When to replace MAFs?
Bad MAFs don't always throw fault codes. An understanding of the Vagcom logs (and not just the MAF values) is the starting point.
Clive
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
-
RS6-4dr911
- 1st Gear
- Posts: 55
- Joined: Sun Jan 25, 2009 12:34 am
- Location: Houston, TX, USA
Don't just tease us, explain. Fine, I've got logs, both banks, multiple runs, the entire rpm range, other variables logged along with them, I'm ready to start. Now what?
The sum of your comments lead me to wonder if you really know how to determine if their bad or not. If you do, spill the beans. If you don't, stop acting like you possess the secret to eternal life. Shopinnit and I, at least if not others as well, are trying to resolve a problem and are looking for concrete answers, not vague generalities like get a proper health check, and unless you understand Vagcom logs . . . . If you have a solution share it, please. Otherwise, sit back and learn like the rest of us.
Shoppinit, some logs are attached to compare to yours, let me know how we stack up. In speaking with a tuner and my mechanic today, I'm pretty much ready to just replace them based on the variance between them combined with their age and mileage.
The sum of your comments lead me to wonder if you really know how to determine if their bad or not. If you do, spill the beans. If you don't, stop acting like you possess the secret to eternal life. Shopinnit and I, at least if not others as well, are trying to resolve a problem and are looking for concrete answers, not vague generalities like get a proper health check, and unless you understand Vagcom logs . . . . If you have a solution share it, please. Otherwise, sit back and learn like the rest of us.
Shoppinit, some logs are attached to compare to yours, let me know how we stack up. In speaking with a tuner and my mechanic today, I'm pretty much ready to just replace them based on the variance between them combined with their age and mileage.
- Attachments
-
- MAF Check Jan24-10 3-10pm.xls
- (25 KiB) Downloaded 79 times
-
- MAF Check Jan24-10 3-13pm.xls
- (32 KiB) Downloaded 89 times
I'm sorry, I'm no expert and was trying to be helpful by saying what I do know 
But I'll take your 'advice' and sit back, unless of course the others taking part in this thread may think otherwise...
But I'll take your 'advice' and sit back, unless of course the others taking part in this thread may think otherwise...
Clive
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
Being the lazy git that I am, made the suggestion to change them (MAFs/N75) between RR runs at the MRC RR day just to avoid all the logging etc. Still think it's a good idea.
If I don't make factory power on my first run, then will definitely go down the route of changing the above whilst I am there and then running again.
If I still don't make factory power then basically Doug explained to me that sometimes cars just don't like the rollers and are perfectly fine on the road - in which case I'll try to do the logging of various values whilst out on a run (there will be plenty of people on the RR day with VAGCOM who know what they're doing).
On the day, I should have a contingency for every scenario. (Just need to make sure Doug has new MAFs and N75s in stock before I go). Oh and remember my credit card
I'm sure many will be watching my antics with interest! All I want is to be somewhere around the 480hp mark. That's all I ask!
If I don't make factory power on my first run, then will definitely go down the route of changing the above whilst I am there and then running again.
If I still don't make factory power then basically Doug explained to me that sometimes cars just don't like the rollers and are perfectly fine on the road - in which case I'll try to do the logging of various values whilst out on a run (there will be plenty of people on the RR day with VAGCOM who know what they're doing).
On the day, I should have a contingency for every scenario. (Just need to make sure Doug has new MAFs and N75s in stock before I go). Oh and remember my credit card
I'm sure many will be watching my antics with interest! All I want is to be somewhere around the 480hp mark. That's all I ask!
Yes, we'll be watching with interest Mo 
Clive
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
The following is a plot of RS6-4dr911 (top) and my logs (bottom). I'm not sure how useful it is to compare them unless the condition under which they were taken are comparable. RS6-4dr911, are these from a power run from idle? Mine are not necessarily from a power run which may explain the difference in the lower part of the graph.
One thing I notice with my MAF readings is that at lower flows, they are close, but at higher flow they deviate significantly as do yours. Also I note that the absolute readings on yours are lower than mine. The puke green plot is the engine load. The others are the MAF readings.
I'll go out again tomorrow and do log some power runs. Starting at say 1500rpm then accelerating to red line. This way my plots should be cleaner. If your plots weren't done this way, perhaps you can do some more using my method.
One thing I notice with my MAF readings is that at lower flows, they are close, but at higher flow they deviate significantly as do yours. Also I note that the absolute readings on yours are lower than mine. The puke green plot is the engine load. The others are the MAF readings.
I'll go out again tomorrow and do log some power runs. Starting at say 1500rpm then accelerating to red line. This way my plots should be cleaner. If your plots weren't done this way, perhaps you can do some more using my method.
Daytona RS6 C5 Avant. Viper'd, Billies, Waggers, MTM box brain, C6 stoppers, xcarlink, R8 coolant cap (woohoo)
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
It's appreciated, Clive.CliveH wrote:I'm sorry, I'm no expert and was trying to be helpful by saying what I do know
Daytona RS6 C5 Avant. Viper'd, Billies, Waggers, MTM box brain, C6 stoppers, xcarlink, R8 coolant cap (woohoo)
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
Clive
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
S2 ABY coupe, S4 B5 saloon, S4 B6 avant
RS4 B7 phantom black saloon, mint, fully loaded, low mileage - FOR SALE!- http://www.rs246.com/index.php?name=PNp ... ic&t=88981
S8 D2 facelift, RS6 C5 saloon, both gone but not forgotten
-
RS6-4dr911
- 1st Gear
- Posts: 55
- Joined: Sun Jan 25, 2009 12:34 am
- Location: Houston, TX, USA
Mine were in fact done as a "power run", I believe on the data I posted I've stripped the non-power-run data out for simplification. I usually start at about 2000rpm in second or third in tip mode and run until it upshifts.Shoppinit wrote:The following is a plot of RS6-4dr911 (top) and my logs (bottom). I'm not sure how useful it is to compare them unless the condition under which they were taken are comparable. RS6-4dr911, are these from a power run from idle? Mine are not necessarily from a power run which may explain the difference in the lower part of the graph.
One thing I notice with my MAF readings is that at lower flows, they are close, but at higher flow they deviate significantly as do yours. Also I note that the absolute readings on yours are lower than mine. The puke green plot is the engine load. The others are the MAF readings.
I'll go out again tomorrow and do log some power runs. Starting at say 1500rpm then accelerating to red line. This way my plots should be cleaner. If your plots weren't done this way, perhaps you can do some more using my method.
You mention the lower readings, my mechanic also said (without verifying facts) that they sounded low for that size engine. Another nail in the MAF's coffin.
-
RS6-4dr911
- 1st Gear
- Posts: 55
- Joined: Sun Jan 25, 2009 12:34 am
- Location: Houston, TX, USA
Relooked at the data, did not in fact strip out part throttle data, but pretty easy to see where Load% goes to 100 or so.
Paul, correct it does not log all data at the exact same time, but if you look at the raw data, reading across the columns, each you time you hit a "Time Stamp" column, everything to the right of that was taken at the same time until you hit another "Time Stamp" column. To my knowledge, all left/right bank data (MAF, Lambda, others) is logged at the same time, so for the variation between MAF's this is not a concern. Even if there are data that are logged at different stamps, the intervals are the same and it is pretty accurate to interpolate linearly between the readings and graph data with adjusted values to eliminate this problem. I've done it with transmission data trying to determine slippage where, IIRC, shaft RPM's and engine RPM's and Vehicle speed were logged at different times. Interpolation made for a much closer correlation between the data, showing that the box was sound.
Paul, correct it does not log all data at the exact same time, but if you look at the raw data, reading across the columns, each you time you hit a "Time Stamp" column, everything to the right of that was taken at the same time until you hit another "Time Stamp" column. To my knowledge, all left/right bank data (MAF, Lambda, others) is logged at the same time, so for the variation between MAF's this is not a concern. Even if there are data that are logged at different stamps, the intervals are the same and it is pretty accurate to interpolate linearly between the readings and graph data with adjusted values to eliminate this problem. I've done it with transmission data trying to determine slippage where, IIRC, shaft RPM's and engine RPM's and Vehicle speed were logged at different times. Interpolation made for a much closer correlation between the data, showing that the box was sound.
OK. I did a couple of clean runs. Here is the data. Both runs starting at 1500 in 2nd. Full throttle, progressively applied, to redline. Interesting results. The first thing I notice is that lambda correction goes to 0 past a certain point. I'm not sure what this point is determined by. The big news (at least to me) is that the car is now relying only on MAF values and knock sensors. So NOW I can understand why bad MAFs would affect performance.
- Attachments
-
- LOG-01-001-002-003.xls
- XLS log values
- (245.5 KiB) Downloaded 57 times
Daytona RS6 C5 Avant. Viper'd, Billies, Waggers, MTM box brain, C6 stoppers, xcarlink, R8 coolant cap (woohoo)
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
-
RS6-4dr911
- 1st Gear
- Posts: 55
- Joined: Sun Jan 25, 2009 12:34 am
- Location: Houston, TX, USA
Interesting. The Lambda goes to zero at or near WOT as it goes into "open loop" mode. Not sure why, maybe a function of processing capacity and/or at WOT who cares if you're efficient or polluting. Even though I already knew this, I failed to make the small logical leap to the fact that then you're relying solely on the MAF for fueling. And if they're bad . . . . So that answers your question that nobody else seemed to respond to about why bad MAF's were a problem if the O2 sensors adjust fueling to compensate - it's only at part throttle that it does that.
Did you notice that your field that is Total MAF, peaks at 364 even though the sum of MAF Banks 1 & 2 is higher than that?? Total MAF for yours is about 10-15% higher than mine. Mine never hits 364 so not sure if that is some sort of built in limit or if it's peculiar to your car.
After all this mathematical masterbation, are either of any closer to figuring out what to do next with our cars?
Did you notice that your field that is Total MAF, peaks at 364 even though the sum of MAF Banks 1 & 2 is higher than that?? Total MAF for yours is about 10-15% higher than mine. Mine never hits 364 so not sure if that is some sort of built in limit or if it's peculiar to your car.
After all this mathematical masterbation, are either of any closer to figuring out what to do next with our cars?
I noticed the discepancy in the bank 1 + bank 2 MAF values. I assumed it was because they were being measured at slightly different times since they are 2 separate measuring blocks.
There are other lamba values that the car uses, too. I'm not clear on how this works, but there are at idle values (correction that is learned over a period of time) and driving values - % of adjustment per bank that is also learned over a period of time. As opposed to what I plotted above which is the instaneous lambda correction. I'm not sure how these latter two interact.
As for what to do next, I think I'm going to have to get some new MAFs just to see. I may never be satisfied that they're working properly until I put some new ones on! Don't forget that I recently had my car rolling roaded and it produced 450hp. But like I've said before, it's not peak output that is be all and end all, but the power on the way up the rev range. Unless you've got a CVT box of course
There are other lamba values that the car uses, too. I'm not clear on how this works, but there are at idle values (correction that is learned over a period of time) and driving values - % of adjustment per bank that is also learned over a period of time. As opposed to what I plotted above which is the instaneous lambda correction. I'm not sure how these latter two interact.
As for what to do next, I think I'm going to have to get some new MAFs just to see. I may never be satisfied that they're working properly until I put some new ones on! Don't forget that I recently had my car rolling roaded and it produced 450hp. But like I've said before, it's not peak output that is be all and end all, but the power on the way up the rev range. Unless you've got a CVT box of course
Daytona RS6 C5 Avant. Viper'd, Billies, Waggers, MTM box brain, C6 stoppers, xcarlink, R8 coolant cap (woohoo)
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
I'm talking about the additive and the multiplicative trim, as described nicely here:
http://www.ross-tech.com/vag-com/cars/fuel-trim.html
So what are the lambda values that we're looking at above doing? Anyone? Bueller?
http://www.ross-tech.com/vag-com/cars/fuel-trim.html
So what are the lambda values that we're looking at above doing? Anyone? Bueller?
Daytona RS6 C5 Avant. Viper'd, Billies, Waggers, MTM box brain, C6 stoppers, xcarlink, R8 coolant cap (woohoo)
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
///M3 E46 | XC90 (V8, natch) | Passat GTE | RR Classic V8 flapper
"The major difference between a thing that might go wrong and a thing that cannot possibly go wrong is that when a thing that cannot possibly go wrong goes wrong it usually turns out to be impossible to get at and repair."
Who is online
Users browsing this forum: audirew, Google Adsense [Bot] and 89 guests