the inertia of the piston is compensated for by the compression in the chamber reversing the piston back to to follow the conrod

  22. What about diesel engine. For example take the land rover freelander 1. When you downshift in a diesel car, is the decelration fuel cuttoff or anythingsimilar to that? Does it save fuel when you downshift/engine brake a diesel car. As EFI petrol cars are like diesel fuel injectors, except for the spark plugs, and as DFC occurs in EFI petrol cars, doesnt it mean it can happen in diesel cars?

Speak Your Mind


SEO Powered By SEOPressor