35 real world mpg from a MK3 VR6...No? The Toyotecwerke economy Ankor w/up to 38mpg.

Discussion in 'VR5, VR6 & Wx' started by Toyotec, Jan 8, 2011.

  1. tshirt2k

    tshirt2k Forum Junkie

    Joined:
    Oct 21, 2004
    Likes Received:
    325
    Location:
    Herts
    Not when the sag causes movent of the wheel when jacked up. Could cause an MOT failure or at least an advisory!
    Those Top mounts don't tend to spring back when there's that much sag on them. Having the same problem with mine at the moment. And my engines not as heavy as the ankor, i wouldnt have thought. Need to find some strong enough.
     
  2. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    At least it's a Crossflow :lol::lol:.

    Hey do not laugh at my magic wand!!!:lol::lol:.
     
  3. Rayhoop Paid Member Paid Member

    Joined:
    Aug 21, 2006
    Likes Received:
    69
    Location:
    Amersham, Bucks
    Ive also found the MFA mpg reading to be quite accurate - tested against the 'brim tank, drive lots till empty, brim tank and calculate' method.

    My ECU is standard (as far as I know), and the only mods are 17's, lowered (60/40) and a few holes in the airbox. The best 'average' I have achieved is 37mpg on a return run. From Amersham to Wembley Ikea and back on a weekend!

    I find 'return runs' more accurate at recording a reliable average mpg reading, as any inclines/declines on the way out are cancelled out by running the same route on the return run. As when I was commuting to Portsmouth, I could get close to 40mpg on the way to Portsmouth (early run, so minimal traffic and very careful driving), but the average dropped to around 35mpg after the return journey (roughly similar traffic, and the same route home). I put this down to the fact my home is approx 150m above mean sea level... my old work place was only a couple of metres above mean sea level on the coast of the Solent.
     
  4. dUff

    dUff Administrator Admin

    Joined:
    Oct 22, 2003
    Likes Received:
    246
    Location:
    Everybody Offline - looking good
    Me and a mate both used to have VR`s i did like mine for driving to work , i had a K plate one with coil packs and my mate had a P reg one with about half the miles on it . In back to back tests his would get over 40mpg where as mine would only get 36mpg . we assumed this to be the OBD1 and OBD2 differences , but his car was a very late model as some of the parts are in mk4 golfs !

    Big engine cars do take a little time to get used to , very small prods of the accelerator do yield good results and just need to use the toque of the engine

    Also the MFA if used for a journey with hills and flats gave me different results in different directions , driving the same style and in at the same time of day , so driving style the same. That may be worth bearing in mind

    Good work tho !!

    Could you do something similar with a 5 cylinder Audi derv engine ?
     
  5. corradophil Forum Member

    Joined:
    Mar 26, 2004
    Likes Received:
    15
    Location:
    Essex
    As a control it would be good to calculate the MPG using the brim to brim method above. I say that because on my Mk4 Golf 1.8t, the stage 1 remap caused the MFA reading to go from around 31mpg standard to 34mpg. Since then it is stage 2 and now regularly shows 40mpg over a 20 mile journey on back roads and B roads. I drove from Leicestershire to Essex and the MFA was reading 51mpg when joing the A127 coming off the M25. I suspect it was fairly accurate when standard, but now is over reading. I am planning on filling it up with petrol tomorrow, after noting the milage and reseting the MFA on the last fill up to calculate the real mpg to compare. The long term reading since the last fill up is 36mpg. From what I have read 'some' Mk4 instrument clusters have a calibration setting to adjust the mpg reading, so depending on the outcome tomorrow, I may be looking into that.

    On the same subject my Corrado with a 9a shows around 44mpg since tweaking the fuel pressure to alter the fuel mixture. It was running very lean, and I have now set it up so the AFR is between 12 and 13 at WOT. Mine has Ke-Motronic with closed loop lambda for part throttle and the AFR gauge shows the part throttle AFR fluctuating between just over 14 and just over 15.
     
  6. Matt82

    Matt82 Forum Addict

    Joined:
    Apr 26, 2006
    Likes Received:
    8
    Location:
    GTI Scene
    from my experience the mk3 mfa is really accurate, those are really quite astounding figures.

    mpg on my mk3 is down a bit lately, new lambda hopefully may cure that.
     
  7. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    "OBD 1/ 2" has nothing to do with any of these European applications. The later M3.8.1 from August 95, does have superior fuel adaptation to M2.9.1 and the very course in M2.7. Also the constant for fuel consumption in the ECU maybe calculated differently for early cars, which had a different make of cluster too. Still all systems can be made to work such that the drive would not be able to tell the difference at either part throttle or at WOT.
    Improvements can be made to any ECU, once you have the resource and know how to extract information from them, understand the control logic and use that hard information to make changes and update.

    I am told your feet are very very light and that you can turn air into fuel. Maybe you need to borrow mine and see if you can get it up to 50mpg. It is a bit snappy though and that might tempt you into less!!

    I have done that before and found it to be fairly consistant with real life. If it was not I have access to the ECU constant that ouputs to the cluster and would have already addressed.
    In MK4s both Pre EOBD (M3.8.3) and ME7 , there are also settings in the ECU map to tweak for fuel flow changes. This may have been left as per std map.


    As my car has a catalyst, the ECU is designed to run λ ~ 0.95 to 1.05 as std in part load and idle. This was always the case. There were many sublte tweaks that were made to the ECU to achieve better FE. CIS-E and KE_Motronic systems also try to target λ ~ 0.95 to 1.05 , but as you found out are dependant on the accuracy of the mechanical metering components and exhaust heated sensors.
     
  8. corradophil Forum Member

    Joined:
    Mar 26, 2004
    Likes Received:
    15
    Location:
    Essex
    Just filled up my Stage 2 1.8t Mk4. Over the last 267 miles the mfa read an overall average of 36mpg, when in fact it actually was 30.6mpg. I'd need to drop the reading by 15% before it becomes accurate.
     
  9. Matt82

    Matt82 Forum Addict

    Joined:
    Apr 26, 2006
    Likes Received:
    8
    Location:
    GTI Scene
    mk4 mfas are a pack of lies

    380 miles on a tank = c.34mpg which is a real rarity.
     
  10. azur Forum Member

    Joined:
    Nov 9, 2008
    Likes Received:
    0
    Location:
    Sussex
    I miss my mk3 and its fuel economy... my 16v used to get over 40 sometimes.

    The spec B gets 25 when you drive like a granny, 22 when you drive like your mum and sub 16 when I drive like a loon [:s]

    wish I could afford to insure the valver as well but insurance has gone crazy. Might have to sell her soon [:^(]
     
  11. aidanb22 Forum Member

    Joined:
    Feb 14, 2006
    Likes Received:
    0
    Location:
    Redditch, Worcs nr Birmingham
    In theory, could a turbo VR6 better those figures? What if it had tdi-style gearing, so 60mph was around 2000rpm?

    Turbo in this case as if it was sized like a K03 is for a 1.8t, for normal driving not max bhp at top end!
     
    Last edited: Jan 27, 2011
  12. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    A turbo VR6 would have a big block in the intake path ( compressor) and exhaust (the turbine) so slighty worse, all other things being equal. Same could be said for a VR6K which would have larger accessory losses and a compressor blocking the intake system. I had a VR6K that was being road tuned. On the same run as the NASP car it would achieve up to 28mpg. The level of calibration was not as mature as found on mine, so it could have achieved more say 30-31mpg. I would expect a turbo version, based on injector size and turbo choice be around this level of mpg.
    Lower revs will help, but you do not want to start lugging the engine where the axle torque required to heave the chassis along is less that the external forces opposing it. This will also hurt FE.
     
    Last edited: Jan 27, 2011
  13. Rayhoop Paid Member Paid Member

    Joined:
    Aug 21, 2006
    Likes Received:
    69
    Location:
    Amersham, Bucks
    I could road test your car on a commute to directly compare against mine using the same route and driving style. Will explore the options - probably wont be possible till end of Feb/early March, as I'm a bit tied up at the mo.

    You're right though - the extra 'zoom zoom' could make it tempting to prod on a little more than required!


    Maybe I could be the Eco-Stig?
     
  14. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    MFA and calculated FE compared. Up to 38mpg achieved!

    In post 1 it was demonstrated that 35+ mpg could be achieved with a Golf Mk3 VR6. This achievement was based on careful driving, ECU calibration changes and looking at the final reading of the multi function analyser (MFA), after a full trip. It is believed by some that the MFA is pretty accurate in determining both instantaneous fuel economy (FE) as well as fuel economy over a longer distance. To validate this as well as add further credibility to the FE achieved with this car so far, more tests were done to performed over a few drive cycles to repeat the last best of 35mpg, this time by observing not just MFA readings over a full trip but also recording the amount of fuel used for miles driving. The aim was the amount of fuel used per mile should tally with the MFA if the analyzer was calibrated correctly.
    A full trip in this draft refers to commute to and from work including cold start and engine off for more than 8 hours. The MFA for such a trip would have reset automatically after 2 hours. A full trip is about 50miles.
    Odometer was reset for refuelling. Engine was hot so MFA was not reset.
    Fuel gauge was on white mark or empty both when first refuelled for the test and at the end. There maybe less than 6 litres of fuel available when the gauge is in this position.
    Golf hatchbacks carry a 55 litre tank including a 7 litre reserve (pp 188 G3 onwer's manual). So far the most amount of fuel ever filled when the gauge has settled at this point was 49.5 litres.

    The Golf VR6 2.8 174PS was advertised as having and average FE of 10.5 l/100km (26.5 UK mpg) over urban and extra urban European Drive Cycle. This can improve somewhat as the vehicle gathers more miles.
    A 150PS 16v MK3 by comparison was advertised at 8.5 l/100km (33.23 UK mpg)

    To burn off the remaining fuel in the tank at the time, a prep run was done on the vehicle, which involved driving carefully on a return trip from work to home. During this return trip fuel economy measurements of up to 38.1mpgs were observed.
    [​IMG]
    [​IMG]

    However as the traffic thickened the final FE for that trip was 37.5mpg. The following morning with the vehicle cold, the journey to the petrol station, some 25 miles away resulted in ~36mpg. The gauge was on the white mark of empty when this was done. The amount of fuel that was dispensed was 26.02 @ 1.29 litre. After refilling the gauge settled to the mark just before half. As this was the beginning of the test the odometer was set to 0 but not the MFA. 25miles were covered before switching off for this part of the commute. During this journey the MFA measured up to 36mpg on some cruises. However congestion did take its toll and the final mpg with the engine off was 35mpg. More than 8hrs had elapsed before the return journey of 50miles was commenced. During this soak time the engine oil, coolant and air path would have equalised with its ambient surroundings. Surroundings that would have been ~ 10 deg C. After the return journey the final reading was 37mpg with up to 38mpg mpg recorded on the journey. Similar was recorded on the commute to work the following morning. So far 125.8 miles were covered on 26.02 that was put in the empty tank a day before.
    [​IMG]
    [​IMG]
    [​IMG]

    When the returing to the car to start the journey towards home it was first thought at this consumption rate the fuel gauge looked as if at least another 50 miles could be squeezed, before the fuel gauge needle would return to where the vehicle would need to be refuelled again.
    [​IMG]
    This was not to be as the route chosen was blocked by a major traffic jam. FE just before stop start traffic was already 36mpg.
    [​IMG]
    A diversion through urban area was taken which added miles to the final destination. This meant refuelling took place earlier than expected. In the end the average FE when stopped at the petrol station was 34.6mpg and the final miles travelled for 26.02 was 158.3
    [​IMG]
    [​IMG]
    [​IMG]

    So how accurate is the MFA?
    Well some calculations were done and shown in the table.
    Please note;
    • It is assumed that on each empty ~ 5litres were still left in the tank

    • MFA readings are directly related to fuel flow from std injectors and ECU mph from speedo sensor.

    Actual FE

    Pence/litre Amount spent Total UK gallons from litres Distance travelled in miles mpg UK l/100km mpg US
    129.9 26.02 4.406 158.3 35.93 7.86 29.92


    Using MFA readings to work out FE for 158.3 miles.

    Journey discription and miles MFA indicated mpg end of trip
    25.8 miles from petrol refill 35
    50 miles to home 37.4
    50 miles to work 37.1
    32.5 miles towards fuel stop 34.6
    Average FE (UK) for 158.3miles 36.03 (less than 3% difference from actual)

    My conclusion is it is pretty accurate.

    This testing demonstrates that is is possible to achieve frugal fuel consumption from what is considered to be a gas guzzler. During this exercise it was observed during traffic condtions where vehicles speed would tansistion between idle and 0 mph to 30mph, before achieve steady state conditions up to 0.3 mpg could be lost. It is thought that the remapped idle speed of 800 rpm ( STD 700 rpm AC/OFF) could be reduced to 650rpm in an attempt to reduce fuel flow at idle. Also for cold starting the cat heating function where spark is retarded and the engine run lean could be reduced or switched off.
    We will see how this develops.
    Thanks for reading.

    Regards
    Toyotec.
     
    Last edited by a moderator: Dec 15, 2013
  15. Rayhoop Paid Member Paid Member

    Joined:
    Aug 21, 2006
    Likes Received:
    69
    Location:
    Amersham, Bucks
    Wow - that is a good result, and with the potential for more by the sounds of your suggestions of the ecu tweaks. Does your route to work take the A406 or the M25? Either way, negotiating commuter traffic and getting those mpg is quite a result!
     
  16. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    Both + a bit of slow B road.
     
  17. prof Forum Addict

    Joined:
    Oct 23, 2003
    Likes Received:
    18
    Location:
    Suffragette City
    interesting findings,
     
  18. Neal H Forum Member

    Joined:
    Oct 18, 2007
    Likes Received:
    5
    So, just to add to this thread, on Friday I did a 180 mile run in my Corrado VR, from cold, with a brimmed tank and properly inflated tyres I had an indicated 39.2 MPG on a 180 mile run. This was done on a short stretch of urban roads, A roads, dual carriageway and motorway. Crusing speed was 70 mph and the average speed for the full journey was 52 mph.

    I will do another cold start this afternoon and re-brim the tank (the only real-world way to accurately check the total amount of fuel used) and confirm the real world MPG and the accuracy of the MFA on this particular example.
     
  19. Neal H Forum Member

    Joined:
    Oct 18, 2007
    Likes Received:
    5
    So, once the tank was brimmed again, the check on the MFA comes to 1 mpg less than indicated, so a calculated 38 mpg. Not too bad for a 15 year old 110k mile 2.9 I think. Unfortunately tihs is an old version of the ECU otherwise it would be over to Toyo for an eco-map and we could see some really good back-to-back testing on a different version.
     
  20. Toyotec

    Toyotec CGTI Committee - Happy helper at large Admin

    Joined:
    Jul 26, 2006
    Likes Received:
    3,320
    Location:
    Creating Pfredstarke
    That is some good going. The longer steady state speeds do help with economy. Goes to show VR6 engines are not that thirsty if driven properly over various cycles.
     

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