1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Hybrid turbocharger k04 build

Discussion in 'A3/S3 Forum (8L Chassis)' started by beachbuggy, Nov 25, 2011.

  1. Prawn
    Offline

    Prawn My other car is a MINI!!!!

    Joined:
    Dec 8, 2006
    Messages:
    9,439
    Likes Received:
    1,281
    I'll do a 3rd gear run and use the spreadsheet to work out a FATs time this evening Doug.

    Mine's an ENJ 02J 5 speed on 225/45/17's with what you guys on vortex would call a 'stock block tune'!

    I imagine being a 5 speed the FAT's time will be considerably higher. Do any of the guys on vortex do FATs with the 5 speeds and F23 combo?
  2. badger5
    Online

    badger5 RIP S3dave Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,262
    Likes Received:
    668
    indeed its a solid gauge on same car same conditions.... Remembering overall weight on board is more of a factor than might be firstly apparent.. (I know our racecars I am involved with, weight is by far the biggest differentiator)

    handy for real world comparisons...
    luckily I have a dyno to see what works for me.. strapped down doing 150mph stationary.

    road logging is how I used to have to do stuff but a 450bhp polo made my mind up, it was daft... and only going to end in tears at some point.
  3. _G_
    Offline

    _G_ Member

    Joined:
    Jul 24, 2011
    Messages:
    872
    Likes Received:
    34
    bump - k04 hybrid project started! parts on order.
  4. Prawn
    Offline

    Prawn My other car is a MINI!!!!

    Joined:
    Dec 8, 2006
    Messages:
    9,439
    Likes Received:
    1,281
    Just been out logging for FATs times to see how they compare to the 6 speed, and I think it's safe to say it's wildly different! haha.

    Best FATS recorded was 4.64s, and 6.11s on actuator pressure only.

    What sort of speeds is the FATS test from on a 6 speed? on my 5 speed it's from 65-101mph. I'd imagine the 6 speed test is at a much lower speed.
  5. Dani_B19
    Offline

    Dani_B19 Audi-sports own special child.

    Joined:
    Nov 9, 2008
    Messages:
    7,270
    Likes Received:
    90
    Can someone explain to me and karl what FATs are? Were only mere grease monkeys haha.
  6. superkarl
    Offline

    superkarl MAN OF STEEL

    Joined:
    Mar 28, 2010
    Messages:
    7,553
    Likes Received:
    1,074
    we are not worthy dani. lets go chew on a spanner
  7. Prawn
    Offline

    Prawn My other car is a MINI!!!!

    Joined:
    Dec 8, 2006
    Messages:
    9,439
    Likes Received:
    1,281
    It's a performance measurement Dani, rather than relying on engine parameters to measure performance, this measures real world acceleration on the road using vagcom.

    FATs time is the time it takes to accelerate from 4200-6500rpm in 3rd gear.

    The little green spreadsheet above is used to work it out using vagcom log data, because recordings arnt always taken at exactly 4200 a d 6500, so it works out the time stamp at those points using the closest values either side.

    Because it measures in gear acceleration, it's only really comparable across cars with the same transmission, hence me asking about speeds at each rpm point. As I imagine on a 6 spd it'll be much lower speeds than the 65-101 on my 5 speed, sometimes are. Of comparable :)
  8. superkarl
    Offline

    superkarl MAN OF STEEL

    Joined:
    Mar 28, 2010
    Messages:
    7,553
    Likes Received:
    1,074
    similar to doing say 30-100 times or whatever.
    surely they would be better gauge of performance, as you are using more gears.
  9. Dani_B19
    Offline

    Dani_B19 Audi-sports own special child.

    Joined:
    Nov 9, 2008
    Messages:
    7,270
    Likes Received:
    90
    Thanks for that nick, didnt know you could do that with vag com.
  10. Prawn
    Offline

    Prawn My other car is a MINI!!!!

    Joined:
    Dec 8, 2006
    Messages:
    9,439
    Likes Received:
    1,281
    Just needs a log with a time stamp either side of 4200 and 6500, and of course the RPM.

    Chances are, if you do your logs in 3rd gear usually, you can already calculate the FATS from an old log

    Mark's S4 imolavant - FATS Calculator
  11. Pedrosousa
    Offline

    Pedrosousa Member

    Joined:
    Feb 11, 2011
    Messages:
    162
    Likes Received:
    1
    My car does the same that this is doing when the boost comes up at 3250Rpm the AFR goes lean, is this normal?
    I tune fuel with Lamfa, it's a narrowband ECU...
    In my case it doesn't goes to lean, because the fuel it's requested to be at 12.6 before the 3000 rpm zone... So I get +-14.00 when the boost comes up...
  12. slappy_dunbar
    Offline

    slappy_dunbar Let's Do This

    Joined:
    Jan 2, 2010
    Messages:
    229
    Likes Received:
    11
    Based on the gearbox ratios, I show the 6spd in 3rd gear running from 55.5mph to 85.9mph. That's a good bit different than with your car. Maybe you should run in 2nd??
  13. beachbuggy
    Offline

    beachbuggy Active Member

    Joined:
    Jan 27, 2010
    Messages:
    1,755
    Likes Received:
    139
    So quick update. New 1bar actuator was fitted yesterday. Basically I bought a T3 actuator , remove the old one from the stock s3 bracket and welded the new one in place. 3 runs and the n75 was tuned to cope with the extra tension and now I can hold 23psi to the red line! Not only that the boost is definitely more aggressive I assume as the waste gate is being held full close rather than blowing open . So running today knocking a few degrees off the timing I managed 263grams, which I am very happy with. Work permitting I'll get down the RR next week to see how it goes.
  14. superkarl
    Offline

    superkarl MAN OF STEEL

    Joined:
    Mar 28, 2010
    Messages:
    7,553
    Likes Received:
    1,074
    excellent.
    whats the full spec of your car?
  15. beachbuggy
    Offline

    beachbuggy Active Member

    Joined:
    Jan 27, 2010
    Messages:
    1,755
    Likes Received:
    139
    Apy engine
    BAM ECU/loom
    ie forge rods
    xs power exhaust manifold
    3" blue flame dp/200cell cat and cat back
    b5 v1 tip
    welly cooler
    hydrid rs6/2283 ( unclipped) k04 turbo with uprated actuator.
    Blue haldex
    blistein springs and shocks
    brembo 4 pot brakes
  16. slappy_dunbar
    Offline

    slappy_dunbar Let's Do This

    Joined:
    Jan 2, 2010
    Messages:
    229
    Likes Received:
    11
    Hah! Just goes to show you that an un-molested and un-clipped RS6 turbine rotor works just fine. So let's see some log graphs! What's your boost onset in 3rd? Airflow curve? :uhm:
  17. beachbuggy
    Offline

    beachbuggy Active Member

    Joined:
    Jan 27, 2010
    Messages:
    1,755
    Likes Received:
    139
    Here you go.
    1) is the maf. as you can see at around the 4750 mark there is a real blip in the maf, this as you can see in 2) is at the point the cam changes from an advance of 23 degrees down to 1 degree..
    3) is the boost map. I have a slight over shoot but it's capped with a MBC at 23 psi ( this logger only goes to 22.5psi hence the flat line) and I found if I go over 21 psi below 4250 rpm the thing will surge if I keep it below that until around 4500 then I can ramp it up to what ever I want with out surge.

    As I now have an uprated actuator if I try and aim for 21psi, the KFLDRL map will always cause it to over shoot and surge because the N75 isn't quick enough. So I've set load map for around 18psi which means that when it tries to react to hold 18psi it will overshoot but stay within the surge limit 21psi and thus by the time is controls it,it'll be at 21psi anyway, if that makes sense?. No need for a clipped turbine. After 4750rpm I'm free to ride the PSI rodeo all the way to the redline..yeeehaaaa :)


    [​IMG]
    [​IMG]
    [​IMG]
  18. Chlippo
    Offline

    Chlippo Member

    Joined:
    Apr 17, 2010
    Messages:
    667
    Likes Received:
    12
    Impressive again! I think that answers my question about the mbc and the N249 bypass.

    Can't wait to start working on mine.
  19. courty89
    Offline

    courty89 New Member

    Joined:
    Jan 27, 2013
    Messages:
    3
    Likes Received:
    0
    had a good read through this thread and fancy doing this to my partners TT, any news or updates at all?
  20. slappy_dunbar
    Offline

    slappy_dunbar Let's Do This

    Joined:
    Jan 2, 2010
    Messages:
    229
    Likes Received:
    11
    Dan -- it'd be interesting to see your airflows compared when you have the intake cam "locked" into the two positions. I came away finding that the VVT system offered no benefit on the cammed FrankenTT. But that's a smaller-wheeled hybrid.

    Here's an example of how I logged the two settings:

    [​IMG]
  21. Brownie888
    Offline

    Brownie888 New Member

    Joined:
    Jul 10, 2013
    Messages:
    3
    Likes Received:
    0
    Thanks for such an intresting read :) looking at doing this myself as my turbo has just started smoking on hard boost. The only bit im woundering about is the machineing, as i dont have the bits in my hard i cant see but does the profile need reshaping aswell or can you get away with just turning it out?Thanks chris
  22. MIGGZ
    Offline

    MIGGZ nanananananananan BATMAN

    Joined:
    May 17, 2012
    Messages:
    1,118
    Likes Received:
    157
    just read thru this and very interesting build!
    did you not need to change inj or was the oem inj good enough?
  23. BoostHard
    Offline

    BoostHard New Member

    Joined:
    Mar 17, 2013
    Messages:
    9
    Likes Received:
    0
  24. BoostHard
    Offline

    BoostHard New Member

    Joined:
    Mar 17, 2013
    Messages:
    9
    Likes Received:
    0
    Just spoke with Dan and now have an answer to the above.
  25. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    Struggling with controlling boost, seeking help here:
    Even so the car is NOT an A3 (2003 A4 1.8T), here is the setup:
    Turbo: K04 turbine housing, machined for RS6 turbine wheel.
    Compressor: RS4 housing, custom fabbed to fit 2283 billet wheel
    550cc Bosch EV14
    High flow log manifold
    3" DP
    S4 MAF
    FMIC
    SAI, N249, N112, EGR/Kombivalve, removed, tuned out.
    Wastegate actuator: stock, starts to move at 4.5 psi
    Goal: constant 22 psi (MAP limit) to 7250 redline
    Tune: we were running into the 'calculated' EGT cut embedded into the Bosch/Audi ECU. Instead just blindly eliminating the threshold, we decided to measure and log the actual EGT at the manifold. Result: showing about 200 deg C lower actual than the calculated ones.
    Consciously raised the EGT threshold so it does not 'cut in'.

    The problem:
    a) N75 control not smooth at high rpm. Tried a few N75 (factory versions), but no significant change.
    b) switched to 'ball and spring' (intergrated engineering) MBC. Now experiencing: when MBC set to 22 psi, boost falls to 18 psi after 4500 rpm. When set to 25 psi (gauge), falls to 21 psi after 4500 rpm. Curves are almost parellel. Why are we 'overboosting' mid range (or NOT holding boost at high rpm)?

    Any suggestion how to achieve our goal to hold 22 psi to redline without too much overboosting (stock rods) at mid range?
  26. Chlippo
    Offline

    Chlippo Member

    Joined:
    Apr 17, 2010
    Messages:
    667
    Likes Received:
    12
    You need a good custom remap and you will be good to go even without mbc.
    You are probably hitting pid limits and some maps would need to be modified so the ecu can accept that much boost without interfering.
    Other thing that can cause that is boost variation.

    My info is based on research, I have somehow a similar prob.
  27. badger5
    Online

    badger5 RIP S3dave Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,262
    Likes Received:
    668
    what ecu?
    whats being used to adjust maps?
  28. badger5
    Online

    badger5 RIP S3dave Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,262
    Likes Received:
    668
    K16,7+7 BBT K04 hybrid, uprated actuator, Badger5 Ported to match the high flow exhaust manifold..
    [​IMG]

    Good work Dan
    jitvv likes this.
  29. Chlippo
    Offline

    Chlippo Member

    Joined:
    Apr 17, 2010
    Messages:
    667
    Likes Received:
    12
    ME7.5

    whats used...well mostly old stuff softwares and demo versions... (u probably know which). I have no access at all to a custom map or a tuner that does those ecu so when left on my own, I manage with what I have, which is un-cool lol

    please do correct me if my info about a custom remap needed for the member above is wrong.
  30. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
  31. StaceyS3
    Offline

    StaceyS3 Active Member

    Joined:
    Dec 17, 2010
    Messages:
    2,108
    Likes Received:
    184
    Could pressure drop accross the cooler create that variance up top (assuming your using OEM positions) as boost control MBC/n75 and map sensor/boost gauge will see different relevant pressures?
  32. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    ECU: 8E0 909 518 AF

    Not sure what my tuner friend is using. But he says PID from scratch would require a TON of work.

    Respective data files from the logs shown are here:

    N75 swing:
    https://www.dropbox.com/s/vuqlb4ryay57vnh/Kody 518AF vt_20140308_215712 3rd.csv

    MBC "up":
    https://www.dropbox.com/s/gsb71oogsvwhwm2/Kody mbc up 3rd518AF vt_20140315_180327.csv


    MBC "down"
    https://www.dropbox.com/s/j1ntjfhbyxydoo2/Kody 3rd MBC down 518AF vt_20140316_000226.csv
  33. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    Both (gauge and MAP) would see losses from FMIC. MAP should read higher (if any) since it does not see potential loss from TB (gauge tap is in manifold, MAP is before TB). MAP is limited to reading 22 psi (that is why the logs are 'cut off'); need to 'visually' extend the actual boost pressure beyond the 22 psi 'cut' line.
  34. badger5
    Online

    badger5 RIP S3dave Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,262
    Likes Received:
    668

    If you are trying to sit closed loop control at limit of map sensor, I would suggest a different approach.
    Request looks to be high... , well at the limit of map sensor..so.. I would be looking at the actual N75% control map(s) personally.

    message me for more info if you want it.
    cheers
  35. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    Some suggested stiffer WG actuator spring (or additional external spring). Possible that WG is partially pushed open by back pressure at high rpm, causing the MBC to NOT hold pressure?
  36. beachbuggy
    Offline

    beachbuggy Active Member

    Joined:
    Jan 27, 2010
    Messages:
    1,755
    Likes Received:
    139
    Stock actuator is too weak to effectively control the boost, maxing the n75 dc and waste gate blowing open are conspiring against you.

    You are right , rewriting the the whole PIDS for an uprated actuator is a pig, but you do need a stronger waste gate.

    Then as said controlling n75 directly is the better/best option to get the boost profile you want...
  37. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    That much back pressure form exhaust valve to turbine housing? Wow. Not sure what the closing force is of the stock actuator spring, but for example purpose: 1 square inch wastegate area, 10 lbs of closing force: back pressure must exceed 10 psi.

    What spring rate (crack open pressure) should we shoot for the actuator revision?

    In this context: what IS the spring rate (in N/mm or lb/inch) of the stock actuator?
    What spring rate (or spring) if adding an external spring?
    Are there any 'donor' actuators with higher (desired) spring rate to 'steel' (make/model/year/engine), even if 'adaption' (bracket, etc) is needed?
  38. badger5
    Online

    badger5 RIP S3dave Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,262
    Likes Received:
    668
    if I read the log posted correctly tho the n75% is hovering aroubnd 70-80%? for the 2550mb level?

    Did I read it right?


    Re Actuators, on k04 hybrids.. I prefer to sit around 12psi crack pressure... as a preference.
  39. StaceyS3
    Offline

    StaceyS3 Active Member

    Joined:
    Dec 17, 2010
    Messages:
    2,108
    Likes Received:
    184
    Sorry was wrote in a bit of a rush and wasn't that clear, but meant that the boost control IE the MBC could be seeing higher boost pressure pre cooler than ECU via MAP sensor and visually on the boost gauge after the cooler
  40. vtraudt
    Offline

    vtraudt New Member

    Joined:
    Apr 4, 2011
    Messages:
    23
    Likes Received:
    0
    Not sure what actual N75 maps are, but here are the 2 N75 DC parameters I logged (before and after linearization) if that helps.

    https://www.dropbox.com/s/c6qbe2e30lvlmf7/Kody 0308 215712 log 3rd N75 swing with duty cycle.JPG


    Here is the boost and N75 DC for the N75 log:

Share This Page