Just get engine hot then do it all parked!

Idle LTFTs, then rev 2500 or so and look again.

 

Well, went out and cleaned the MAFS. Let it sit for a while. Then went for a bit of a drive without hooking the connector back up, car was not happy, and readings were really weird. Realized what I had done. Went for another drive. Still getting high positive readings for LT and ST, but the ST is not going below -10% anymore. So I ordered a smoke machine, and will see what I find. May be a helpful tool to have in any case, just have to rig up a battery of some kind.

The trouble with just reving the engine, is that most of the numbers are relatively normal there. I'll occasionally get a real quick spike of the ST, but it goes away quick, even if I hold the throttle open at 3 grand or so. It's only when I have it out on the road that I get it to really climb. I'm not sure what that means for leak potential, but I guess I'll find out.

I did do a trip log this time, and both brought the data into google earth, and a CSV file into a spreadsheet. I honestly am not an expert (or know much at all), but I'm not seeing any correlation to anything else when the fuel trims start going high. MAFS data doesn't seem to be a factor, acceleration isn't usually a factor, engine work load, or rpm don't seem to be factors. I will say it is interesting to me that acceleration seems to affect the MAFS data more then speed. I even included o2 sensor data, but that didn't seem to correlate either.

I'm still less then a tank in from filling it up myself with 91 octane, if the previous owner was driving with 87(don't know one way or the other), would that mess with any of this?

 

So long as they've settled you can ignore STFTs.

They migrate into LTFTs (the clue is in the names) and should be near zero (i.e. have settled, oh I just said that).

So, that's why LTFTs are examined.

Hot engine parked means:
1. nice stable situation, easy to compare with others
2. easy to do

Idle means low air flow (through the MAF). Rev 2500 means quite high air flow.

Any air leak will show up as raised LTFT(s) at idle.

Other issues may or may not show up but at least you can rule in or out a ton of things AND IT'S EASY

 

While you are cleaning your MAF sensor, be sure to check on the connector.
There have been cases where the pins make slightly less contact, sending erroneous info to the ECU.
On the X308 forum, there is an interesting thread going on, where the TPS setting caused havoc.
Was your's ever replaced or worked on?

 

Thanks again for working with me on this and educating me. I had originally written another response to this, and thought I had posted it, but didn't and then I went and did some more testing and reading. When relooking over the data earlier, and this seemed to be confirmed by another lap, was that the engine had to be under load (at least 20%) before I would get high LTFT readings. From doing some reading on that condition, and looking at the data, I'm thinking I probably do not have a significant air leak (although I will still probably test, since I ordered the smoke tester already), but instead probably have an issue with fuel. I did measure fuel pressure on this latest trip, and that seemed OK to me, but I'm no expert, so maybe someone else can let me know if it sounds correct. Most of the time both at idle and while driving they stayed at 55psi +/-1psi, including most of the time I had high LTFT readings. But under high acceleration it did drop down to about 51psi for one reading, and under a bit higher deceleration it went up to 63psi for one reading, but in both cases normalized soon after. I don't know where the fuel pressure is measured, but to me this seems like the fuel pump is probably fine, and if the filter is before this measurement, it is probably fine as well. So does that point to possibly partially clogged injectors? Or is there somewhere else that I should look? The thing is both banks are reporting about the same thing, so does that mean I have injectors on both sides that may be partially clogged? If the previous owner didn't use 91 octane gas, and I started to, would this cause this issue? I had almost forgotten, but I had also added a 12 oz bottle of Techron about a tank ago, could that cause any of this?

Oh, the MAFS was reading about 30-40 g/s most of the time it was at 70mph, but did go up to 80 g/s during hard acceleration, and dropped down to about 6 g/s when idling or not under any load. I don't know if that clears that sensor as a potential issue, or not.

Thanks again for all the help. I don't consider this a crippling issue yet, more an annoyance, but I would like to solve it, especially before it does become a bigger issue.

 

MY maf reading under hard acceleration is 312.08 g/s ..

But this was WOT from a standing still start.
Dont know what your hard acceleration condition was but 80 g/s dos'nt seam a lot.

 

Some quick general remarks:
- During the years, I noticed that the LTFT readings fluctuates much quicker on our cars than you think, some 1/3 of the rate the STFT change.
That is making it harder to draw conclusions.
I also observed that the underlaying average LTFT only really settles after a long trip (50-100km), even after a hard reset.

- The MAF should read some 5g/s at hot idle, that seems correct.
However, your maximum MAF reading should be some 320-330 g/s for a S/C engine, your readings are farrrr too low.

 

Caution: the car stores many LTFT readings per bank. It is a more advanced thing to start looking at them than just the ones I've stated.

 

Remember, fuel trims are values stored as the "height" of a 3D elevation map, with RPM and Load as the other 2 coordinates. Then, there one such map per bank. It means that for a range of rpm (I do not know how wide), and a given Load, there is a trim on each bank. Then, there is another trim for the next band of RPM, same Load. Then another number for the same rpm, but a different Load. To my knowledge, Long term trims are updated every 2 minutes with the average values of the Short Term trims over those same 2 minutes, at the same rpm and Load. Over time, these Long Term trims become what they become, and the Short Term trims average zero (assuming no new air leaks develop). In other words, there is no such thing as a single trim number for a given engine. This is why you can have high Long Term trims at idle and low Long Term trims at a higher rpm a few seconds later. They are just different values in a matrix, each one "evolving" on its own. I hope this helps in your diagnostic.

 

Can you share what you found? I am also on the market for an inexpensive smoke machine. I have also looked into re-purposing an old soldering iron to make smoke with baby oil in a jar, and pipe it with a bicycle pump for air pressure and an outlet for the smoke:

BEST-Automotive-Smoke-Machine-You-Can-Build

 

Thanks folks. I had a long day at work today, and possibly a few more this week, so I don't know when I will next get to look into some of these things. I will probably take a second glance at the data tonight and try and figure out how hard my accelerations were in comparison to the low MAFS values I got.

The smoke machine I got didn't seem any better or worse then many of the lower priced ones you see on eBay or the rest of the web. It looked to have everything, and it was on the lower end of the price spectrum, but otherwise I just made a choice that it was easier to spend $100 to buy a tool then it would be to try and make one. Here is the amazon link.

 

I knew this, I knew this ...... and never made my brains to connect the two ....
Thanks for the revelation of today!

 

Looking at the data again I guess I better recalibrate my butt dyno. My "hard acceleration" was about 0-60 in about 10 seconds. I guess that is about twice as long as it should have taken, but it seemed fast. This was in sport, but it seemed to take a long time for the first shift, then the second shift came right away. I don't know if that matches up with my MAFS reading of 80 g/sec or not, but it was what it was.

 

Got home and had a little present in the mailbox. My new license plates. The reason this was particularly cool, was because my temporary plate was in my rear window, and that meant I couldn't drive around with the top down. So, I had to go for another lap around the country block I live on, with the top down.

I got a bit better acceleration this time, but I had the peddle floored, and still only got about a 0-60 in 8 sec (according to the gps) but I did get it to rev a bit higher before shifting up to 5200 rpm, and a load of 67%. This also gave me a bit higher MAFS reading of 225 g/sec. The fuel pressure also bounced around while doing this acceleration, down 44psi for one reading then up to 62psi 2 readings later, before settling down to the normal 55psi. As far as the fuel trims, I don't know. The LT ones still seemed to climb higher than I've been lead to believe they should be, maxing out at about 17%. But that is down from the last few times when it hit 19%, and they still like to live in the lower teens for the few miles I was at 70MPH. This is again a few percent lower than it was 2 days ago, but higher than I'm lead to believe they should be at. But this time when I kept it under 55 on the back roads with a engine load barely hitting 25% and an rpm never above 2k the LTFT stayed under 8% for almost the whole time (much of the time closer to 0), with just a few occasional readings going up over 10%, last time it still liked to hit mid teens on these same conditions.

Looking over the stats, I also really just noticed the vacuum/ boost gauge readings. Should this help tell if I did have a vacume leak, or not really? It bottomed out at just below -13 for this trip under that good acceleration. I know this can be influenced by air density (barometric pressure was at 29.77 in and the temp was about 68F), but I still thought that seemed like a good number.

I'm still thinking that this seems to be more a fuel issue then an air issue, or even a MAFS issue, but you folks know more then me. A new fuel filter seems like a relatively cheap, but tedious project. If I have to start looking for a slightly clogged injector that seems like a possibly more costly adventure. I'll wait a tank and then send another bottle of Techron through the system, see if that helps anything. And I'll still use the smoke machine when I get some free time, but it is less a priority now.

Otherwise I think all I need to do now, is stop over thinking it, start driving longer trips, and just see if it just all sorts it self out on its own.

 

What results for posts #21, #23 ?

It may be fuel but let's do the basics before you assume anything.

 

Sorry, I had written that up in the post that got deleted, and then forgot to reinclude it in the new one. At a warmed up idle of 650 rpm, then revved up to 1650, and 3400 (each held for at least 30 sec, the 3400 for close to a minute. The LTFT's were both between 1.5 and 4.5%. Bank one STFT was about 1% the whole time. Bank two STFT did very more; starting at 0, going up to 3.125, and then eventually 7.8125%. This is what lead me to thinking there was a correlation between engine load (had to be at least 20%), and the really high LTFT numbers I'd get while actually driving, and what had thrown the engine codes earlier. And this lead me to start thinking it was possibly more fuel related. Although I do admit that the STFT on bank 2 of over 7.5% at the higher revs does stand out now on looking at it again.

 

At this point, it would not hurt checking the basics:

• The AFM works by sampling. It does not measure everything. The condition of the air filter is critical as it can affect the path of the air. Maybe just replace it.
• There is an o-ring around the base of the air flow meter. Sealing this off is critical. Make sure this o-ring is there and not pinched/broken/missing.
• The XKR air box has a special flap to admit more air under acceleration. Check this is operational. Maybe the ECU expects this and "corrects" air flow readings.
• Check the intercooler water pump works. It is on when the accessories are on (engine still off). It is by the air box and apparently hums pretty obviously. Hot water jackets will not help.
• To my knowledge, there are 2 critical seals on the supercharger that had a poor initial design and slipped out of position under pressure, creating air leaks. A later design corrected this.
• The fuel pressure is measured relative to the intake pressure. There is a vacuum line that forwards intake pressure to the sensor (at the end of the fuel rail on an XK8). If this line is somehow cracked, the fuel pressure will net incorrect.
• Try and get a reading of air temperature (off of the same air flow meter) AND manifold pressure. If the pressure reading is off, fueling will be off (Pressure sensor has a vaccum/pressure line to it I think)
• Try and get a sense of the condition of the exhaust gaskets. Fresh air intake there will fool the O2 sensors. Any weird ticking? Maybe try and blow smoke in the area to look for leaks.
• Definitely change the fuel filter. Beware that the parts store have this part wrong, as it lists a filter with barbed fittings when it needs the quick-connect type. You need the filter listed for the S-type 4.2.

Just my thoughts. I have an XK8, so no direct experience with the particulars of an XKR. Hope this is of any help.

 

So long as the engine was HOT (rather than just warm) then if the LTFTs at idle were also quite small you don't have an air leak at idle. Were they also small?

It's then worth looking at things in more depth to see why it ever flagged lean codes - the freeze frame data is there to help. Bear in mind in this scenario that "lean" just means there was unused (aka excess) air unexpectedly in the exhaust as seen by the O2 sensors. You can't tell, so far, if there was any unburned fuel as well. So you'd be trying to find out if you've a lack of fuel or a failure to burn it.

 

Very nice list!

Would like to add the following:
* The problematic seals of the S/C were in the earlier models, should be ok in your car,
* Check if your TB is clean,
* Check the S/C pressure (easy to tap into the hose on the right side of the S/C housing),
* This will also confirm if the S/C pressure valve (located under the housing) is working / fully closing,
* Check if your TB has the full opening range (~2% at idle, till some 92% at full throttle),
* Check if the ignition has full advance (Torque Pro, from memory the range is ~4 till 35 or something).

However, after that, if you have such an imbalance between the left and right bank, I should concentrate on those items which are bank related.
* Injectors (actually never read of problems of blocking on a x308),
* O2 sensors,
* Cats.

I did not read through all your posts, but your problem brought back a project I handled 1 1/2 year ago, it should still be lingering around on the x308 forum.

One of my friend's Daimler Super V8 (4 liter), had several of the same issues; slow acceleration at ~9-10 sec, a slight vibration at idle, no mid range push, trim levels unbalanced (but no codes), and some others.
We checked about all the possible causes, only to finally open up the covers.
The timing on the left bank was about 1 tooth off, probably because of an improper job done while they replaced the CAM tensioners (previous owner).
Re-timing fixed everything, making it the smoothest and most powerful x308 of the many he has.
Not saying this is your problem, but just as one for your check list.

 

I have air and fuel filters coming on Saturday, and when I find some time I will start working through the lists. I do appreciate that info.

I’d like to think I’m mechanically decent, but I’ll admit, I’m more of wrench turner then a true technician, especially on more modern cars. I’ve always done a lot of work on my cars, but I’ve never had to delve to far into the engines. Sure I can replace a starter, or alternator, or I’ve even rebuilt old motorcycle carbs. But getting this deep is a bit new to me, so I really do appreciate the education you all are providing. A big part of my issue with some of this, is because I’ve never looked this far, is I’m not sure what is normal, and what isn’t, and same with this particular vehicle. But, since this is my “fun” car, I figure it’s a good one to learn on.


Thanks again folks.