Volkswagen Passat Forum banner
1 - 20 of 40 Posts

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #1 · (Edited)
EDIT: Due to miscalculation, the original graph is no longer accurate. Please see below for the new graph.

I calculated duty cycle percentages by dividing the injector pulsewidths by 1000, and multiplying by the number of intake cycles per second.

Then, I calculated fuel consumption by taking the 100% duty cycle injection rate (210cc/min), converting it to grams/second, multiplying it by the actual duty cycle percentages, and multiplying by 4 (the number of cylinders I have). I also adjusted the flow rates (measured at the 3 bar standard) upwards by 33% to account for the 4 bar stock FPR.

Finally, I calculated Lambda by dividing measured airflow in grams/second by fuel consumption in grams/second, then dividing by Lambda 1 (14.7:1).

Did I do it right? The data looks good, but I'm worried about the fanout at low airflow rates. According to things Rusty has said in the past, it's acceptable for Lambda to dip as low as .85 (12.5:1) under heavy throttle, but regardless of the way I drive the car, Lambda shouldn't go above 1 -- or at least, it shouldn't go as far above 1 as it appears to have.

Oh, and the handful of data points at the bottom of the graph aren't really Lambda 0, they're actually Lambda Infinite. Yes, I have data points showing that, on occasion, my fuel injectors were COMPLETELY OFF. What the hell does that mean?
 

·
Founder of the STFA
Joined
·
8,667 Posts

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #5 ·
Okay, I corrected the data to account for the higher-than-expected fuel pressure.



The data looks good, the average Lambda is .97, but I'm worried about the fanout at low airflow rates. According to things Rusty has said in the past, it's acceptable for Lambda to dip as low as .85 (12.5:1) under heavy throttle, but regardless of the way I drive the car, Lambda shouldn't go above 1 -- or at least, it shouldn't go as far above 1 as it appears to have.

Oh, and the handful of data points at the bottom of the graph aren't really Lambda 0, they're actually Lambda Infinite. Yes, I have data points showing that, on occasion, my fuel injectors were COMPLETELY OFF. What the hell does that mean?
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #7 ·
how about just getting an A/F ratio meter?
For starters, my problem is more complicated than just the A/F ratio. I also need some idea of what would cause the injectors to have a 0ms duty cycle periodically, for durations up to two seconds at a shot.

As far as the A/F gauge is concerned:

All that would do is give me a readout from the front oxygen sensor, which won't do me any good above 50% throttle, because on my engine the oxygen sensors don't have a wide enough range of operation to detect the richer full-throttle A/F ratio. Believe me, I had definite plans to get an A/F gauge until I found out this dirty little secret.

Can anybody please tell me whether the ratio fanout at the left side of the graph is normal or abnormal? I'll replace all my air sensors if I have to, but I want to know whether it will make any difference first.
 

·
Registered
Joined
·
569 Posts
The fanout is normal. If you want a more reliable evaluation of whether or not you are running lean at low airflow, check your additive trims in group 32. Or get a custom downpipe with a bung for a wideband sensor.

Cool graph though!
 

·
Banned
Joined
·
3,535 Posts
For starters, my problem is more complicated than just the A/F ratio. I also need some idea of what would cause the injectors to have a 0ms duty cycle periodically, for durations up to two seconds at a shot.

As far as the A/F gauge is concerned:

All that would do is give me a readout from the front oxygen sensor, which won't do me any good above 50% throttle, because on my engine the oxygen sensors don't have a wide enough range of operation to detect the richer full-throttle A/F ratio. Believe me, I had definite plans to get an A/F gauge until I found out this dirty little secret.

Can anybody please tell me whether the ratio fanout at the left side of the graph is normal or abnormal? I'll replace all my air sensors if I have to, but I want to know whether it will make any difference first.
I know what a wide band an narrow band sensor is, no need for the lecture I was talking about a real A/F meter with a wideband O2 sensor.

http://www.grid.co.jp/en/prod/prod342.htm
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #10 ·
The fanout is normal. If you want a more reliable evaluation of whether or not you are running lean at low airflow, check your additive trims in group 32. Or get a custom downpipe with a bung for a wideband sensor.

Cool graph though!
Any idea why the fanout is normal? Is the MAF less accurate under low-flow conditions? Are the injectors harder to control at short duty intervals? Or is it something else?

If you want the spreadsheet I used to generate the graph, I could post a link if I only had a place to host files.
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #11 ·
I know what a wide band an narrow band sensor is, no need for the lecture I was talking about a real A/F meter with a wideband O2 sensor.

http://www.grid.co.jp/en/prod/prod342.htm
This would do nothing to make my car run better. The ECU will still auto-adjust to whatever it thinks is "normal". I'm interested in knowing whether my ratio is off so I can fix it, not so I can have another dashboard readout.
 

·
Founder of the STFA
Joined
·
8,667 Posts

Then, I calculated fuel consumption by taking the 100% duty cycle injection rate (210cc/min), converting it to grams/second, multiplying it by the actual duty cycle percentages, and multiplying by 4 (the number of cylinders I have). I also adjusted the flow rates (measured at the 3 bar standard) upwards by 33% to account for the 4 bar stock FPR.


The flow rate of the 210cc rated injectors @ 4 bar base pressure is approx 243cc, about 16% higher, not 33%. The 4 bar FPR is rated without vac or intake pressure applied.

What is the problem you're trying to track down? Fault codes, bad add or mult trims or is it just a reverse engineering project?
 

·
Registered
Joined
·
28,344 Posts
Your car cuts fuel when you let off the throttle. It will even shut off the injectors.
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #14 ·
The flow rate of the 210cc rated injectors @ 4 bar base pressure is approx 243cc, about 16% higher, not 33%. The 4 bar FPR is rated without vac or intake pressure applied.

What is the problem you're trying to track down? Fault codes, bad add or mult trims or is it just a reverse engineering project?
Dammit, I thought all you had to do was divide the real fuel pressure by the pressure that the injectors' flow rating is based on.

What's the real formula for adjusting the flow rate based on pressure, then?
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #15 ·
Your car cuts fuel when you let off the throttle. It will even shut off the injectors.
Completely? I didn't think the car would ever completely cut the fuel supply. Good to know that my cutouts aren't (guaranteed to be) a problem.
 

·
Registered
Joined
·
3,482 Posts
What's the real formula for adjusting the flow rate based on pressure, then?


courtesy of:
http://www.rceng.com/technical.htm

to use the formula easily set the old flow rate to "1"...the new flow rate will be a ratio relative to the old. so for example for a typical K04 system which uses stock injectors and a 5 bar fpr the increase in flow rate is about 12%.

SQRT(5 bar/4 bar)*1=1.118 ==>11.8% more fuel.
 

·
Founder of the STFA
Joined
·
8,667 Posts
Dammit, I thought all you had to do was divide the real fuel pressure by the pressure that the injectors' flow rating is based on.

What's the real formula for adjusting the flow rate based on pressure, then?
!
 

·
Banned
Joined
·
8,975 Posts
Discussion Starter · #19 ·


courtesy of:
http://www.rceng.com/technical.htm

to use the formula easily set the old flow rate to "1"...the new flow rate will be a ratio relative to the old. so for example for a typical K04 system which uses stock injectors and a 5 bar fpr the increase in flow rate is about 12%.

SQRT(5 bar/4 bar)*1=1.118 ==>11.8% more fuel.
The square root of the ratio? I'd be interested to know the physics behind that.

Unfortunately, this means my car IS running lean after all. The "new correct" average Lambda, taking into account the correct fuel pressure adjustment, is 1.12. So what I'd like to know is, why does block 032 say the auto-correction from the O2 sensors is -0.8%?

Is my MAF reading too low, or are my O2 sensors reading too high? Should I just give up and take the brute-force approach of replacing the sensors one by one?
 
1 - 20 of 40 Posts
Top