Taking bets on the limp mode problem

[aliencurv]

Re: Taking bets on the limp mode problem

the o2 sensor is used in "closed loop" mode. the computer uses the o2 sensor to determine stoich (which is 14.7 afr) during cruising .

if the o2 sensor is defective then cruising will be very rich and get bad gas milleage. but your WOT (wide open throttle) will be fine. during WOT, the computer goes to Open Loop which references a few maps to determine the final afr. WOT does not use the o2 sensor.

if there was a boost leak, then you would be running too rich. (unless the boost leak was affecting the charge pressure sender G31 and causing over boost). a boost leak would mean measured air was escaping when the computer expected it to be in the engine.

interesting thing i noticed while examining the monotronic programming and available senders tonight. it's interesting how the monotronic determines air density and humidity. the maf measures air in lbs/min as expected. then the monotronic knows the volume of 1 cylinder which it uses to calculate load. but interestingly, the G31 sender is referenced now to ensure the expected ammount of pressure is reached. if the pressure is too low (boost leak) then it will soft limp because it thinks the G31 or MAF is possibly defective.

so that's interesting that it can compare the two to determine "something is wrong". most pcm/ecu's have only one or the other. monotronic having two is awesome.

i dig that

[R-Audi]

Re: Taking bets on the limp mode problem

Your MAF shouldnt be 300... you can go with the V8 A8/S8 and it will work.... I think the one i got from George was in the 150~ range? Probably a re-man.

[RSS]

Re: Taking bets on the limp mode problem

I am going to run the car a couple days and watch my LTFT's and see where they climb to.

[The_Jerbel]

Re: Taking bets on the limp mode problem

Those first options where my guesses on the problem i checked it and it seemed to be tight. I did fully open it up just to make sure tho i will run it the next couple of days and see if any of the codes return.

If it was a deviation wouldn't i get a code for that as well?

Oh ok, I thought you were saying that it actually had happened.

This code is usually the result of a boost leak, but can also be caused by a primary O2 sensor that is way out of wack:

17538 - Fuel Trim: Bank 2 (Multi) System too lean
P1130 - 35-10 - intermittent

[The_Jerbel]

Re: Taking bets on the limp mode problem

the o2 sensor is used in "closed loop" mode. the computer uses the o2 sensor to determine stoich (which is 14.7 afr) during cruising .

if the o2 sensor is defective then cruising will be very rich and get bad gas milleage. but your WOT (wide open throttle) will be fine. during WOT, the computer goes to Open Loop which references a few maps to determine the final afr. WOT does not use the o2 sensor.

if there was a boost leak, then you would be running too rich. (unless the boost leak was affecting the charge pressure sender G31 and causing over boost). a boost leak would mean measured air was escaping when the computer expected it to be in the engine.

interesting thing i noticed while examining the monotronic programming and available senders tonight. it's interesting how the monotronic determines air density and humidity. the maf measures air in lbs/min as expected. then the monotronic knows the volume of 1 cylinder which it uses to calculate load. but interestingly, the G31 sender is referenced now to ensure the expected ammount of pressure is reached. if the pressure is too low (boost leak) then it will soft limp because it thinks the G31 or MAF is possibly defective.

so that's interesting that it can compare the two to determine "something is wrong". most pcm/ecu's have only one or the other. monotronic having two is awesome.

i dig that

A boost leak usually causes you to run lean because the MAP (G31) sensor reads lower than it should, so the ECU adjusts fuel injection accordingly. If it reads low enough, it will put the car into soft limp mode, as you pointed out.

[aliencurv]

Re: Taking bets on the limp mode problem

A boost leak usually causes you to run lean because the MAP (G31) sensor reads lower than it should, so the ECU adjusts fuel injection accordingly. If it reads low enough, it will put the car into soft limp mode, as you pointed out.

boost and airflow are seperate items that are related by density, temperature and humidity. boost is a measurement of air your engine/pump isn't using. but the ammount of air in the charge is measured by the MAF. the maf is what is referenced to calculate the required fuel. the g31 is used to control boost (not a primary use for fuel). our computer is Load Based which is calculated by the maf.

1) air passed by the maf (computer now knows how much air there is in the charge pipe)
2) air leaks out of the charge pipe
3) computer thinks it knows how much is there so puts in fuel respectively

^ as you can see the computer puts in fuel due to what it measured on step #1 which results in a rich condition because there is more fuel then the "expected" air

now the map sensor is used to control boost levels. the computer knows how much "air" is in the pipe, but that doesn't mean pressure (at least not accurately). so the simpliest way is to have a map sensor. when the computer wants 20 psi, it opens the N75 solenoid which opens the wastegates.

the computer doesn't use the G31 for airfuel (reference page 59 of the Audi Engine Management Level 1 Course test). this same procedure applies for other maf/load based computers. the sensor is used to monitor boost and not used for final air fuel calculations.

[The_Jerbel]

Re: Taking bets on the limp mode problem

Yes I'm well aware of what you just explained, except that you're only partially correct.

- Boost IS the airflow after it's been compressed.
- The MAF is used to calculate load, but there are other variables and inputs as well.
- The G31 is a secondary input for calculating load, which directly affects AFR.

I can tell you that, from experience, when you have a boost leak, it can cause LTFT to decrease, and you run lean. BTW, this was verified by using a WB AFR gauge, not just ME7.1's output readings.

[RSS]

Re: Taking bets on the limp mode problem

Well i am going to do some logs after work to get some data then hopefully we can narrow it down.

Blocks 31 , 32 , 003

and anything else any one suggests?

[Mantis]

Re: Taking bets on the limp mode problem

yup, fix your boost leak, lol, I want that cookie

[RSS]

Re: Taking bets on the limp mode problem

Well here are the logs from tonight i am starting to think that the limp mode was from my LTFT's when i had the boost leak that i recently fixed? The O2 sensor does look like it is dying tho.

LTFT's looked like this after 1 day of driving around before logs

Idle Wot Idle Wot

2.8 3.1 2.4 3.9

After Logs / Hard Runs

2.8 2.3 2.3 2.3

[The_Jerbel]

Re: Taking bets on the limp mode problem

Looks like Ryan (nothing77) gets the cookie, but I think I deserve a bite

[RSS]

Re: Taking bets on the limp mode problem

Does he get the cookie for me fixing it a week before the problem occured?

[The_Jerbel]

Re: Taking bets on the limp mode problem

Yup

No offense, but there is a chance you didn't fix it good enough. Have you done a boost leak test since you did the fix?

[R_Ethug]

Re: Taking bets on the limp mode problem

hey guys to do the boost leak check do you need an air compressor??

of could you use a bike pump or 12v car pump to do it?? i think i want to do mine

[Mantis]

Re: Taking bets on the limp mode problem

your pushing it with a 12v car pump, if you need a compressor lt me know, i'll lend you one of my old ones