Help - Knock, LTFT, and low mpg (stock cal) - 3400

[chri0029]

My fuel mileage has been lower than it should be for a little while now. The mileage was especially poor in the winter. I do the basic routine maintenance like I should. I did an initial OBDII scan/log with a different program before I got HPT. My LTFT levels were always really low at idle (-15 to –20%). So I did the basic troubleshooting (checked fuel line pressure, compared fuel pressure drop in each injector, air filter good (new K&N filter), no air induction restrictions that I could see, sprayed MAF sensor screen clean, etc). No help. I recently bought HPT to begin to do some tuning and scanned a trip (I want to figure out what the current problem is before I start any tuning). I was definitely noticing some KR with my normal 87 octane (up to 9 degrees of KR at times, mostly at loading – hills, etc). I then tried a tank of 91 octane fuel. I did see less, there were still times that my engine did KR.

I did some reading already and learned the V6 knock sensors are particularly sensitive. My PCM calibration is stock (for now) except I lowered the fan-on temps. I am thinking the LTFT leaning-compensation, KR, and reduced mpg are all related but I can't quite pinpoint why. I have no codes, so I assume no EGR problems. Please let me know your thoughts. I have attached 2 HPT scans: 1 with 87 octane and the other with 91 octane fuel.

[Azrael]

How old is the O2 sensor? It's possible it's indicating rich when it really isn't and the pcm is leaning the mixture as a result which can cause knock under load.

[3400-95-Modified]

Azrael is on the right path, I was trying to think of why there would be a high negative adjustment to the fuel and also cause KR and I just couldn't pinpoint it right away till I read his post, but that's a definite possibility.

[chri0029]

That was my initial thought as well. Was the primary O2 sensor defective or was there something 'tricking' it into reading rich? I did try a different O2 sensor with the same result (at least with LTFT). I did't have the ability to read KR before. So that is why I am thinking something is possibly doing the 'tricking' the PCM into adjusting lean (i.e leaky injector, etc.)

Plus, an old O2 sensor wouldn't necessarily read incorrectly, right? It would just have less cross-overs and/or slower response time.

I wonder if possibly the knock sensor is overly sensitive or defective. I went to a used car lot and took a different Grand AM (same year, etc) for a test drive and hooked up my laptop. It also read low LTFT at idle but not as extreme as mine. Maybe just a coincidence? I didn't read knock. My thoughts are that somehow the very negative LTFT and knock are not necessarily related?

[robertisaar]

O2 sensors CAN be skewed and still look to read correctly.

and from what i'm told, GM went nuts with the timing on the late N-bodys with 3400s and they could generate quite a bit of KR stock.

[86FieroSEv6]

Improper PCV valve? Dirty oil? Nasty spark plugs, poor plug wires? Exhaust backpressure issues? EGR leaking?

[chri0029]

Thanks, Guys, for your help!

Improper PCV valve? Dirty oil? Nasty spark plugs, poor plug wires? Exhaust backpressure issues? EGR leaking?

New PCV valve, new plugs, new wires. EGR leaking could be a possibility. Regarding the backpressure, that was kind of what I was thinking as well; maybe a partially plugged converter.

and from what i'm told, GM went nuts with the timing on the late N-bodys with 3400s and they could generate quite a bit of KR stock.

This could be true as well. I found an exact used car on a lot and took one for a 'test drive' and was surprised to see the LTFT's low (but not as extreme as mine). I might need to find another car on a lot and test for knock. Or now since I have HPT, I have to look on this site where you post scans and look at some similar late 3400's.

Good stuff. I'll do some more testing and let you know what I find.

[Azrael]

The coolant sensor can also affect fueling to a degree. Very easy to diagnose, compare its reading to the IAT sensor's whenever the engine's cold.

[86FieroSEv6]

The normal "range" for fuel trims can be -10 to 10. I personally prefer to see them -7 to 5, but with higher mileage engines it can push closer to the range limits. Have you taken a cold and hot cranking compression reading (throttle held wide open, don't forget to disable spark and fuel for the test)? Carbon buildup can rear it's ugly head. If the intake valves are carboned up, it acts like a sponge and the fuel will "drip" off the valve causing the leanout of the air fuel charge even though the fuel is hitting the cylinder. This then causes knock and a false rich to lean ratio. I know we're reaching a bit here, but you seem to have covered the basics already. As for the PCV valve, did you use the correct Delco valve? I have yet to meet an aftermarket unit that was properly calibrated for GM engines. They work, but not perfectly.

[chri0029]

Thanks for all the good info. After a small break (swiching vehicles with my wife for a few weeks) I'm back on diagnosing this thing again. I didn't do any compression test yet. But the PCV valve comment is interesting. I thought they were all the same ..? I have a cheap AZ one on there now. Would that affect my mpg that much?

I am really thinking the problem is related to a somewhat plugged CAT. My engine seems to run hotter than it should, there seems to be a little drop in power compared to what there used to be, definitely less mpg, and I had a head gasket issue awhile ago which the engine burned coolant (which the AF silicone could have done its share on the cat). Since I will be doing more performance stuff in the future, I am contemplating a small circular cat replacement since there is no sniffer test here in MI.

I'd rather go properly diagnose it first. I did a quick log (attached) between idle and 2500 rpm. Then engine idles with around 9-10" of Hg at idle, then at 2500 rpm drops slightly to around 7" of HG and stays. Not significantly different than idle, but enough..? What do you guys think? I can't find any engine logs posted in the 'downloads' of this site to see if that's considered normal or not. Also look at the LTFT (~ -18 or so). Not normal for basically a stock cal.

One more comment about my mpg: I only average 23-25 if I baby it on the fwy anymore. When I first got the vehicle, my best freeway trip was 31 mpg. On country roads (45-55 mph), it's about the same: 23-25 (used to be around 27). It seems to be worse in the winter when it's cold. Avg 21-23 mpg. Again, not horrible, but I really think it should be better.

[Azrael]

9-10 in-hg at idle isn't right. It should normally be between 16-20". At ~60mph I see 10-12". You might have a possible vacuum leak.

[chri0029]

Yea, I thought that didn't seem right either. From what I read, idle MAP should be at 18 to 22". I wasn't sure if for some reason the 60*V6 engines had a special intake configuration or something which lowered the MAP.

But wouldn't that be an intake restriction rather than a vacuum leak? I will try pulling the TB air duct off and confirm I get a higher MAP.

If that doesn't work, my thought is that the exhaust has high backpressure, or the cat could be plugged.

[bob442]

maybe the timing chain is getting loose? my car had 90,000 km, and regular oil chainges, and stock everything, and it had a bit of slop, then again, the CAT just plugged up and was covered under warranty, but what i noticed with the plugged CAT was that the intake was termendously loud, and the exhaust was extremly quiet, great bottom end but after 3000rpm it fell flat on its face, and fuel consumpsion was getting worse, and power lacked more prior to it suddenly getting really plugged up.

[chri0029]

Okay, I was really convinced before my CAT was plugged, now I'm not so. Over the weekend I pulled the airfilter and opened the airbox plus disconnected the exhaust from the manifold. Basically the same results: 8-9" of Hg at idle, 7-8" of Hg at 2500 rpm.

Now I'm stumped. I don't believe this is normal, but I haven't seen any other logs. Can anyone else send their engine log my way? Either .hpl (HPT) or exported (xls or csv) for a somewhat stock engine?

Regarding the engine performance (assuming the MAP is normal), the next step I think is unfortunately pulling the injector rail to check for possibly a very slight injector leak. I think that could 'trick' my O2 sensor into reading false rich readings and causing the computer to lean the engine out (-15 to -20 LTFT) and give me my lower mpg and more knock than normal..??

The other suggestions are good as well, but I still wonder. I think I would notice more performance loss if my timing chain had slipped a gear. Remember that the car really drives pretty well, but I only slightly notice a problem with knock & mpg. I think the kicker is the very negative LTFT at idle (maybe this is normal - I haven't seen other logs, but I doubt it). Regarding the compression check, I had the head off for a CHG about 30k miles ago and there was very little carbon then.

[chri0029]

Okay, I took-off from what I started writing here: http://60degreev6.com/forum/showthre...563#post433563

Even though this is more of a performance-related topic, I am still trying to improve on mpg. Now I'm just experimenting with my cals since I am getting more comfortable with this.

Update: I installed the 28# injectors and recalibrated my IFR accordingly. Now my fuel trims are significantly closer to 0 than they used to be. Plus I don't see the PCM adding a lot fuel at high load where it did before. My thought is that my old 100k + mile injectors simply needed to be replaced. A few of them were probably partially plugged so the computer had to compensate to make-up for the uneven fueling & unburnt fuel in the exhaust. That probably was why my trims were off. I went with the 28# ones only since I want to do more performance stuff in the future.

Now I am playing with the AFR and spark advance. I have run in OL and have changed the AFR to 14.3 (to compensate for the 10% Ethanol) and upped my timing advance to where I get > 4 degrees of KR in some cells. Then I can change the spark table accordingly (lessen the KR areas, advance the non-KR areas)

Next I will try the same with maybe 15.3 AFR and see what difference it makes.

I have 2 questions regarding this: I have been reading theories about 180 degree thermostat temps not properly atomizing the fuel. Then, on the flip side, others can get more advance with cooler temperatures. 3400-Mod, did you see less mpg as a result of the 180 T-stat?

Secondly, what is the 'proper' way to change AFR? I could only do this in the COT area of HPT since maybe my vehicle doesn't have the AFR table in the 'general' area of fueling (only cold start). But, then again, don't your O2 sensors return it back to stoich anyway in CL? Hell, I'm probably running at 14.3 AFR stoich due to the 'up to 10% Ethanol' winter blend. So do you have to run in OL all the time (which defeats the mpg issue) if I decided to go to 15.3 perminantly?

Sorry, I'm just a rookie tuner. I'm just trying to make sense of all I have read and put it to good use.