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

Boost leak testing

Discussion in 'Engine, Drivetrain, Fuel and Exhaust' started by alexanderfoti, Apr 21, 2017.

  1. alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Hi all

    My 221 does not make the proper amount of boost through the rev range.

    Some thoughts are that there is a boost leak and I am trying to test for this.

    I am using a boost tester which is is basically this:

    [​IMG]

    I have 2 of them (one for each turbo) but am only pressurising up through one.

    I found a few leaks, but something struck me as odd. All the info says to start with the compressor set to 5psi and then work your way up to full boost pressure. I could not build any pressure in the inlet side of the system until I turned the compressor reg up to 80+ PSI. I Initially think this is due to the restriction that the tyre inflation valve provides.

    Even then, I could only build 10Psi in the inlet system. (I had a few leaks but nothing gaping!).

    I believe a few one way valves are leaking as well, so have ordered replacements.

    Obviously some air will leak past the throttle body, but is this normal? I though this is probably due to a large volume of air being required, but not so sure now.
     
  2. V6Matty

    V6Matty Senior Member

    Messages:
    4,512
    Likes Received:
    39
    Joined:
    Aug 1, 2012
    Location:
    Newark, Notts
    Your Mercedes:
    Peugeot 508 RXH (His) W169/2005/A180 (Hers)
    Would it be possible to do a smoke test on your Alex maybe with a fluorescent die in?
     
  3. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Not sure, hadn't thought of that! I am doing this outside, so would probably have most of blow away :p
     
  4. V6Matty

    V6Matty Senior Member

    Messages:
    4,512
    Likes Received:
    39
    Joined:
    Aug 1, 2012
    Location:
    Newark, Notts
    Your Mercedes:
    Peugeot 508 RXH (His) W169/2005/A180 (Hers)
    Doh, i always forget your working in the wind Alex, pressure gauge it is then.

    Reading the original post the throttle body sounds a reasonable place to start looking, I can imagine a little leak but should be minimal?
     
  5. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    I would expect some amount of air to leak past the throttle body, but not sure if its closed with ignition off. Im sure it is.

    That said, I had a sizeable leak from the rear of the inlet manifold (brake booster hose), which maybe means its open or letting quite a large amount of air through.

    Its frustrating, I am finding it very difficult to find a "tuner" to take on assisting me, they are all either A) not bothered b) flakier than the flakiest chocolate.
     
  6. Botus

    Botus Senior Member

    Messages:
    1,362
    Likes Received:
    18
    Joined:
    Jul 29, 2014
    Location:
    UK
    Your Mercedes:
    S500/2010/500
    Unless the idle circuit has some other way to get air in the engine, throttle bodies would have to let air past (auxiliary air valve etc.) I expect they flow "loads" in relation to what you are trying to do.

    To actually build pressure through the inlet system (I guess with the engine off ???) I find it remarkable you can get any pressure built up. The sheer volume the system holds would be massive and the quantity of air a running engine uses is off the scale and almost incomprehensible... If 50 litres vanished at tickover it would only just register after 2 decimal places as a % of air used. Under full boost at 3000 rpm you'll run out of fingers to count with.

    I'd be looking at turbo impellers, waste gate control and other mechanical failures like the intercooler. I gather modern intercoolers are almost built to fail units and make a nicer little earner. I believe its std to last around 5 years on a Porsche and a mate needed one on his A8 recently
     
  7. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Of course, the throttle body would let some past, however during testing the ignition is off, so it should be in the idle position. Allowing some amount of pressure to be build in all the pipework before hand. During testing the engine is not running so we are trying to build positive pressure in all the inlet system before the TB ideally.

    The inter coolers on this are air/water intercoolers rather than AIR/AIR and they sit at the top of the engine:

    [​IMG]

    So I am essentially only testing the section between the turbo outlet and the throttle body, with a few bits after (like one way valves on PCV's etc)

    [​IMG]
     
  8. Botus

    Botus Senior Member

    Messages:
    1,362
    Likes Received:
    18
    Joined:
    Jul 29, 2014
    Location:
    UK
    Your Mercedes:
    S500/2010/500
    nice drawings thanks :)

    if throttle butterfly "closed" completely its possible it could jam in that position... so I expect there's a physical stop leaving a small gap at all times. At idle it (or with the engine off) if that's the case, across both bodies I wouldn't be surprised it could still flow a litre every second.

    the other maybe more sensible thought on the issue you have, is it the turbo boost managed intentionally? I wouldn't be surprised if the gearbox gave up and or the back tyres just light up if they give it loads of boost all the time?

    with modern electronics and traction control could easily be trying to cut back power on purpose... on my 335d a software upgrade on the transmission opened up a clutch or band on purpose to limit the torque the gearbox had to suffer on full throttle takeoff - I didn't like that tweak felt like the box was letting go (which is exactly what it was doing) in my mind increasing the likelihood of failure.

    here's another wild idea, have you toggled the steering wheel buttons and put the car in "Dyno Mode" and checked what happens to your boost?
     
  9. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Indeed. It closes to the idle position. I'm sure.

    The reason I doubt the numbers is that these are known to boost 20/21psi very early (2500rpm) in the rev range then tapers off towards the top, as they are small turbos for this size engine.

    This graph is from my car:
    [​IMG]

    Purple line is absolute boost in inHg, the peak shown on the graph ends up being about 21 PSI, as you can see it reaches 15 then sits there for a good 1000rpm or so then sort of peaks out. It should be straight smooth line to peak at about 4200. To me looks like torque management but what do I know.

    Somebody on the american forum who "knows these engines" says primary suspect is boost leak. I have found numerous leaks, so this may not be that far off. GAD seems to think possibly the waste gate is not closing properly.

    JF automotive say they cant access this standard ECU which is odd. (suspect they don't want to get involved)
    Bad experience with eurocharged (both USA and GB arms)
    and GAD tuning, whilst knowledgeable and friendly, are a nightmare on the customer service/time management front

    Its a possibility that its a got a bad map on it, but no idea who to turn to really.
     
  10. V6Matty

    V6Matty Senior Member

    Messages:
    4,512
    Likes Received:
    39
    Joined:
    Aug 1, 2012
    Location:
    Newark, Notts
    Your Mercedes:
    Peugeot 508 RXH (His) W169/2005/A180 (Hers)
    Big fish tuning? Welcome to pop in here on your way past Alex ;):rolleyes:
     
  11. Botus

    Botus Senior Member

    Messages:
    1,362
    Likes Received:
    18
    Joined:
    Jul 29, 2014
    Location:
    UK
    Your Mercedes:
    S500/2010/500
    well now I can see the boost pressure, I'd say GAD is in the right place.

    Never having played about ( aside from disabling the wastegate on my Mitsubishi Evo :) ), I'd suggest wastegate opens up to stop things going mad early on and for some reason never really regains control till some magic happens later in the revs (past 3200rpm). If the turbo's run out of puff (like they do) how come the boost is highest at high revs... that's all a bit odd.

    Nah, changed my mind.... the jaggedy trend could be the wastegate chattering managing boost... but then why would it continue to climb higher and higher. What are they measuring it with and how much noise does that reading pick up and things start working hard.

    What about dyno mode in the IC ? you didn't respond to that bit. Hope u have lots of air in those back tyres making that much go on rollers !!!

    Aside from the boost reading being a bit odd, the power and torque curves are a nice shape and plenty big enough. Wot numbers are we supposed to see? Don't forget if the dyno mode is in play (no traction control) at 4250 rpm that dip could be the tyres letting go on the rollers
     
  12. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    So the boost numbers were read via OBD which has a relatively slow refresh rate which gives the choppy look to the graph.

    The ECU manages boost by adjusting the boost control solenoid and diverting varying amounts of boost to it at a 30hz duty cycle.

    I checked the waste gates yesterday, they are solidly closed with the engine off, so unless the boost control solenoid is leaking, its more likely that the ECU is opening them for whatever reason. I adjusted them so they are the same length.

    Additionally, if I clamp the boost control pipes, I get an overboost code straight away, so they have no issues creating boost :)

    That dip at 4500 is probably intake restriction. If you remove the filters these cars make 35+ hp at the top end.

    I am struggling as various people say that it should boost 20PSI early and then taper off, but this could be normal for this car! Hence the requirement for a knowledgeable tuner.

    Numbers are about 90hp down, torque is spot on a 980 odd crank NM though.

    Results are the same in dyno mode :)
     
  13. Botus

    Botus Senior Member

    Messages:
    1,362
    Likes Received:
    18
    Joined:
    Jul 29, 2014
    Location:
    UK
    Your Mercedes:
    S500/2010/500
    what fuel are we using ?

    if its on 95 that might explain why the engine management is dialling back the numbers?
     
  14. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Always 98, no evidence of ECU Knock count rising either.
     
  15. OP
    alexanderfoti

    alexanderfoti Moderator

    Messages:
    8,025
    Likes Received:
    98
    Joined:
    Mar 25, 2010
    Location:
    West London
    Your Mercedes:
    W221 S65 AMG
    Vac tested the Vac system today, no leaks. This system actuates the divert er valves ETC.
     

Share This Page