The definitive typical 16v dyno plots parts 1 and 2 von Toyotecwerke

Discussion in '16-valve' started by Toyotec, May 14, 2011.

  1. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    Part 1

    Introduction

    During the various studies written up on Club GTI previously, others and myself have tried to further our collective understanding of engine outputs, and the interpretation of information from dynos. It is very easy for a casual reader to focus on the peak numbers, accept them as correct and have no basis to question the result apart from a shoulder shrug. With a little more scrutiny, further validation techniques enable a reader to understand far more.

    It has been asked many times what are the typical power and torque of standard VW 16v engines. This question has 2 visual answers, 1) the peaks seen, and 2, the shape of the power curve.

    The purpose of this thread is to illustrate how engine outputs can be read, understood and correlated via:
    • by the use of a g-meter
    • on the profile of a dyno graph
    • the driver's 'backside dyno'
    • published OEM information

    The first three detection systems are reading the same information, and the first two and latter one give a graphical output.

    When driving, if you feel an acceleration peak at particular engine speed, so should the dyno, and so does the G-meter. You are feeling torque. This leads to a recognisable shape of any dyno torque output.


    Thanks to the recent successful NASP 16v rolling day we have seen the standard characteristics of an EA827 16v engine illustrated on the forum recently, and what to expect when this engine type is tested on any good modern dyno.

    [​IMG]


    Folklore

    Everyone likes big power numbers. Having an engine that is more powerful than "std" justifies that the big bore exhaust or shiny chrome intake manifold and cone filter are making a 'huge' difference.

    We have heard suggestions on numerous occasions that std ABF engines "are good for an easy 170 bhp and 150 lbft", enough to put fear in the release of VAG's Ankor weight 174bhp VR6 range. Also mentioned are KR engines with a mere "WUR mod" achieving 155 bhp. Usually these numbers tend to come without a plot or rpm references, allowing them to be elevated, unquestioned, to the pub or forum "wall of fame".

    If they do have a dyno curve, the numbers, which to some may look impressive compared to OE published plots, often display results that actually show errors relating to the power and torque peaks and generally the shape of the measured torque. By all means check other dyno prints posted around on the forum, but I shall not point specifically to any. This can be detected by a seasoned tester through actual vehicle feel.

    When I am asked to analyse performance data, the first question is usually: "where does this typical 'XXXbhp' or 'XXX lbft' fall on the plot profiles?" In the background, I am referring to standard engine output information eg 16v. This then leads into other questions relating to fuel mixture logged during the dyno run. Armed with this information, the validity of the dyno result can be validated. To move engine performance peaks around significantly requires hardware changes. If they are misplaced, the plot is wrong.


    The G meter

    • Axle torque transmitted to a vehicle's tyre results in a force to move the vehicle. The force when divided by the inertia weight of the vehicle results = acceleration a G force.

    Automotive G-meters can observe this acceleration and in certain cases engine speed can be acquired to determine where the peak points of G are during a full throttle pull.

    After setting vehicle weight, the device reports maximum G, which is sometimes conducted in second gear of a test vehicle to minimise opposing forces i.e. rolling resistance, gearbox friction and aerodynamic drag.

    We know that the tyre patch force is transmitted through the tyre radius, axle and gearbox multiplication from engine clutch torque. Maximum G force represents the engine at max torque.

    The rpm enables the G meter to plot Gs vs RPM which can be used in creating a plot of road torque. 'Road power' can be derived from looking at the rate of acceleration and speed with a known inertia weight (mass). So Power in a straight line = Force x Velocity or Power = ( mass x acceleration) x vehicle speed). If the vehicle weight is entered in the G meter it will still measure power without any rpm input, and is very similar to what is seen on the DD 450DS dyno at Garage Streamline when engine speed is not acquired. Power is still measured but there will be no plot.


    G force vs real dyno plot

    G-meter tests have been done on two Mk2 Golfs one with a KR KJet engine and the other with a standard ABF engine controlled by a digifant 3.2 system. The engines and vehicles were well maintained and were working in correct order.


    Here is a screenshot of the Gs in time from Chris Eyre's 1988 Golf 2 KR with OE 14in wheels and 2Y gearbox.

    [​IMG]


    This data with a weight and rpm is entered into the device results in a power and torque curve. Note the values are not absolute numbers due to estimating weight. Purely done to illustrate the profile

    [​IMG]


    The raw data from the G meter was downloaded and entered into a spreadsheet to compare with dyno data from an actual test done on the same vehicle.

    [​IMG]


    A similar test was done on Gad's Golf 2 GTI 8v fitted with an ABF digi 3.2 engine and 8v AUG gearbox a and G60 15in steelies

    Here is the screenshot of Gs in time.

    [​IMG]


    Gtech Screenshot of non absolute power and torque due to estimated weight.

    [​IMG]


    and Gtech raw data combined with actual dyno test data on a spreadsheet chart.

    [​IMG]

    In both cases, the data from the G meter clearly shows a direct correlation to the dyno test.

    In the plots we can see that the KR engine tested, this maximum G force and TQ occurs at ~ 5500 rpm an and in the ABF engine this occurs at ~ 4500rpm. Max power occurs at 6400 rpm for the KR and 6200 rpm for the ABF. There was some noise that entered the G tech data which accounted for the shift compared to actual dyno numbers.

    Real world and dyno testing, looking at profiles and peaks, are therefore validated!


    Gs correlated to the what the butt dyno feels

    How does this translate in terms of acceleration between an optimised KR and a well maintained MK2 ABF digi 3.2?

    Even with larger wheels and a slightly longer 2nd gear, the Golf MK2 w/STD ABF engine is able to out accelerate the other vehicle fitted with an optimised KR engine every where as the plot below shows in Gs.

    [​IMG]

    In the second part to this exercise, I will show how this dyno data correlates with the OEM published vehicle information, long replicated on the 16v outputs link above.

    I'll let this post digest and add part 2 in due course.

    Thanks for reading :thumbup:
     
    Last edited: May 15, 2011
    Nige likes this.
  2. tshirt2k

    tshirt2k Forum Junkie

    Joined:
    Oct 21, 2004
    Likes Received:
    325
    Location:
    Herts
    Great write up as usual. Maybe all these 10,000 word essays can go towards your 16v doctorate? Then you can change pastor ventile to Dr ventil. :thumbup::lol:

    Good to see how the G tech replicates the shape of the dyno curves. Just wondering if there's is an error in a setting entered into it that would cause that shift in output? Maybe it's the "estimated weight"? Wondering how an increase or decrease would affect calculated power/ torque? You mentioned noise, but would that affect profile shape rather than calculated output?
     
    Last edited: May 14, 2011
  3. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    Thanks for the kudos.

    For the Gtech, the force at the wheels would be affected by weight input. This is probably wrong and so the amplitude of the pwr and torque will be affected globally.
    Some skewing will happen to the profile due to unknown road flatness, vehicle suspension and more time to accurately sync the engine speed to alternator noise.
    In the end the Gs are Gs and this represents a quantity that cannot be messed with. This is what you feel.
     
    Last edited: May 14, 2011
  4. Admin Guest

    Yet again another cracking read, keep it coming Eddie.
     
  5. Mike_H Forum Addict

    Joined:
    Oct 16, 2004
    Likes Received:
    17
    Location:
    iQuit
    Let the science continue!

    It's a bit off topic, but at some point it would be great to correlate results from your G-Tech meter against the Dynolicious app on my iphone. Interesting to see if they're worth using for a bit of budget home tuning.
     
  6. tshirt2k

    tshirt2k Forum Junkie

    Joined:
    Oct 21, 2004
    Likes Received:
    325
    Location:
    Herts
    I was thinking this too.
     
  7. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    Would be good to do actually. I am all up for it.
     
  8. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    Definitive typical 16v dyno plots part 2

    Part 2

    Now that we have seen similarities in the data from the dyno and real life acceleration via the G-meter, and established where the typical peaks are for std engines, we can move forward to compare this information to the information issued by the manufacturer.


    VW published plots.
    There were many plots, published by VW to market their vehicles back in the day.

    We have seen several of them on this forum over the years.
    Here are some examples I have collected.

    KR
    [​IMG]

    9A
    [​IMG]

    ABF
    [​IMG]

    VW published plots - made legible!
    It is very difficult to compare these plots, with the size, appearance and scale being so different. Plots for the KR amd ABF are very smooth and lack many of the rough bumps in the curve as the engine accelerates out of certain conditons at WOT. The 9A curve found in VW Self Study publications does seem to be more representative.
    To enhance comparabilty, these results have been magnified and transposed by hand into a spreadsheet and all the OE data curves are on one chart seen below.

    [​IMG]

    An interesting comparison, one that enables validation of the respective power units, but the purpose here is to compare dyno output and general profile with the information distributed from OE.

    Next the the dyno raw data is combined with the OEM graphs, to see how actual data compares. It should be noted that the OEM results from an engine bench dyno would have been performed to a much more controlled Deutsches Institut fr Normung (DIN) 70020 standard compared to the data retrived from the Dyno Dynamics 450DS chassis rolls which, in the open enviroment, is correcting and modelling engine output to a "ATMC2" std - a correction std very specific to this type of chassis rolls when power tests are conducted in the open environment.
    Besides the measuring device attempting to correct to a pre-set nominal, the vehicle when subject to ambients, will also react independent of data acquired. As the method to acquire dyno data was very well rehearsed and robust and the specific correction standard by the tool seemed to result in repeatable results from other unmodified vehicles in different ambients within reason, the actual number generated is accepted as representative quantity to OE advertised power an torque peaks.

    VW KR plot vs actual KR car.
    In this plot Chris' barn stored KR MK2 was compared to the original. Like the data from the G meter, torque and power fell at the same points and interestingly the actual data wrapped its self around the factory datum points.
    [​IMG]

    Unlike the 'smoothish' factory plot there was a loss in torque as the engine ramped up to 4000rpm. It is believed that pulse frequency changes from the intake manifold are at work in this area. Post 5000 rpm torque does not drop off as suggested by the factory plot and this may have to do with the removal of the airbox snorkel in the ,lower air-box half. Ultimately the engine runs out of huff as the revs quickly rise to the 6800rpm soft cut.


    Power Torque Conditions
    VW claims 139PS (137bhp) @ 6100rpm 168Nm (124lbft) @ 4600rpm DIN70020
    Actual 139bhp @ 6200rpm and 124lbft from 4500-5400rpm Dyno Dynamics ATMC2 correction


    VW 9A plot vs actual 9A car

    The 9A engine is controlled by a KE-Motronic system. This system allows stoichmetric operation at part loads and is used with vehicles that have a cataylist fitted. The engine also runs 'mild' cams. This plots shows how this engine compares to the factory plot.

    [​IMG]

    When this KE-Motronic system is removed and the engine is run on K jet, the same torque and power profile that would be present on a KE-Motronic would be present at WOT.


    Power Torque Conditions
    VW claims 100kW(136bhp) @ 5800rpm 180Nm(133lbft) @ 4200 rpm DIN 70020
    Acutal 132bhp @ 5800rpm 132lbft @ 4500 rpm Dyno Dynamics ATMC2 correction used.


    VW ABF plot vs actual ABF car.

    [​IMG]

    In this plot the vehicle tested also displayed close relation to the factory published plot. However what the factory plot does not show is ABF engine as seen tested with a G meter and in this actual test display maximum torque at ~4500rpm vs a smooth rise to 4800rpm as shown in the OE documentation. As the revs rise extra fuel is injected to cool the catalyst when it exotherms, and exhaust components from danger. Unfortunately this fuel mixture can be richer than 11.5:1 and has a penalty to torque generation. This occurs in all std ABF engined vehicles and is the reason why the torque drops after 4300 rpm until the engine runs out of air just after 6400rpm as shown in the plot below.

    [​IMG]

    This is one reason why ABF engines in a factory state of tune cannot achieve peak torque post 4500rpm. Dynos that suggest +5000rpm torque peaks on a factory tuned ecu with the aforementioned level of enrichment displayed would have results that are suspect.
    It also believed that this level of enrichment maybe accounted for in the factory plots and thus the 150PS @ 6000rpm claim by VW.


    Power Torque Conditions
    VW claims 110kW (150PS) @6000 rpm 180Nm (133 lbft) @4800 rpm. DIN 70020
    Actual 154bhp @ 6300rpm 138lbft@4300 rpm Dyno Dynamics ATMC2 correction used


    Conclusions.

    Next time you are at a dyno meet, remember what you are looking for as reference points when measuring a standard or near standard ABF, 9A/6A/ACE or KR engine, or indeed any other engine. Check the peaks are in the right places, or that you have the hardware to justify a peak in a different than 'standard' place.

    To achieve high power on standard 16v engines, torque needs to be kept as high as possible, > 133lbft at 5000+ rpm for larger 2.0 engines and > 124lbft from 5500+ for 1.8 KR units.

    It would appear the KR engines can peak their torque higher than their 2.0 relations, once fuelling is near to LBT ( lean best torque), due to valve size to bore ratio.
    From the data shared here, a hypothetical 155+bhp KR engine will need significant amount of work and will only be capable of standard 2.0 16v torque at ~133lbft at best. This of course will mean reduction in pumping and frictional loss and other hardware improvement to allow for a torque peak in the 5.8 to 6.0k rpm range.

    Likewise the "std" 170bhp ABF is a myth. To achieve near to 170bhp+ on an ABF engine controlled by an OE Siemens ECU, you will need to work on the over fuelling at WOT that occurs post 4300rpm. This has been addressed mildly on some after market eproms and does help the torque in this condition. All-out recalibration of vehicles that do not require a catalyst will mean torque upwards of 160lbft and an engine that can achieve 180bhp once the surrounding ambients and air density are favourable. Supporting bolt on modification will be necessary for this level of performance to be achieved.

    [​IMG] :thumbup:

    Many thanks to those who contributed to making this essay possible.
    You know who you are.
     
    Last edited by a moderator: Dec 10, 2013
    Nige likes this.
  9. tshirt2k

    tshirt2k Forum Junkie

    Joined:
    Oct 21, 2004
    Likes Received:
    325
    Location:
    Herts
    Great essay dr. Sticky needed. :thumbup:
     
  10. geordiegar Forum Member

    Joined:
    Oct 16, 2009
    Likes Received:
    6
    Location:
    Welsh Wales
    Superb Pastor.
     
  11. vw_singh Events Team Paid Member

    Joined:
    Oct 29, 2003
    Likes Received:
    781
    Location:
    United Kingdom
    Really nice write up Eddie. Showing those VW plots are not quite 'fake' as they seemed, just a little smoothed.

    Gurds
     
  12. seventynine Forum Member

    Joined:
    Sep 30, 2006
    Likes Received:
    26
    Location:
    Wales/Essex
    thankyou for that, very interesting read
     
    Toyotec likes this.
  13. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    ^^^Good ol engine in your sig - The Sigma
     
  14. seventynine Forum Member

    Joined:
    Sep 30, 2006
    Likes Received:
    26
    Location:
    Wales/Essex
    Sigma has the intake on the other side, that ^^^ is a black top zetec i'm afraid.

    i hear you work downstairs from me?

    (PM Sent)
     
  15. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    You are right it is an old Mondeo motor.

    Replied on email (DTC).
     
  16. Brian.G

    Brian.G Forum Member

    Joined:
    Feb 3, 2008
    Likes Received:
    446
    Location:
    West of Ireland
    Great stuff Ed,
     
  17. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,102
    Location:
    Creating Pfredstarke
    Thanks mate.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice