3rd gear pull... weird results

[tstkl]

so I've been tuning these 1000cc fic injectors for about a week, and finally was able to make a pretty clean 3rd gear pull. My new computer doesn't have a serial port, so I can't hotsync my palm, so I'll just discribe the log.

starting at 2687 rpms with a tps of 99.6% and timing advance of 30 degrees, my o2 volts were .87, with 1 count of knock (just a guess, timing) and ending at 7093 rpms at 25 degrees timing with 3 counts of knock and an o2 volt of .83, (the injector duty cycle is 48.4% with 1944 hz...but thats with the maft.. so thats probably wrong

The reason why Im posting this is because of what I felt while making the pull. The car feels sluggish and somewhat slow from 2687-5k rpms, but at 5k rpms, there is a sudden, extremely significant pull that continues on up until 7k rpms. The first thing I noticed while looking at the logs is that at 5k rpms, the knock count has shot up to 14, and trails off to 3 at 7k rpms. This is weird to me, because the timing advance at 5k rpms is 13, it goes down to 11, then goes up to 25. Upon further inspection, I noticed the narrowband o2 sensor values (btw, I have a wideband, but it stopped working misteriously, I'll be attempting to fix that on tuesday with paul) start at .87, go to .91 by 2906 rpms, and then hit .85 at 4187 and stay there til 7k with one exception. The one exception is that at 5k rpms, 5062 to be exact, they dip down to .83 for a second, then go back to .85. This happens twice more at 7062 rpms and 7093 rpms.

So to me it looks like I need to richen up the 5krpm zone by a very small amount, and lean almost everything else out.

following the injector duty cycle, there are slight changes throughout the pull, which may be due to airflow changes, but may also be the cause of changes in afr.

the whole pull took exactly 14 seconds. 14.000 seconds.

any thoughts as to why when I have the lowest timing advance and the most knock, I feel it pull the hardest?

 

[GVR4592]

Can you post the datalog?

 

[GVR4592]

Whoops I missed the first part about the hotsync.

 

[pneumo]

Just to clarify for others, tstkl has an eprom chip set for 1k injectors and extended maps, plus a maft to tune it.

Usually when you see the O2 volts blip up and down it's because the engine is misfiring slightly. When it misfires the oxygen remains unburnt, so the O2 sensor reads it as a quick burst of lean. These rich misfires can also lead to knock since portions of hot unburnt gas stay in the corners of the combustion chamber where they're more likely to cause knock on the next cycle. Naturally, misfires don't produce any power, so the car feels sluggish until it clears up.

I'd recomend leaning it out by another click and try another pull. Keep logging and see what happens. Could you post airflow and timing advance for the whole pull? I have a feeling that airflow per rev is peaking around 4-5krpm, which puts you in the richest part of the map.

 

[tstkl]

I'll do it the only way I can think of at the moment, digicam...

that really, REALLY didn't work, you couldn't see anything. So I'll just try and discribe the airflow and timing better.

timing:
between 2687 and 3375 the timing is decreasing from 30 to 17 degrees, where it turns around and starts going up to 20 degrees at 4000 rpms. this holds until 4406 rpms, and starts decreasing at that point to 12 degrees at 4750. From there it goes up one degree at 4906 rpms and again turns south, to 11 degrees at 5062. It stays at 11 til 5250 then proceeds upwards at a non constant rate until it hits 25 degrees at 7093.

airflow:
I will be using the airflow values displayed on the logger, which is half the airflow value that the maf is actually seeing.
At the start of the log, 2687 rpms, the airflow is at 225 hz, and is almost always increasing. Ok, this is really hard to discribe, but basically at around 5k rpms, there is a small jump in airflow, which flatens out til 5500 rpms, at this point it starts increasing and then has a small dip and flattens out until the end of the log where it dips a little and stays there.

the trouble is, the boost isn't very constant in these pulls. It flys to 20 lbs where it sits for a while, then it starts fluttering between 21 and 20 lbs, as if there is a flux in the exhaust or something...maybe some unburt gases getting in there. I'll try and get the logs posted, but seeing as I've never been able to do that before, I doubt I'll get very far.

 

[ForcedNduckshn]

If I were in your position, I would hold off wracking my brain too much until you get the wideband o2 calibrated and working properly. You could still be running off-of-the-chart rich, and then maybe leaning out a bit around what would probably be peak torque for you. I assuming you are running a large frame turbo w/ the 1000cc injectors? I would get the basic fuel tuning done first, perhaps shoot for 11.0 and see where that leaves you in the logs for fuel and timing, then make adjustments from there to combat any knock.

 

[tstkl]

14b POWA!!!!!!!!!!

I got the 1000's for later...

as for the wideband, I have no idea why it doesn't work, so worst comes to worst, I have to buy a new wideband unit for 150 and find someone with a laptop I can borrow to program the outputs in again. Im getting an error code 1, which is a short in the system, my guess is the sensor itself went out. Calibrating the damn thing takes two seconds, so that isn't an issue whatsoever.

 

[tstkl]

dave, another click puts the wot at A and the mid at D, thats -30% and -15% respectively. I only have one more click to go on wot before I have to go into rpm mode and start taking out fuel across the board. Any way I can convince you to burn another chip with all the same data and just corrections for a leaner afr up top? I think I might just stick to pump gas on wednesday then, since I don't have many options for leaning it out on race gas.

I'll make some pulls tomorrow and borrow my friends laptop after we go to the gym.

I noticed my engine bay had an unusual amount of oil sprayed all over the place, Im assuming its my breather filter, and me just not used to actually making wot pulls in this car, but she was low about 1/3 a quart of oil, so Im somewhat worried, although I still have like a 1/3 a quart left over after filling her up, so I think I should be fine with some often checks at the track. I really should get a catch can.

 

[GVR4592]

14b POWA!!!!!!!!!!

I got the 1000's for later...

as for the wideband, I have no idea why it doesn't work, so worst comes to worst, I have to buy a new wideband unit for 150 and find someone with a laptop I can borrow to program the outputs in again. Im getting an error code 1, which is a short in the system, my guess is the sensor itself went out. Calibrating the damn thing takes two seconds, so that isn't an issue whatsoever.

Sounds like you have the LC-1, I would look for something different if you plan on replacing it.

 

[pneumo]

I'm tracing the load level based on your rpm/timing data. The log dips into the highest load level around 4750rpm, where it's between level 7 and 8 out of 12. The timing progression looks normal, so I'd say the maft is reporting accurate airflow numbers. The fluctuations you're seeing in boost and airflow are obviously related. Something I've learned from tuning my 14b to run 12.0's is that high timing helps keep boost up, misfires will lower boost. The jump in airflow at 5k rpm is also where the power picks up and I'm guessing where the misfires stop.

Tracing the load levels on the fuel map shows a fairly consistent AF ratio of about 9.8:1 Before you freak out, that's actually a little leaner than stock. See if you can get the wideband working so you know what's going on. After that I'd be glad to make some adjustments to the fuel map.

 

[pneumo]

http://dsm-ecu.com/archive/msg03255.html check this out, will those sensors work with your wideband unit?

 

[tstkl]

whats wrong with the lc-1?

and dave, Im taking the car to paul on tuesday and were going to try his wideband sensor on my lc-1 and see if the thing still works. If not I dunno what else it could be except the unit itself. I think it was probably the sensor though, as I dropped that one like three times, and calibrated it before every start up (normally 2-3 days between each start up of my car)

Hopefully I just have to replace the sensor, and the garage I work at uses only bosch parts or whatever, so Im sure I could get a sensor from work with the employee discount.

I dunno if innovate used a generic plug, but I know they used a bosch sensor.

http://www.innovatemotorsports.com/xcart/customer/product.php?productid=16169&cat=253&page=1

 

[pneumo]

I don't know why I didn't think of this before. I was probably too focused on tuning the chip. You car shows all the signs of having a boost leak: It's lean at idle and gets very rich on boost. Have Paul do a leak test while it's at the shop.

 

[tstkl]

that might be because my maf is mounted on my uicp and the chances are quite slim, but I'll have paul check it again.

and its not lean at idle, its CRAZY. Honestly, what you saw while we were sitting there wasn't even half of it. The low fuel trims go from 81.2 to 130 whatever (the max) on a daily bases. And sometimes it even idles at 850. But most of the time it sits on 1100 rpms....

I think injectors this big are just not designed to run well at idle.

 

[larsrya8]

I think injectors this big are just not designed to run well at idle.

Most DSMLink users rave about how well their 1000's idle... especially FIC 1000's. They do require some adjustment to get them running correctly, but once they do it's smooth sailing. I don't know if the GM MAF-T can offer you fine enough tuning to get them dialed in like you can with DSMLink.

 

[tstkl]

heres the first pull,...

http://www.geocities.com/sdktheta/log.xls

 

[tstkl]

leaned it out 5% more, more knock, higher o2 voltage, dave said it might be because the car isn't misfiring from being too rich, so now it doesn't have all that extra fuel supressing the knock, yet its reading richer because its not misfiring. Either way, Im taking it to the track today and I'll probably take out the other 19% just for kicks and giggles....

 

[tstkl]

holy hell batman, my wideband decided to work today.....

the results were no good though, at idle, the afr's go from low 13's to high 19's, and at wot, from 0->~3k rpms the afr's MAXED OUT at 21.9:1. After that, they shot down to low 8's and low 9's, then once they hit about 5k rpms, it was up to 13's again.

overall, it was too sporatic to really make an sence of.... Im HOPING that is due to the fact that the sensor hasn't been calibrated since it stopped working (instructions say calibrate once every 1-2 weeks.)

Its at least some insight to what is going on, since it reflects the o2 trims pretty well sometimes...

 

[tstkl]

Changed my mind, the wideband wasn't working accuratly. I ran some test and hooked it up to the laptop and the results were just TOOOOO off to be right.

ANYWAYS

leaned the car out another 5% and took some more logs to about 5k and saw this: less knock, and much, MUCH leaner o2 voltages. Before I was seeing around .87, now I was seeing .80-.82.

before anyone says don't tune off the narrowband o2 sensor, I already know that, Im just using it as a guidline. I don't know why I had less knock though with such a lean afr, and why 5% made such a huge difference. I think I might be right in between the sweet spot, so I'll hook the tach wire up and start using the crappy rpm mode the maft comes with to tune. Also, I have some other logs I could post, but they are long and I need to trim them down, which I need excel to do....

edit: fuel trims settled in at 81.2% high, 95ish% mid, and the low will never settle in...

 

[tstkl]

funny stuff:
www.geocities.com/sdktheta/wtfmate.txt

I think I know whats happening though, it was uber hot today and the gm maf has no intake air temperature sensor. Everyone keeps telling me its easier to tune without one, since thats one less variable you have to deal with, but in my opinion, I think getting an iat would be very helpful. I want to put one on my throttle body elbow so I can know what my 14b is up to temp wise.

www.geocities.com/sdktheta/3rd1.txt

thats a 3rd gear pull with (I think) a slightly richer afr and a boost creep of over 20 psi. Again, thats a guess, since it very closely matches another more recent log I took, but the file got corrupted somehow, so I can't get it off the logger.

both logs are with a straight vac line going from the throttle body to the wastegate, and Im hitting 10 psi by 3100 rpms. The creep varies from pull to pull greatly.

 

[pneumo]

Odd logs, for sure. It looks like the wideband a/f ratio and the O2 volts are saying about the same thing. They don't deviate from each other much. I'd recomend turning the mid knob richer by a click or two. That'll help richen the area where you're running lean around 2500 to 3500 rpm.

I wouldn't worry about that part at the bottom of the log where you let off the gas and get back on it again- the last few datapoints on the 'wtfmate' log. I can see the WB and O2 volts show vastly different things, but when you quickly lift off the throttle and quickly open it again the readings will lag behind slightly. It's possible the WB has more or less delay than the ECU signal, so don't expect the readings to line up exactly.

 

[tstkl]

or the fact that the wideband o2 sensor is much farther down the exhaust track than the narrowband o2 sensor. Either way, you did a hell of a job with that new chip and I honestly can't thank you enough for your time.