By on December 17, 2012

Chris writes:


In a couple recent Piston Slap articles you’ve mentioned that when driving car with a manual transmission its most efficient to accelerate with the engine near its torque peak, then cruise in the highest gear possible. This raised two questions in my mind:

1. Does the engine’s torque peak vary based on throttle position? From what I understand, power and torque curves are generated at wide open throttle. But would the torque curve look different at, say, 50% throttle? I’ve heard that exhaust backpressure can affect the torque curve (maybe this is a myth). Could throttle position have the same effect via intake vacuum? Speaking of intake pressure, that leads me to my real question:

2. How does your strategy of accelerating with the engine near its torque peak apply to a turbocharged vehicle? My car has a turbo and according to the manufacturer the torque peak is 2000 rpm. But clearly it’s not always capable of generating max torque at 2000. If I’m loafing along at 1800 rpm and floor the throttle it takes a very laggy second or so for the boost to build and its definitely past 2000 rpm by the time it starts really generating power. I’m thinking there must be a different torque curve for part-throttle acceleration, when the engine is either off-boost or not making full boost. I think this would also apply to an engine like Audi’s supercharged V6, where the supercharger can de-clutch from the engine under low load. Any thoughts on the most efficient way of accelerating in a turbo? Better to accelerate “on boost” at relatively low rpm and relatively wide throttle? Or accelerate with less throttle, keeping it out of the boost (but probably winding the tach up more to avoid moving at a snail’s pace)? Or just forget the whole thing, floor it and enjoy the wild turbo-torque surge?

If these are stupid questions, please disregard. These are just things I ponder while sitting in traffic… Keep up the great work!

Sajeev answers:

This is a fantastic question that I am totally not qualified to answer…but that hasn’t stopped me before, and it hasn’t stopped you lovely people from reading, so let’s do this thang!

Point #1: Yes, throttle position will affect the torque peak. Because an engine is basically just an air pump, if you have less throttle you have less air, less fuel and therefore less power.  Thankfully, with the advent of electronic fuel injection there are multiple mappings: older systems have a full and a part throttle program, and newer systems probably have several.  So I betcha you can maximize an engine’s efficiency at just about any throttle opening. Every application is a little different, and many are tuned to maximize performance with a computer reflash from an aftermarket programmer.

As a rule of thumb, and I’m ready to get slammed by engineers for saying this, backpressure (or a lack thereof) does indeed affect the torque output of an engine.  More importantly: backpressure isn’t a good thing, finding the ideal exhaust velocity to minimize backpressure while keeping the speed “slow” enough to not hurt torque output is crucial.  That’s why, in the past 10-15 years, we see far higher quality exhaust systems in all OEM applications: no crush bends in the tubes, cast iron manifolds that are shaped more like aftermarket tubular headers, and mufflers/catalytic converters that aren’t a significant restriction.**

Point #2: turbocharged motors are just like point #1 when it comes to power in part throttle applications. And every boosted application out there is different. Once again, and even more so, tuning makes ALL the difference in the world.  Because the turbo is a muffler/restrictor in the exhaust system, you want as little restriction behind it to ensure maximum efficiency: hence why the Dodge SRT-4 is muffler-less from the factory.

My gut feeling is that with any modern car, turbo or not, you need to give it more gas to cut through the slop of electronic throttle control/torque management to get into your torque peak quicker.  Spend less time accelerating and more time cruising, with traffic conditions in mind of course. That doesn’t mean you run wide-open throttle, either. There’s a happy medium out there, somewhere.

Off to you, Best and Brightest: I’m ready, I’m wearing my flame suit.

**Grab a catalytic converter from the 1970s-early 1990s. They neck down, restrict air flow, etc far more than the goodies I see today in cut-away diagrams at the auto shows.  We have come a long way, baby.




Get the latest TTAC e-Newsletter!

22 Comments on “Piston Slap: Of Power Curves and Turbo Boost…...”

  • avatar

    To reinforce the point on restrictive cats…The FIRST and one of the few performance mods on my Gen-1(1989) S.H.O is a new catalytic converter assembly that is much less kinked and has smoother seams.

  • avatar

    The restriction in the exhaust by the turbo is a myth at light throttle. Comparing turbo vs normal aspirated(NA) 4-cylinder shouldn’t really be compared in the same article, at least when it comes to low rpm efficiency.

    At cruise and light throttle my SAAB 9-5 is creating a slight pressure on the throttle body where a NA engine doesn’t. The positive pressure on the throttle body helps push the 3400 lbs 5-speed to around 44 mpg @ 60 mph. The biggest difference is the ECU is tuned differently with today’s turbo charged 4-cylinder. And even more differently with direct injection than NA. Throw in electronically controlled automatic transmissions and throttle bodys and every one blames turbos for lag which might not be the root cause.

    • 0 avatar

      My read was Sajeev was pointing out that in some factory apps, the turbo (and the cat) are the only mufflers. In the bad old days 30 years ago, before everyone figgerd out how to properly size a turbo, they could be restrictive until fully spooled. You’re right, that for the last 10+ years, they aren’t a restriction at all.

      (BTW- Sajeev, SRT4? The Dodge Shelby GLHS Omni used the turbo and cat as mufflers too. In 1986.)

      • 0 avatar

        et tu, GLHS? Very cool.

      • 0 avatar
        Felis Concolor

        Check out Honeywell’s site for some new developments in the turbosupercharger arena; the double sided intake compressor and axial flow exhaust turbine promise a new level of responsiveness, and my GLHS is pleased to see the return of the VNT to the arena.

        Even in the late 80s, catalysts had improved to the point where dyno tests were showing them to flow better than open headers, and the decades of 1s and 0s thrown at them have only served to provide equally impressive flow rates in ever more compact packages.

      • 0 avatar

        Glad to hear about another one still going. Primitive by current metrics, but they were a hoot back in the day. And with some modern tools, and parts they can be made fast even by today’s standards.

        One of (if not the) earliest automotive apps of VNT, and squandered like so many other technologies by the sheer hubris, ignorance, and short-term world of D3 corporate culture.

        All of the great EFI tech that Bosch sold to much of the world back when the D3 finally couldn’t make carbs work anymore? Developed by Bosch with the patents they licensed from Bendix. Chrysler sold a few 1958 cars with the Bendix Electrojector system, but it needed more work. In the already broken world of the D3, why bother? Gas is cheap, carbs are cheap, who cares if cars run better?

        Detroit. Ceding the future since 1955…

  • avatar

    Well, what really matters is where the fuel conversion efficiency is maximized, not where peak torque occurs. The two are related, though. Torque essentially tells you how much work an engine puts out with each revolution. In a nice, ideal world, the air (and thus the fuel) pumped through the engine with each rev will be the same; thus, when you get maximum torque, you are converting the most fuel into useful work.

    One of my textbooks (which a roommate walked off with, I think) had a sample plot of fuel conversion efficiency as a function of engine load & rpm. The area where it was maximized was relatively low rpm & high (but not max) load.

    I really can’t comment on where to shift to produce the best efficiency, but given that autos are programmed to give the best efficiency, and they shift at the lowest possible speeds, and that would seem to jive with the plot from my textbook, I suspect that’s the best method to get the best mpg.

    • 0 avatar

      “One of my textbooks (which a roommate walked off with, I think) had a sample plot of fuel conversion efficiency as a function of engine load & rpm. The area where it was maximized was relatively low rpm & high (but not max) load.”

      IIRC, that’s usually a point in the torque peak of a motor. The torque peak is the engine’s maximum “power efficiency” point. I think.

      You want to accelerate in the most “power efficient” manner, and then enjoy cruising at low rpms for the most fuel efficiency.

      • 0 avatar

        While closely related, I doubt the two are necessarily the same point. (Torque is just the total work output per cycle, so more air + more fuel = more torque, even if it isn’t burned as efficiently.) Again, going back to auto transmissions, if going all the way to the peak torque was the most efficient method, that’s how they’d be programmed, but they aren’t.

        Also, for a true optimization, you have to consider the ‘path traveled’ by the engine & gearing. Since you want the whole time in each gear to maximize efficiency, you would actually go past your peak & shift at the point where going into the next gear keeps you close to the optimal point. The exact best point would depend on the next gear you are shifting into.

    • 0 avatar

      “plot of fuel conversion efficiency as a function of engine load & rpm. The area where it was maximized was relatively low rpm & high (but not max) load.”

      That’s usually called a bsfc map. An example is found here:

      • 0 avatar

        All the BSFC charts I have been able to find for turbocharged engines are for WOT and are used for sizing or tuning. To be useful for “efficiency” predictions in real world situations the part throttle data is necessary (i.e. for “hypermiling”).

    • 0 avatar

      “Torque essentially tells you how much work an engine puts out with each revolution.”

      That should be “how much force”

      Torque by itself does not produce work (probably not better example of this would be an electric motor producing 1000 ft/lbs of torque at zero RPM – lots of force but no work is being accomplished).

      Two engines producing “X” amount of horsepower coupled to something like a CVT where the engine could be kept at its peak power the entire time despite substantially different torque values would produce the same amount of work.

      Its in the imperfect world of limited gear sets and varying RPM where torque becomes a valuable commodity as it can offer more power “under the curve” provided the other engine does not have enough gear sets to compensate.

      Also, I’ve always been baffled by the “back pressure = torque”?

      Years ago I had read some articles (daygonit, can’t remember the guy’s name, but he did a series of online lectures and Hot Rod articles – the best were properly sizing exhaust systems using engine size, cam timing events, and so on – Professor Power, whats his name??????) and there was going to an article on back pressure and torque but it never materialized.

      I’ve always maintained that an exahust system than minimizes or elminates back pressure without killing exhuast velocity or over scavenging the engine is the best state to be in.

  • avatar

    Fuel Injection (especially electronic fuel injection) uses mathematical formulae for mixing the right amount of fuel to the right amount of air to get a desired power outcome.

    The factors considered included intake air temp, intake air mass, engine temp, throttle position, RPM, and sometimes manifold absolute pressure and/or EGT (especially in a force inducted scenario).

    All these are taken in to account and ultimately determine how long the injection pulse (squirt) will be.

    EFI 101 is a few pages that I have done wayyyy too many times over the last 20 years, and there is no space here for Sajeev or anyone else to do an article that long, soooo….

    A great explanation of EFI basics and terminology (with drawings!) can be found over at Megasquirt. For those who don’t know, they offer a DIY EFI computer.

    They’ll get you through the basics. If someone has questions about the evils of deadhead rails, the importance of fluid dynamics, properly sizing rails, where to place your dampeners, how to balance your pressures, or how to fix your D-Jet Benz from 1973, those I can answer more succinctly.

  • avatar

    I am no engineer but I like to understand things so this comment is generalized and based on my understanding.
    In the old days before fuel injection, variable valve timing and electronic ignition, torque and power peak were affected by the physical design of the motor. That meant the size and placement of the valves, the valve timing, carburetor/s, the ignition timing, flywheel and even the size and shape of the pistons and crank.
    Torque peak was, and still is to some extent, the point on the rev range where the engine was at its most efficient, no matter the throttle setting. The best acceleration would be some ware in the rev range starting with the highest torque and power, up to just passed the power peak.
    Today these clear lines are blurred by VVT, fuel injection, electronic ignition, timing and turbos. They all lead to flatter (production cars) torque curves meaning the engine is as at maximum efficiency at a MUCH wider rev range.
    Physical engine design still plays a role though and the higher the engine revs the more fuel it will use.
    To drive a manual at it’s most efficient, keep the revs as low as possible within the area of the cars torque peak. Knowing the torque curve of the vehicle is extremely helpful.

  • avatar

    As a theoretical question, this is very complex. The practical answer is much simpler and hasn’t really changed for decades.

    It’s most efficient to cruise at the lowest engine rpm. Top gear is better than a lower gear. However, at low speeds on steep hills, there comes a point where you need to use a lower gear. It varies according to whether you are going up hill or down. Full throttle in top gear at 1,000 rpm isn’t a good idea. Downshift instead.

    When you need to accelerate, accelerate. Upshift when you reach a road speed that will still give you reasonable acceleration in the next higher gear.

    When you need to accelerate hard, accelerate hard all or most of the way to red line. Because torque drops off at high rpm, there may be a crossover point below red line where acceleration in the higher gear is as good or better than acceleration in the lower gear. Unless you have special instrumentation, you will have to judge this by the seat of your pants.

  • avatar

    Gear ratios matter to. Not all transmissions have nice close ratio, well matched gearing. Some have weird gaps and are more often designed with economy in mind.
    Sometimes, when changing up too soon leaves the car trying to accelerate with the revs too low and out of the engines working range. That can result in too much fuel in the cylinders leaving some of it unburned. That is very uneconomical and dilutes the oil…
    For most cars and engines though, any driver with even a small amount of mechanical sympathy will be able work out when the motor is not happy.

  • avatar

    Sajeev, thanks for taking my question. I think you (and the other respondents) are probably right that optimal mileage is going to depend on a lot of factors including ECU programming, throttle calibration and probably dozens of other factors. I guess “your mileage may vary” applies here.

    In the past week or so I’ve tried keeping the car around the 2000 rpm torque peak when accelerating. This generally means driving my car like its a diesel, usually up-shifting by 2500 rpm. I’ve noticed an increase in (indicated) mpg of about 20% (from 18 to 22 mpg – almost entirely stop-and-go, in and around downtown, so how I accelerate has a big impact on mileage). However, I don’t know how much of this increase is due to accelerating in the optimal rev range and how much of it is due to the fact that it forces me to accelerate a lot slower than usual. In any case, I’m too impatient to keep driving this way :)

    • 0 avatar

      Actually, one thing I realized this morning is that my car doesn’t so much have a torque “peak” as a plateau from around 2000-4500 rpm, at least judging by the seat of my pants. So probably best to just accelerate in this range and then cruise at low rpm.

      • 0 avatar

        The wastegate will stay closed until boost builds and will start to open(spring pressure) to limit over boosting. This is some of the plateu you feel the turbo spools up giving a rush then tampers off when still adding throttle input.

        Some turbo systems you can put your foot into it from cruise and let up right away as the turbo is starting it’s spool up and you can hear what sounds like someone stretching the open end of a full balloon and letting air out.

        For better economy make sure spark plugs and fuel system are up to date, tire pressure at or higher than recommended on the door jam(higher for high speed driving and carrying heavy loads like passengers), good fuel is hard to find with only 91 octane offered out west and oxygenated winters fuels in the north east. Cooler temperatures don’t help fuel economy but are great for power.

        I’d recommend a ECU tune from your favorite tuner for your car model(VAG). The ECU will allow more timing ang boost for a given throttle input yielding less fuel consumption.

  • avatar

    This is a very interesting question. All the usual babble about torque peaks presumes full throttle, which rarely happens in road car use.

    These days, prodding the accelerator pedal to some position is interpreted by the ECU as a request for x amount of torque. The ECU reads road speed, engine temp, engine rpm, outside temp, position of the variable cam phasers, sun spot activity, cornering g, and issues an order to the throttle position motor to assume a certain initial position, where its lookup tables have said the required torque will be output by the engine.

    Turbo motors will have low torque capabilities when the turbine isn’t spooling very much, so the throttle may be fully opened by the ECU, even though you have only given the accelerator pedal a small push. That is to disguise the low output capability of the engine at that moment at that apparently small accelerator pedal push.

    As the turbo spools up, and engine speed also increases, the ECU dials back the throttle opening because the engine can make the desired output torque at this lower opening.

    It’s quite a dynamic situation, and nothing like pulling on a throttle cable and hoping for the best. Modern ECU tuning is quite the arcane art. Unburned fuel? Not likely at all, since all parameters are constantly being checked.

    Cobb tuning has more in-depth discussion on their site, if you don’t go cross-eyed reading it and decide a Big Mac would be good right now instead.

    The bottom line? You as a human want the best economy while accelerating? Get in the highest gear you can as soon as possible. It’s pretty obvious when you watch the instantaneous mpg display on your dash. Low engine speeds minimize friction losses.

    Or you just forget about it, and let the ECU do the worrying part while you sip your latte.

  • avatar

    I just had a giddy moment looking at that Knight Rider picture.

    Off topic I know…

Read all comments

Back to TopLeave a Reply

You must be logged in to post a comment.

Recent Comments

  • EBFlex: “According to a T-shirt I saw a young man wearing on June 19th at church celebrating the 4th of July...
  • EBFlex: “His keyboard doesn’t have spell- or IQ-check.” “Certainly no IQ check. But probably an...
  • Inside Looking Out: Happy July 4th to all!
  • DenverMike: It’s not just the parking thing. Most Americans simply can’t drive something that big, scared...
  • DenverMike: Unless you were already struggling to feed the beast, yeah a few small changes and you’re good. Yet...

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Jo Borras
  • Mark Baruth
  • Ronnie Schreiber