Pages: [1]
Author Topic: Can Charge Air System Leaks be Detected/Inferred by Logging?  (Read 7346 times)
TTQS
Guest
« on: March 29, 2011, 05:40:29 PM »

Hi.

I don't believe I have any charge air system/boost leaks on my car, but leaks were very frequently detected during pre-tune health checks on Audi TTs so I became quite paranoid about them.

I haven't done a great deal of logging on my car either pre- or post-remap but the general consensus on the car-specific forum I was previously a member of seemed to be that rich running caused by loss of intake air already metered and accounted for would not manifest itself on any loggable parameters. Therefore, charge air system leaks would have to be detected by physical inspection and a pressure test of the pipework...

Is this indeed the case, or are there in fact obvious ways of detecting boost leaks or subtle ways to infer their presence by logging particular VCDS/VAG-COM parameters? I'm afraid I haven't given it a lot of thought, but would anything show up at the lambda sensor in the exhaust system perhaps?

Clearly, if there was a gross leak, then actual boost/MAP would not be able to meet the boost demanded by the ECU, but I presume this kind of fault would be easily detectable physically. I'm referring to smaller leaks that could be detected at early onset and then rectified without causing long-term stress to the turbocharger components.

Thanks in advance.

Doug
Logged
Matt Danger
Full Member
***

Karma: +17/-1
Offline Offline

Posts: 116


« Reply #1 on: March 29, 2011, 10:00:15 PM »

One way is to look for a deviation between requested vs. actual boost.
Logged
TTQS
Guest
« Reply #2 on: March 30, 2011, 02:01:45 AM »

One way is to look for a deviation between requested vs. actual boost.

Yes indeed, I recognise that, but I am under the impression (rightly or wrongly) that there would be no difference if the turbo still had enough capacity to meet requested boost. Or, is it the case that during a full load WOT run, there would always be a discrepancy with a boost leak of any magnitude?

Doug
Logged
Tony@NefMoto
Administrator
Hero Member
*****

Karma: +119/-4
Offline Offline

Posts: 1389


2001.5 Audi S4 Stage 3


« Reply #3 on: March 30, 2011, 11:36:44 AM »

In my opinion the best way to detect leaks would be to inspect the long term fuel trims.
Logged

Remember you have to log in if you want to see the file attachments!
Info or questions, please add to the wiki: http://www.nefariousmotorsports.com/wiki
Follow NefMoto developments on Twitter: http://twitter.com/nefmoto
TTQS
Guest
« Reply #4 on: March 30, 2011, 07:15:57 PM »

would anything show up at the lambda sensor in the exhaust system perhaps?

O.k. Thanks for confirming my guess Tony. I looked into it and realised I had read about LTFTs before on the S4 wiki and elsewhere, but didn't realise what they were because the term is different from the measuring block title. Some notes of my own I made by condensing information available elsewhere are included in the next post.
« Last Edit: March 30, 2011, 07:19:05 PM by TTQS » Logged
TTQS
Guest
« Reply #5 on: March 30, 2011, 07:17:04 PM »

LTFT (long-term fuel trim) is shown in VAG-COM measuring block 032 [Oxygen sensors learning values (maximum value)] at both idle which is an additive trims and partial load which is a multiplicative trim. Negative values indicate that the engine is running too rich and oxygen sensor control is enleaning it by reducing the amount of time that the injectors are open. Positive values indicate that the engine is running too lean and oxygen sensor control is enriching by increasing the amount of time that the injectors are open.

Specifications for normal operation are around +/- 10% although the Bentley manual states +/- 25%. The ECU will throw DTCs for values outwith this range. E.g. 17544/P1136/004406 - Fuel Trim: Bank 1 (Add): System too Lean or 17545/P1137/004407 - Fuel Trim: Bank 1 (Add): System too Rich

Off-nominal values in the first field (additive) could indicate a leak in the vacuum system since it is mostly present at idle, when vacuum is highest. Clearing DTCs will yield zero values, as will a faulty or disconnected sensor.

Note 1. An additive trim (idle) is used to make the injectors stay open a fixed amount of time longer or shorter because the fault (e.g. vacuum leak) becomes less significant as engine speed increases.

Note 2. A multiplicative trim (partial loads) is used because the fault (e.g. partially-blocked fuel injector) becomes more severe at higher engine speeds. This parameter is dependent on basic injection timing.
« Last Edit: March 30, 2011, 07:19:39 PM by TTQS » Logged
silentbob
Full Member
***

Karma: +30/-1
Offline Offline

Posts: 141


« Reply #6 on: April 03, 2011, 11:49:34 PM »

Depending on how the four adaption fields are calibrated they won't tell you much on possible air leaks under boost conditions (Sections LRA and LRAEB in the Funktionsrahmen). For example on the binary lambda sensor cars no adaption can take place if you run rich (what you usually do on boost unlike the stock calibration). Also it's calibrated that the adaption is blocked with TPS higher than 99%.
Even on linear sensor cars adaption is calibrated only to run within lambda 0,95<x<1,05.
So information you can get out of this is very limited or even useless.

Other thing I can think of is the N75 DC. If it's significant higher under the same conditions the wastgate has to be in a more closed position to get the same boost meaning the turbo has to spin faster to get the requested air mass.
 
« Last Edit: April 03, 2011, 11:51:12 PM by silentbob » Logged
nyet
Administrator
Hero Member
*****

Karma: +441/-62
Offline Offline

Posts: 10092


WWW
« Reply #7 on: April 04, 2011, 12:30:56 PM »

I looked into it and realised I had read about LTFTs before on the S4 wiki and elsewhere, but didn't realise what they were because the term is different from the measuring block title. Some notes of my own I made by condensing information available elsewhere are included in the next post.

Just a word of warning about s4wiki content: pretty much all of it predates my having the Funktionsramen, so it is all synthesized from logging and empirical evidence Smiley

IOW, take it all with a grain of salt. In particular, if you see something in the s4wiki that conflicts with the Funktionsramen, PLEASE correct it (or drop me a note) so I don't continue spreading false information...

Thanks in advance!
Logged

ME7.1 tuning guide (READ FIRST)
ECUx Plot
ME7Sum checksum checker/corrrector for ME7.x

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience.
TTQS
Guest
« Reply #8 on: April 04, 2011, 03:20:00 PM »

Thanks for your thoughts guys. Just a note about the N75 duty cycle; for that kind of comparison to be useful, you'd need a baseline duty cycle for a sound charge air system which might not be available.

Say I went to check out a car for possible purchase and took it out for a test drive, plugged in the laptop with VCDS/VAG-COM. What could I log to give me an indication of whether there are charge air system leaks that might not be evident on a quick survey under the hood?

Nyet: S4 wiki is still good material despite pre-dating your exposure to the funktionsrahmen!

Doug

P.S. Just checked the LTFTs and the idle value was -0.9% and the partial load value -1.6%. Happy days!
« Last Edit: April 05, 2011, 05:39:28 AM by TTQS » Logged
Jason
Sr. Member
****

Karma: +36/-0
Offline Offline

Posts: 498


Breaks everything!


« Reply #9 on: April 08, 2011, 01:22:38 PM »

In my opinion the best way to detect leaks would be to inspect the long term fuel trims.

I agree 100% - this has been the tried and true method for me.
Logged
julex
Hero Member
*****

Karma: +77/-4
Offline Offline

Posts: 923


« Reply #10 on: April 08, 2011, 05:44:58 PM »

unofrtunately the only reliable way would b wideband sensor.

Medium-severe leaks would show as CEL/limp mode and code for "negative boost deviation".
Small leaks would be pretty undetectable unless you had previous logs when car was healthy, you'd probably notice different WOT o2 voltages on your narrowband sensors which would tell you that something is off. You'd notice higher n75 duty for similar rpms/boost and probably decreased performance too from running rich. Lack of pinging would be another clue.
Logged
TTQS
Guest
« Reply #11 on: April 09, 2011, 09:33:40 AM »

unfortunately the only reliable way would be wideband sensor.

TT has wideband oxygen sensor.

Doug
Logged
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.024 seconds with 16 queries. (Pretty URLs adds 0s, 0q)