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

WMI and Throttle Bodies - Possible but unconfirmed effects.

Discussion in 'A3/S3 Forum (8L Chassis)' started by Welly, Oct 11, 2011.

  1. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    Morning all.

    It seems there may be a potential problem with the post intercooler WMI setup often used, including by myself...

    There is a possibility, and it's not for definite, that the Methanol over a period of time can eat through the Throttle Body's seals to the electronics.

    Methanol is well know for being corrosive to rubber seals - Methanol Molecule - hence why if you were going to run a car on Meth alone, you would have some serious work to do...

    Anyway, Bumble packed up yesterday morning, and wouldn't even idle, let alone get me to work. EPC light was on on the dash and there was no throttle response at all from the pedal. Checked for fault codes using liquid and got 17976 - Throttle Body Mechanical Malfunction.

    Bearing in mind, the largest amount of meth I have ever used in one go went through the engine on Saturday round the track at Castle Combe, Bill suggest that that the above may be the cause.... and it would seem he is right unsurprisingly.

    [​IMG]

    Little droplets of liquid, unconfirmed what, all under the electronics board and visible on the plastic in the photo above.

    It's unclear exactly what happens here.

    There are two possibilities:

    1: The water/meth is still liquid in the charge air at the point it gets to the throttle body. This should be the case as it should be evaporated. If so, the liquid may be literally eating through the seal. Should be reasonable to resolve.

    2: The water/meth does evaporate properly as expected, but the methanol even evaporated is still corrosive enough to eat the seal, air gets through the perished seal, and the condensates when the engine gets cold. Result, water in the electrics.

    I would say based on what happen to my TB, that the second is most likely. I parked it up the night before and turned it off. When I started it up again, the TB was dead, so condensation of air is definitely possible. IMO.

    Now, my TB is as old as the Cotswolds and has done a million miles, so I'm not saying that this will always happen to others. The seal may well have been old and knackered anyway, and this was always going to happen.

    Throttle bodies do just fail, my mate had one go a few months ago running a stock S3 with no meth at all.

    Coincidence? Possibly.

    I am thinking about how I plan to future proof a TB on an engine running meth... :-/
     
  2. sportstractor

    sportstractor Chugger

    Joined:
    Nov 25, 2009
    Messages:
    2,715
    Likes Received:
    32
    Could you not have a nozzle bossed directly into the inlet manifold?
     
  3. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    Yes, and this done quite often, however that comes with it's own set of considerations.

    The vacuum can draw the meth through at idle which isn't the best. You can use check valves and such to stop the flow though.

    Also, the water/meth doesn't get as long to evaporate...
     
  4. sportstractor

    sportstractor Chugger

    Joined:
    Nov 25, 2009
    Messages:
    2,715
    Likes Received:
    32
    Yeah makes sence, although im sure with a decent nozzle and pump im sure it would evapourate fairly well, when warm.

    Your going to have them all worrying now...
     
  5. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    Not my aim, but something worth considering IMO.
     
  6. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    It's not just about the evaporation though. You also have to consider the amount of time needed for the water and the meth to absorb heat from the air charge, which is the point.

    It's a bit of a delicate balance. Too far and the meth will have become ineffective. Too close and it won't work as effectively. Halfway between the two, and the water meth works at it's best for cooling the air.
     
  7. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    This is all sounding a lot more complecated then I first thought. Are there any alternatives to WMI? Also how much low is the charge temp with the use of WMI?
     
  8. AARON77

    AARON77 Well-Known Member

    Joined:
    Nov 14, 2007
    Messages:
    1,322
    Likes Received:
    26
    Interesting stuff mate me and vince have thought bout this mod, and now were now rethinking!!!
    Have you any pics of the set up? I'm interested to see the set up mate
    Good post mate, have many others had wmi fitted to there motors, and what are the actual gains I ask cos I know you guys like fact and figures so I'm sure you will know

    I'm now wondering bout my nos jet as I've moved on to the TB
    [​IMG]
    I guess I'll soon see
     
  9. DaveA3

    DaveA3 Audi A-Trizzle!

    Joined:
    Feb 25, 2010
    Messages:
    2,375
    Likes Received:
    15
    can you just use some silicone to seal the unit properly. cheap fix problem solved
     
  10. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    It'll eat through that too I would have thought.

    Welly you need to stop hosing down your TB with meth :)
     
  11. Prawn

    Prawn My other car is a MINI!!!!

    Joined:
    Dec 8, 2006
    Messages:
    10,300
    Likes Received:
    1,899
    A mate of mine ran WMI On his st2 LCR, after about 6 months the exact same thing happened to him. Tb full
    Of water and broken :(
     
  12. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    What has he done about it Nick?
     
  13. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    Fitted another one I expect, what else can you do really. You can't really get it apart to do anything with the seals.
     
  14. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    Ok Welly, so the only other option is to tap the jet into the inlet mani. Can it be put in straight after the TB?
     
  15. Broken Byzan

    Broken Byzan Photographic Moderator
    Staff Member Moderator VCDS Map User

    Joined:
    Apr 26, 2004
    Messages:
    12,962
    Likes Received:
    381
    the cure as i see it, is to fit the jet to the inlet manifold, via a solenoid that is only actuated when you are expecting the WMI to be on. That way it cannot draw the liquid willy nily. However this adds cost to a reasonably priced mod, can't win.
     
  16. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    All depends on how much a TB costs you and how often I suppose.
     
  17. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
  18. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
  19. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
  20. <tuffty/>

    <tuffty/> Badger 5 Edition...Its all about the flow...
    Staff Member Moderator

    Joined:
    Sep 14, 2008
    Messages:
    17,430
    Likes Received:
    1,994
    From what I understand the water/meth should be pretty much vaporised by the time it reaches the throttle body so shouldn't be an issue... reading through the issues on the devils own forum, I don't subscribe to the devils own rep suggesting the position is incorrect as pooling will happen anyway given thats the lowest point of the charge system... the problem here is clearly the silicon hose adaptor for the nozzle which I know are rubbish... the videos the guy links to after shows the problem... the fan of the spray pattern is hitting the sides of the adaptor where is too far away from the pipe... if he took the nozzle out of that and ran it with just the nozzle it would atomise fine...

    The other prob is making sure you have the right nozzle size and controller settings....

    Wellys controller is not a progressive one its pressure operated so switches it on or off at a certain boost pressure (IIRC) this kinda controller isn't going to give you the same control as the WMI controllers in other kits...

    Flow of WM with the controllers will mean that there is less WM in the airstream at lower boost/airflow so "shouldn't" pool if its set up right... the issue the LCR owner had was the crappy adaptor making the atomised spray clump together as he used the same nozzle on the adaptor plate

    Have a read of this too for some interesting info..

    Where Do I Position My Water Injection Nozzles

    <tuffty/>
     
  21. superkarl

    superkarl MAN OF STEEL

    Joined:
    Mar 28, 2010
    Messages:
    8,202
    Likes Received:
    1,477
    So many things to consider in that article tuffty posted.
    if you inject pre or post throttle body how would you make sure its being distributed evenly? im thinking a nozzle on an angle so it sprays straight as can be down the plenum.

    the idea of a plate inbetween the TB and inlet mani with holes for nozzles seems a good idea
     
  22. <tuffty/>

    <tuffty/> Badger 5 Edition...Its all about the flow...
    Staff Member Moderator

    Joined:
    Sep 14, 2008
    Messages:
    17,430
    Likes Received:
    1,994
    Bill's 300hp Lupo has an AEM kit with the nozzle fitted into the inlet mani...
    [​IMG]

    Ended up putting 2 check valves in as it was syphoning WM on idle... inlet mani was well cold though :)

    <tuffty/>
     
  23. s3dave

    s3dave TFSI Hybrid
    Supporter

    Joined:
    Nov 11, 2009
    Messages:
    10,120
    Likes Received:
    732
    I will be looking at the best nozzle you can buy to vaporise it the best it can, welding a tube in the middle of the manni around 2" diameter and about 3" in length, with the nozzle fitted at the end of the tube, then in theory it should be sucked into each branch and not go near the TB.....:readit:
     
  24. StaceyS3

    StaceyS3 Well-Known Member

    Joined:
    Dec 17, 2010
    Messages:
    2,240
    Likes Received:
    265
    That article on where the nozzle is fitted has made me wonder if ive put mine in the best place? :think:
     
  25. <tuffty/>

    <tuffty/> Badger 5 Edition...Its all about the flow...
    Staff Member Moderator

    Joined:
    Sep 14, 2008
    Messages:
    17,430
    Likes Received:
    1,994
    Nozzle size determines ultimate flow rather than anything else on WMI set ups... the nozzles typically have quite a wide fan...

    Devils own have a 'calculator'...
    Methanol Injection : Alcohol Injection : Water Injection : DevilsOwn

    ...and a discussion about it here..
    Alcohol Nozzle size Calculator - Devilsown Methanol Alcohol Water Injection

    Should be noted that cyl 4 is the one that typically pulls the most timing... this is partially due to the proximity of the water elbow and flow of coolant at that point... in most if not all the times I have seen 20v's on the dyno its Cyl4 that that pulls more often than the rest..

    Also bare in mind that its still not an exact science... have seen a lot of variance in suitable nozzle sizing and settings on the same make/model of controller on similarly specced levels of tune... there is still a need to dial in WMI on a dyno and be conservative with timing advance etc...

    Personally mounting post IC is still a preferable choice for charge cooling... this means that the IAT sensor see the correct charge air temp and won't pull timing when things get too hot... while inlet mani injection does help if the IC setup isn't up to spec and IAT's (without WMI) are getting into the 50deg+ area then the ECU will pull timing and direct WMI won't help as much... if the ECU sees lower temps from the outset then you could potentially use inlet mani injection to maximise timing advance etc with better cylinder cooling and octane enhancement

    <tuffty/>
     
  26. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    Ah yes! I forgot all about the AIT sensor being before the TB. This is a right dilemma!
     
  27. s3dave

    s3dave TFSI Hybrid
    Supporter

    Joined:
    Nov 11, 2009
    Messages:
    10,120
    Likes Received:
    732
    Move the AIT as well?
     
  28. Westy

    Westy Double Dark Side! Diesel & 8P

    Joined:
    Dec 2, 2008
    Messages:
    17,358
    Likes Received:
    1,007
    I'm loosing the plot! The AIT is after the TB in the inlet mani. So surely fitting the WMI jet where Bill has in the above pic is the way to go to avoid ruining TBs

    [​IMG]
     
  29. <tuffty/>

    <tuffty/> Badger 5 Edition...Its all about the flow...
    Staff Member Moderator

    Joined:
    Sep 14, 2008
    Messages:
    17,430
    Likes Received:
    1,994
    I think we are drifting away from the point a little here... seem to be too worried about fitting the WMI to save the TB rather than for optimum effect...

    As Welly has mentioned the TB failing could well be incidental... would seem to me that when fitting WMI post IC like Welly and a lot of other people have (including myself when I get around to it) if the settings are just a little out then the charge air is still going to contain moist WM by the time it hits the TB... this could then affect the TB as it would appear to have in Welly's case...

    Correct atomisation and flow is key here I reckon... Bill fitted the nozzle where he did on the Lupo purely due to there being no where else physically yot fit it... also bare in mind that it is essentially firing at the IAT sensor so at lower airflow this could potentially provide false readings for the actual IAT's until the airspeed picks up and draws it past the sensor enough for the charge air to provide a true readiing of charge air temps...

    Remember that as the WM is injected at this point then the sensor will still see the hotter charge air as the WM will not have an effect soon enough for it to be picked up with the sensor in the short time the air is in the plenum... there will be an accumulative effect of course as the WM will help cool the inlet mani keeping everything cooler generally...

    <tuffty/>
     
  30. Welly

    Welly VX220 SC Driver :)

    Joined:
    Aug 5, 2009
    Messages:
    10,699
    Likes Received:
    120
    Bottom line, this is a bit of a balancing act.

    I am happy with where my WMI jet is. It is half way between the IC and the TB basically.

    It is there because that is the best balance of detonation suppression, and charge air cooling. These are the two effects I want to see from WMI and I currently get them both.

    I am more concerned with finding a fix for the problem, as opposed to working around it by moving the nozzle to a different location. The closer the WMI nozzle is to the inlets, the less heat will be absorbed. That is mainly why I use water meth, to keep the temps down...

    So, having discussed this with Bill last night, I reckon that there are two possible fixes. Both of them being reliant on the fact that the water in the TB is from condensation of humid air (caused by the WMI) on engine cooling, which I genuinely believe is the case.

    Option 1:

    Take off the black cover and install a silica gel pack to absorb the moisture inside the TB mechanics. This should draw the moisture in, meaning that the electronics never see it. (Bill's idea)

    Option 2:

    Drill 2 small breather holes inside the plastic cover itself. This will allow to moisture to escape and for engine cooling to happen without the condensation effect. (Obviously my idea as it involves breaking stuff and drilling)

    The main problem here is that no-one really knows HOW the TB operates. Is it seals, or does the TB operate on a bearing system which will naturally allow some air/water in anyway?

    Plans to destruct my TB to find out are being forged....
     
  31. DaveA3

    DaveA3 Audi A-Trizzle!

    Joined:
    Feb 25, 2010
    Messages:
    2,375
    Likes Received:
    15
    i find it hard to believe it would munch through it that quickly though if at all. if u bought a tube u could do it every 5-6 months and it wouldnt be an issue.

    how thick is the current seal?
     
  32. badger5

    badger5 www.badger5.co.uk
    Site Sponsor

    Joined:
    May 28, 2004
    Messages:
    7,930
    Likes Received:
    963
    there is some evidence that some kits will allow syphoning which is clearly not going to be good..
    also, if the system switches on too early where there is insufficient airflow to carry the spray into the inlet, it would pool and droplets could amass onto pipes, throttle plate etc... Too large a nozzle also, too close to tbody upstream... all *could* influence this.

    Lupo's install was post tbody cos there was no space upstream to stick it.. BUT, as tuffty said, its subject to vacuum there, so will allow the wmi to be sucked thru by vacuum. uber cold mani - lol, but not as designed. Solenoid control would eliminate this to valved off until switched on. check valves have'nt fully resolved the issue, just reduced the amount sucked thru. lupo's getting an fmic currently and will have a nozzle mounted post IC like normal.. I may play with running dual nozzles for the crack.. who knows.. when/if I get spare time
     
  33. <tuffty/>

    <tuffty/> Badger 5 Edition...Its all about the flow...
    Staff Member Moderator

    Joined:
    Sep 14, 2008
    Messages:
    17,430
    Likes Received:
    1,994
    Great minds etc... was thinking about this for when I get around to fitting to the S3 ;p

    <tuffty/>
     

Share This Page