Bought the Durametric but I need help understanding the numbers
 

I received the Durametric I ordered from Pelican and now I have some actual data about my car, a 2000 S Tiptronic that has been giving me codes about running lean (1126 and 0410 most recently).

I also bought the ODBII maintenance manual which tells me the Porsche terminology for running lean - "Oxygen sensor adaptation range 2 (cylinders 4-6) enrichment limit". I found some data points in the Durametric pull-down list of actual values. I cleared the codes and took a 30 mile drive last evening. I tried to take a long log file but it apparantly stopped soon after I departed so I obtained the following values while idling in my driveway after the drive. At that point the car showed no fault codes and seemed to be running perfectly. Typically the codes return within 200 - 350 miles. Here are the numbers...

Loss adaptation idle =.08
Adaptation range 2 FRA Bank 1 =1.27
Adaptation range 1 RKAT Bank 1 =-2.58
Adaptation range 2 FRA Bank 1 =1.24
Adaptation range 1 RKAT Bank 2 = -2.39
MAF voltage at idle =1.30

Are these the right points for monitoring fuel trim?
What do "range 1" and "range 2" mean?
Waht do "FRA" and "RKAT" mean?
What are the units of measurement?
How do I interpret the numbers (what do perfectly running motors show)?

What would be the best strategy to employ the Durametric as I proceed with repairs to this car? I have had a local shop smoke test the vacuum and have a replacement vacuum line to the brake booster on order. I can see that the Durametric will come in handy if I decide but buy a fuel pressure test gauge but is there any value in activating the evap valve or the secondary air injection to see if the fuel values change? Or would those activations be useful if I need another smoke test?

Page B-10 and following of the DME 7.2 manual explains. Good luck understanding it all.

You might find the following scanned pages from my OBD-II Boxster manual handy: http://www.pelicanparts.com/techarticles/Boxster_Tech/19-FUEL-02_Sensor/OBD-II-scan.pdf

However, if you have a big vacuum leak with the brake booster, then fuel injection system's method of fuel metering will be off. So that may fix the problem right away.

-Wayne

I bought the "Boxster Maintenance & ODBII Manual" from Pelican at the same time as the Durametric but I don't see anything in it (or the Durametric unit documentation) that explains the meaning of the values coming off the car, at least in regard to the fuel adaptation. If I did this kind of work all the time at least I would know typical "normal" values. It's pretty hard to use the Durametric without understanding what it's telling me.

If anyone has a link or part number to the DME 7.2 manual I'd be interested, although apparantly it's pretty hard to understand.

The basic question I'm trying to answer is whether my car has a flaky sensor vs good sensors and significant vacuum leaks. My car always runs well and the CEL stays off for a long time when I clear it. If I have a flaky sensor I'd expect the Durametric data values to be near normal except if I caught it during the brief period when the sensor was acting up. If the sensors are OK I'd expect to find data values that are obviously abnormal and they would get better each time I found and fixed another vacuum leak. I can't really use the Durametric until I understand the data it's giving me and my lack of knowledge is frustrating my scientific mechanic mindset.

For example I see readings hovering around 0% in the adaptation data points coming off the Durametric. Some are positive values and others are negative. Totally on guesswork it seems like maybe zero is the desired optimum DME fuel compensation and the positive values mean rich compensation and negative lean (or vice versa). If that were true my car is showing reasonable levels for fuel trim and I could have a bad sensor or intermittantly bad electrical signals, in which case the repair focus should be on the sensor components rather than a hunt for leaks. I'm still planning on finding and fixing leaks regardless.

As many fuel trim CELs as these forums experience I'd think somebody must have been down my road before and used the Durametric to help with the diagnosis and repair.

These are good questions, and even the people who do this all day long have trouble explaining it. I will see if I can get a write-up done on it...

-Wayne

Thanks. I posted this on another forum and didn't get much info there either. The lead expert there was very assertive in saying my car has vacuum leaks and he could be right. Setting aside the high-tech Durametric, what kind of vacuum reading should I get at idle on a Boxster that's running correctly and free of intake leaks? I was thinking about buying an old fashioned mechanical vacuum gauge.

I've had the car smoke tested once already.

For a DME 7.2 car, code 1126 can mean lean or rich running depending on whether it is above or below the limit. Code 0410 is not a mixture error it is a secondary air injection failure.

RKAT is the adaptation value near idle.
FRA is the adaptation value under load and further divided into FRAU (lower load range ) and FRAO (higher load range) ranges (perhaps the durametric doesn't display these divisions).

The RKAT values near idle are used to change the length of the injector pulse to account for deviations from ideality (ie intake leaks) to maintain the proper O2 sensor reading. The RKAT value is an additive factor and can deviate by +/- 4.5%. The FRA value is a multiplicative factor and can deviate by 1.32 to 0.70. The positive values or values greater than 1 increase the injector pulse length to account for a lean mixture, and negative values and values less than 1 are to compensate for a rich mixture.

As you might expect the RKAT values are most sensitive to intake leaks and would be greater than 1, adding fuel to account for the false air. As you can see, you do not have an intake leak as your RKAT values are negative, suggesting that the car is running rich and the DME is trying to shorten the injector pulse to get the O2 sensor readings in spec.

In contrast, under load the car is running lean. Thus the FRA values are greater than one and near the threshold of 1.32. This lean mixture at high rpm is not likely due to intake leaks as I stated above, false air has little effect at high air volumes under load.

For your 2000S, at idle, at operating temp with no accessories on the air mass should be 10-20kg/h. At 3000 with no load in neutral the air mass should be 30-60 kg/h.

Based upon the limited information you have provided and the rich running at low air flow and lean running at high air flow I would try and clean/replace your MAF sensor. However, if you have recurring secondary air faults I would suggest repairing the other problems first before investigating the mixture problems.

-Todd

durametric is very limited on what it can do not worth the money. look for used factory scanner on e-bay

Where do I study to learn stuff like this answer! I can do a lot of mechanical stuff, but I know someday my car will have an issue that isn't an obvious mechanical issue, I'll take it to an Indy and I would be surprised if they can diagnose the way you just did.

Todd to the rescue again! Again, I think a tech article on what all of this means would be useful - I'll have to work with my buddy Tony Callas on something like this.

-Wayne

If the smoke test didn't reveal anything, then you don't have vacuum leaks. It's a pretty comprehensive test. Did it show any leaks?

-Wayne

Todd, I hate to use a cliche but you are THE MAN. Thanks very much. Is that info available to me in some kind of repair book? I'd like to buy whatever book it is, so give me a name, link or part number if you have it.

If the car is running lean under load I'm guessing the fuel system needs to be checked out so I'll buy a fuel filter and a fuel pressure test gauge and maybe perform the flow rate test shown in the service manual. The car doesn't "feel" like it's running out of gas under load - it pulls really strong all the way to the next gear on the Tip. But it has 104k miles and I can't rule it out so I'll check the fuel first, then buy a MAF sensor if the fuel seems OK.

The smoke test showed a significant leak on the vacuum line leading to the brake booster. It was rubbing the bodywork and I have a new line on order. It didn't really leak until he wiggled the line. The mechanic doing the test didn't have any Boxster experience (he didn't even know how to put the top in the service position and did his work over the seat backs). He only looked in the engine bay so he might have missed something. At my request he applied a little RTV compound to the leak to possibly temporarily reduce it until I could get the new part. I don't know how successful the silicone was.

So my plan of attack is - pressure test the fuel rail before any other work, pull the undertray, replace the vacuum line, replace the fuel filter, pressure test again to see if anything changed, clear the codes and drive the car. If the CEL comes back, which is what I expect, replace the MAF sensor, drive the car some more and get new fuel trim figures from the Durametric to compare to the original values.

Dale...don't replace the sensor before cleaning it with MAF cleaner (any auto parts store). Risk $7 instead of multi-hundreds. Then if that doesn't work...spend the big bucks.

You are running rich at idle and lean at the higher end. It is probably not an air leak as the rkat would show lean at idle.
I actually had the same values in a 99 10 days ago. It turned out that the ecu was toasted (the car was in a body shop with a dead battery). One of the symptoms that mine had was that the fuel trims would not return to zero after a code erase. (i pulled the maf sensor plug to create a code). The other symptom was that I created an air leak to drive the oxygen sensing values (which is the instantaneous lambda) very lean and the rkat would not move from the rich negative value. I also drve the manifold vacuum signal to the pressure regulator low and high to vary the fuel pressure and it had no effect on the fuel trim at idle (rkat).

Thanks for those tips. I already tried the MAF sensor cleaner with no improvement. Geez, I hope its not the DME. I know my fuel trims changed the first time I had an issue and cleared the codes. The car ran like crap for the first 1/4 mile but cleared up. I've been disconnecting the battery after attempting MAF sensor cleaning but never noticed any change in operation.

BTW, how does the fuel pressure regulator interact with the vacuum level? Less vacuum - more pressure?

One more thing - are there any systems on the car, such as the secondary air injection or EVAP systems, that might create a vacuum leak that could be concealed during a stationary smoke vacuum test but would become active leaks when the sub-system turned on? Would it be important to activate those systems via the Durametric while the smoke machine is hooked up?

Update...I haven't received any of the parts I've ordered so nothing has changed mechanically on the car. Today the CEL came back on again at 220 miles since clearing it the last time. Codes were 1126 and 1128. The 0410 code that showed the last time wasn't there. The Durametric gave the following readings, which are pretty similar to the last time.

Loss adaptation idle = -0.25
adaptation range 2 FRA bank 1 = 1.31
adaptation range 1 RKAT bank 1 = -3.98
adaptation range 2 FRA bank 2 = 1.27
adaptation range 1 RKAT bank 2 = -2.56

I cleared the codes and fired up the Durametric again immediately afterward and obtained the following values for the same points...

loss adaptation idle = -0.62
adaptation range 1 FRA bank 1 = .99
adaptation range 1 RKAT bank 1 = 0
adaptation range 2 FRA bank 2 = .99
adaptation range 1 RKAT bank 2 = 0

I'm interpreting this as good news in the sense that the car reacted to clearing the codes and therefore it's probably not a bad DME (heaves sigh of relief). The other values presumably take some time to sample data and adjust. I'm assuming I received default values since I got them within a minute or two of starting the car after clearing the CEL.

I'm going to start getting the Durametric values more often, possibly after every drive. If the numbers show a good pattern for many drive cycles eventually followed by a CEL and one set of bad values I guess that would point toward a flaky MAF sensor. If the values trend worse and worse after each drive, then eventually crossing the CEL threshold it might mean something else. Kind of odd that the car runs so long, at least 200 miles, between CEL episodes. You'd think a genuine mechanical issue would come back quicker, more like 30-50 miles. Could be wishful thinking on my part hoping for a miracle fix with a new MAF sensor, which I haven't ordered yet pending more conclusive data.

In the olden days (let's say 1985-89), the cars ran engine management systems that controlled fuel and spark, but didn't give out any trouble codes. With those cars, you have to work on the cars often in order to "feel" what the problem is, and also you have to have a shelf full of "test components" that are known good components to swap in for testing. I personally have found that era of car difficult to work on, as you need to test each component, some with oscilloscopes, etc.

With the newer cars, they started to tell you what sensors were bad / defective and indicate when they weren't reading values within the specified range. This is great, however, it almost feels like it's gone way long in the opposite direction. There aren't too many components in the Boxster fuel injection system (mass air flow sensor, crank sensor, cam sensor, a few temp sensors, an electronic throttle body, and a few others), yet the diagnosis is sometimes very complicated. If this were an older car, the test would be to swap in a good MAF sensor and see if that fixes the problem.

That said, I would do the following:

1 - test / check the entire system for vacuum leaks, in particular your secondary air injection system

2 - replace the MAF sensor with a new one

I don't think that you have a defective DME, as it would most likely not be able to send you this information, and the car probably wouldn't start or run either.

-Wayne

More new data. I drove about 25 miles last evening with the Durametric hooked up. This was the first running under load after clearing the codes. The car ran fine (always does) but the data contradicted some of the previous readings I've posted. The low load RKAT points were very near zero during the entire drive. The higher load FRA points started at .99 and gradually drifted higher to the 1.2 range.

I had an opportunity to give the car about 3/4 throttle for a few seconds and the FRA points immediately shifted to .99. In other words they adapted back to a perfect value. Then they went right back to 1.2 when I returned to steady state 40 mph. I'm assuming the fuel pressure increased during the brief 3/4 throttle period but that's only my guess. I have a fuel pressure gauge on order and that should be revealing.

Can anyone explain how the Boxster fuel pressure regulator system works? Does the engine vacuum vary the final fuel pressure? If so in what way? And does the DME turn on a second stage fuel pump output?

Also, does anyone have a theory about what kind of malfunction could make the low load RKAT fuel trims go out of whack intermittently? Every time I've looked at the data after a CEL the RKAT values are up around 3 but apparently that is an anomaly because last night's drive showed more normal values of -0.05. Here's a picture of the data screen.

http://i19.photobucket.com/albums/b1...tricapril9.jpg

Fuel pressure in these engine is affected by vacuum, as the pressure regulator has a vacuum line plugged into it. There could indeed be a problem with your fuel pressure regulator. I'll have to think about this a bit. Here's a diagram that shows how it plugs into the manifold. You might want to verify that it's still properly connected:

http://www.pelicanparts.com/techarti...mall/Pic07.jpg

-Wayne

Thanks for the diagram. The fuel pressure tester I ordered from Amazon arrived yesterday but now I'm waiting for the replacement test port cap as I understand you're supposed to use a new one whenever you take the cap off.

I've gathered the fuel trim readings from the car a couple more times after 30 mile drives.

After the first 30 mile drive following the screen shot above the higher load FRA trims were approximately 1.25 while the RKAT trims went to approximately -0.55.

After the second 30 mile drive the FRA trims stayed roughly the same at around 1.25 while the RKAT trims went even more rich at -1.50 (bank 1) and -1.27 (bank 2).

If the fuel pressure was too high at low loads and too low at higher loads it might explain the numbers. Possibly the fuel pressure regulator is not responding to manifold vacuum (broken or disconnected line?) but I don't actually know which way a change in vacuum influences the pressure. You'd think the regulator would be using the vacuum as a reference signal, more vacuum, less pressure needed and vice versa. If that is right a disconnected vacuum line (or broken internal diaphragm) would tend to give too much pressure under load load, high vacuum conditions, which is just what I'm seeing. Truth is I'm not really a mechanic so all this is guesswork.

Another possibility would be some kind of weird interaction with the vapor recovery system that continuously fed the motor with fumes at low load (stuck open EVAP valve) but that wouldn't explain the lean condition at higher loads. Of course it could all be a flaky MAF sensor. The sensor readings look perfect at idle though. I'm going to visually examine the fuel presure regulator vacuum line today just to see if something obvious is wrong.

This Sherlock Holmes detective work is sort of fun in an analytical way but I'd really like the car fixed so I might just throw some parts at it and hope for the best.