Toyotec's 2.0 16v MK2 dyno comparison. ABF (Digi vs Kjet vs MS) vs 9A MS. P3 14-11-10

Discussion in '16-valve' started by Toyotec, Oct 6, 2010.

  1. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    Abstract
    I have always wanted to test the popular 2.0 16v variants back to back on the dyno. These standard engines are now very popular updates to MK1-2 Golfs. Finally, an opportunity came around that I could not refuse and I thought I would share it with my fellow enthusiasts on this forum. Please read on.

    Introduction
    A pal with a 9A 16v MK2 had converted to MegaSquirt a long time ago and had issues with the calibration settings that were delivered elsewhere. He got in touch and handed the car over to start the mapping process. The another old buddy, who's car forms part of the "How to ABF your MK2" thread, came and offered his car as a dyno reference. This was the red car that is controlled by a Digifant 3.X sytem. Of late Ben S, who owned a 100bhp, Webber'd 1.6 8v, threw that engine in the bin and installed an ABF engine powered by MegaSquirt.

    With the 3 vehicles achieving an acceptable state of tune, all 2.0 16v, all live tuned, with the exception of the plug and play ABF Digifant, Garage Streamline was contacted and a date was set for back to back testing on the dyno.

    This topic details my experiences leading to and during dyno testing at Garage Streamline in London.

    First some notes.
    • All vehicles were in ~ **85% state of tune and never seen the dyno before.

    • All vehicles run in 4th gear

    • All vehicles MK2 Golf.

    • All vehicles had basic engine hardware, such as factory exhaust manifold, unported throttle body and so on.

    • All vehicles where driven by the tester prior to the event and on the event.

    • Dyno used is a Dyno Dynamic 450DS unit.

    • MS = MegaSquirt.

    • Two consistant tests were performed back to back on each vehicle for the posted result.

    • While care was taken to ensure the infomation seen in this thread was accurate and the vehicles and personel were safe, it should be stressed to use such information one's own risk. If you do not have the skills and blow up your own engine and damage yourself by attempting what you see here, nobody said it will be easy.

    **85% state of tune means the ECU was delivered as per factory spec, or in the case of SEMs, tuned to achieve average lambda for full power +/- 6% with WOT spark advnace set very clear of det threshold or MBT for the given vehicle weight and gearing

    Dyno Dynamics.
    [​IMG]

    I have always believed that a good OE VW 16v base engine, such as KR, ABF etc will, when optimised deliver torque and power peaks at ~5500rpm and ~6500rpm respectively. This trend of thought was slowly developed before there was dyno access and as a result of compiling data from my Gtech-Pro acceleration device over the many years of owning one.

    The torque traces on this dyno does not skew the torque readings at higher speeds as seen on other dynos visited and as torque is directly proportional to vehicle acceleration or vehicle forward force, this profile can be and has been replicated on an accelometer such as a Gtech-Pro in second or third gear. So I am confident that the torque profiles generated on this dyno as a measurement tool do correlate to real life vehicle acceleration and thus the resultant power that is reported.
    [​IMG]


    Mike's 9A MK2 on MS.
    [​IMG]
    When this car first came in to start mapping, it had numerous issues with the hardware, both electrically and mechanically.
    The drive feel was very flat and felt worse than a misfiring SDI diesel. The engine struggled to get to 6500rpm. Some recommendations were made to the garage that built the vehicle to get it to a good state of health. Once concluded tweaks where made to correct the cam timing. With the hardware now in a good baseline state, the drive map was established one evening over a few hours. The result was a vehicle that pulled much more urgently to the 7000rpm cut out. I still thought, for the time spent this power plant was not on the same level of responsiveness as an ABF unit. I was then told post mapping, that ABF cams were installed. With dyno time still required to smoothen the WOT torque curve this was the result when it was strapped down on the rolls.

    [​IMG]

    You can see that there is still work to be done at certain break points of the torque curve, yet still not a bad result for an old engine such as this and a good reference for those still wanting to use such an engine in a conversion.

    Owners comments- "I am well happy with that"

    Dav's MK2 ABF on Digifant 3.2.
    [​IMG]
    As previously said This car was the very same vehicle that part contributed to the the "How to ABF your MK2" thread" MK2+ABF or OE controls. This car still lives and is well looked after by its owner. After many years since the ABF thread, this car drives just as the first time the engine was installed. Pulls very strong from low revs but seems to struggles a bit from 6000rpm to 6800rpm cut out. With just a recent oil change and new plugs, this was the next vehicle to be evaluated after the 9A MK2.

    [​IMG]

    Like the Mk3 vehicle used in our 16v airbox study , Dav's engine, on its OE Siemens controls and cal, displayed the same behaviour and torque profile. Torque peaked at 4400rpm and as the vehicle over fuelled slightly at 5000-6000rpm there was a -ve delta in torque and finally the std set up just runs out of puff as the torque drops off post 6100rpm. A good result as this stock engine achieved 140lbft peak and 152bhp@6100rpm. On this dyno I would expect a very similar torque profile on most good working MK3 16vs. Just to remind the readers, below is the plot from 16v airbox study MK3, after some hardware test iterations.

    [​IMG]
    151bhp@6200rpm and 138lbft@4600rpm was the result back then. Note how similar this MK3 16v airbox study plot is to Dav's plot.

    Considering that this engine has a OE tune, there is more to be had from proper remapping which is hinted at in the next participant. A good power and torque reference for a typical ABF into MK2 or MK1 swap when the factory ECU and software is used. MK3 16v owners can also use this a reference for a healthy working engine.

    Owners comments - "Standard is the new modified"

    Ben S ABF MK2 on MegaSquirt.
    [​IMG]

    Ben's MK2, in its latest incarnation, brings back quite few memories. Back to a time when an 8v GTi MK2 Golf, was converted to an ABF engine, by eatonMk2 and the vehicle was handed over and calibrated live on my way to work. The addition of the home built ECU did not transform the vehicle per se, but the calibration that was delivered after did indeed make this combination of engine conversion and remapped engine parameters a most unforgettable one. I knew from K-Jet tuning that a std ABF in an MK2 can be made to be fairly angry but this car was a bit more than K-Jet angry and all who went out for a demo in this car did not believe that the engine was standard and never opened. Well that car went on to sport an Eaton M62 charger, from a W210 Merc and tested at 230PS@6500 and 0.7bar on this same GS DD dyno. However it was soon sold but the sprit of what went on back then now lives in Ben's car. After a few hours were spent to establish the vehicle's character as well as part shaking the flywheel slack on a live mapping session, Ben's car was road tuned and ready for the rolls. This was not without great effort though, but he made it in one piece.
    Having driven Mike and Dav's cars I was sure that this vehicle was a bit more edgy. So how did it fear on its road tune.

    [​IMG]

    We can see like all the previously tested vehicles, there is still work to be done to smooth out the torque profile but even in this state of tune you can see why this car is much more aggressive than the rest as the torque is higher generally and wider in its peak average. With more tuning both on rolls and road the torque can be increased from 3500rpm and also at 6500rpm as seen in other well developed examples.

    Owner's comments- " Engine sounds angry but I need more "

    Next steps
    There is another vehicle that will be visiting the rolls that could not be present due to an unforeseen reason and that is Don's GerTIe.
    [​IMG]
    This is the original K-Jet car from the "How to ABF your MK2" thread. This car has also been driven and in my opinion seems to feel as if it would slot in under the MS car but be -ve the Digifant car pre 4000rpm and be +ve post 4000rpm. Would my feelings be right? We will soon find out when this car is tested.

    Conclusion
    Here are the vehicles power/torque runs compared.

    [​IMG]

    From this it can be deduced that the best drive feel at WOT from the 3 was from the vehicle fitted with the MS and the ABF engine.

    Of course if this vehicle was fitted with another standalone controller or if the std Digifant unit was carefully recalibrated then the result would have been the same.


    Hope this thread helps to answer common questions related to these engines and put more fact where there was fiction.

    Regards
    Toyotec.
     
    Last edited by a moderator: Aug 19, 2017
  2. A.N. Other Banned after significant club disruption Dec 5th 2

    Joined:
    Oct 23, 2003
    Likes Received:
    447
    Reading through and got stuck here ^^.

    This is Dav's car, but a previous run on a different day? Clarify above?
     
  3. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    No that is the plot from the MK3 16v airbox study thread. The torque profile from that car was very close to what was seen on Dav's vehicle. It should be as both run the same OE ECU.
     
  4. A.N. Other Banned after significant club disruption Dec 5th 2

    Joined:
    Oct 23, 2003
    Likes Received:
    447
    Thanks.

    The last graph is a great illustration - I'd been looking at the horizontal torque row of all graphs up until then :thumbup:
     
  5. Ess Three Forum Member

    Joined:
    Oct 28, 2003
    Likes Received:
    84
    Location:
    Aberdeenshire, UK.
    Some cracking work there Toyotec. :thumbup:
     
  6. Mike_H Forum Addict

    Joined:
    Oct 16, 2004
    Likes Received:
    17
    Location:
    iQuit
    Nice work Eddie. I notice there's a fair difference between the ABF on MS there, and Gurds' Mk1 beastie. Is there a bit more calibration needed to get to that level, or do you reckon it's down to Gurds' Supersprint Manifold and System?

    It would be interesting to swap the calibrations over between these cars and Gurds' and see how the plots change, or do you think there are too many variables for that to be a valid comparison?
     
  7. vw_singh Events Team Paid Member

    Joined:
    Oct 29, 2003
    Likes Received:
    793
    Location:
    United Kingdom
    Great work Toyotec :clap:
    A lot of effort there just for compiling a reference :thumbup:

    Gurds
     
  8. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    Gurd's car is at an advanced level of calibration compared to the 2hrs of road time only spent on Ben's car. In fact the calibration for Gurd's short ratio'd and lighter car does not work in this heavier, CDA'd gearbox MK2. So swapping cals is out of the question as the cal is matched to the car to get the best out of the engine. In theory the engine will output the same but the control settings are different. It is for this reason the tests where limited to MK2 Golfs with similar wheel force requirements. There is a great possiblity that the extractor exhaust system on the MK1 contributed to the 160+lbft of torque achieved on the MK1. But to analyse I would have to undertake back to back tests on one known car to keep errors down to a minimum.
    There is more work to do to both MS'd vehicles particulary on the ABF MS'd car where I would expect the torque plateau to include 3500rpm and increase on amplitude. I would need dyno time to fix.
    What is not shown in these the torque plots is the time to max torque whe the accel pedal is pushed. Once again the MS'd car was the most aggressive followed by the Digi car and then the 9A version. Hopefully the K-Jet will be ready soon and the whole thing can be concluded properly.
     
  9. paultownsend Forum Member

    Joined:
    Jun 23, 2006
    Likes Received:
    0
    looks like my mk2 abf digi 3.x is pulling healthy std figures on a similar D_D machine. Albeit with a "2 jetex system and no cat
     
  10. paultownsend Forum Member

    Joined:
    Jun 23, 2006
    Likes Received:
    0
    infact that was well over a year ago. home service today, and booked in for a D_D shakedown tomorrow on the same machine at 3.

    head has since been rebuilt and seats re-cut. a challenger to the gauntlet
     
  11. Mike_H Forum Addict

    Joined:
    Oct 16, 2004
    Likes Received:
    17
    Location:
    iQuit
    I had a funny feeling it might not be that straight forward. Might still be an interesting experiment, to see how transferable the results are from Gurds' map. MS certainly transformed my car, and probably still has potential for more, with tuning of the cam timing. It'll be interesting to see what it does on the rollers.

    I notice that the tested cars were on standard manifolds. Were they on std exhausts too, or at least similar aftermarket setups?
     
  12. paultownsend Forum Member

    Joined:
    Jun 23, 2006
    Likes Received:
    0
    did ess_three's work not prove that de-catting and 'performance' exhausts contibuted little to abf output? mine certainly proves that.

    if everyone interested in abf development wanted to donate 5, id quite happily use my engine as a development car. proceeds would be forwarded to mr_hillclimber on exchange for a worked abf head and matched manifolds [:$] and then on to digifant calibration.
     
  13. Ess Three Forum Member

    Joined:
    Oct 28, 2003
    Likes Received:
    84
    Location:
    Aberdeenshire, UK.
    On a set ECU calibration, yes.
    I did not try recalibrating afterwards, so although gains are possible there, it proves that the mythical 5-10 BHP the exhaust sellers try and tell you you'll get by bolting on a noisy exhaust, just aren't there...

    I've tried two (Milltek & Supersprint) and several cat combinations (de-cat, Milltek 100 cell cat for a Corrado VR6, and a std ABF cat) and never found a tangible, repeatable gain any which way.



    Or people could sent it to Herr Toyotec in order to bribe him into mapping my ABF and we'd try std cams vs Schrick 268s vs Scrhick 276s vs all combinations of the aforementioned, with intercam verniers fitted to allow cam swinging, and cam to crank swinging, and 4 vs 8 injectors as well as inner only vs outer only. :thumbup:

    All in the name of ClubGTI science, of course.

    I'm even prepared to make the journey from the frozen north to the big smoke for the occasion.
     
  14. A.N. Other Banned after significant club disruption Dec 5th 2

    Joined:
    Oct 23, 2003
    Likes Received:
    447
    The idea of development funds has been discussed briefly before and the issue is cost, particularly rolling road time.

    However, if folks are interested in pooling towards such work (generically, not necessarily this one) CGTI can hold funds to allow development sponsorship like this to progress. A budget would need to be decided, contribution per participant decided and then a set time for monies to be collected. Once in, the project is green lighted, CGTI hands over the funds on a strict receipt basis, so everyone knows where their funds went. For the good of all, we couldn't have a situ of uncertainty there.
     
  15. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    As hinted before a less aggressive iteration of vw_singh's cal was flashed to this car, as an experiment, and it caused det issues at WOT and certain transients. So the latest MK1 cal would be even worse. The cal for this car set-up is custom and would be even more when the final release is completed.

    Dav's car has a STD VW 8V exhaust system and a factory cast manifold.
    Ben's car has a STD VW 8v exahust system and factory cast manifold.
    Mike's car has a backbox, unknown centre section, OE 16v downpipe and cast manfold ABF cams, but 051***373 head.
     
    Last edited: Oct 7, 2010
  16. Ben S

    Ben S Forum Junkie

    Joined:
    Aug 1, 2005
    Likes Received:
    346
    Location:
    Essex
    Mike, I can only speak for my vehicle, which is as follows:

    Corrado airbox, then everything from intake duct to exhaust manifold is standard 1996 ABF from a Mk3 Golf 16V. Exhaust downpipe & exhaust system are standard 90 spec 8V items

    Calibration needs more work and an hour or so on the dyno, but is great for everyday use at present.

    not going to change any components in near future as there have been a few teething problems, this build was initially put together in feb / march but has been taken apart and put back together twice since due to various reasons.
     
  17. Mike_H Forum Addict

    Joined:
    Oct 16, 2004
    Likes Received:
    17
    Location:
    iQuit
    Ah, missed that bits about the calibration. Thanks.
     
  18. badger5

    badger5 Club GTI Sponsor and Supporter Trader

    Joined:
    Oct 22, 2003
    Likes Received:
    15
    Location:
    nr Glos
    some nice, sensible comparisons there..
    I am wondering why the torque of all of these looks so smooth, without the 16v rise then dip then rise again they all normally show? Is this a dyno thing?

    The plots I see are quite different.
     
  19. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    I would not say the torque curve is smooth as there are 'holes' in the profiles, something that would require development time to rectify. The dyno software does take out the curve jaggedness seen in some machines. As said I can back those traces up with an accelerometer in 3rd or 2nd. Plus having driven all vehicles at WOT from 2500rpm they do feel just as measured on this device.
     
  20. danster Forum Addict

    Joined:
    Apr 2, 2008
    Likes Received:
    15
    Are you using a 90spec mk2 8v Digifant downpipe or 90spec mk3 8v Downpipe?
    Also, have you got a Cat fitted?

    The longer secondaries of the mk2 8v Digifant downpipe may be helping. It is certainly the done thing in a few one make race series I am aware of to have good long secondaries on the downpipe.

    This back to back test does show that there can be large differences in performance and output even whilst running similar specs.
     

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