Brian: Loaded the altered binary onto the flash chip. Ran the engine with it and it seems really behaved at first. Pushed the pintle back to full open position, started the engine, ran at higher RPM for short time, then settled down to good idle RPM. After it warms up at idle, goes from BPW of 1.x to 2.x. At the same time the wide band O2 says it is going to nearly lean cutoff, approx 16-17:1 AFR, then back less that 14:1. Does this in relatively slow cycles but continuous at idle and at the same time retards timing to after TDC according to timing light.
Suspect, maybe radiated interference from spark plug wire to TPS and MAP wiring which are on top of spark plug wires. Am going to splice in extensions of TPS and MAP wiring to reroute away from spark plug wires and the run BJYL-test-rsicard.bin again to see if the cyclic idle situation goes away. What does CCP stand for? Darn TLAs.
Made a recording of the run and attached to this post. Also attached image file of binary comparisons.
CCP is charcoal canister purge. Part of the emissions control. I will open your datalog and have a look. Will post what I see in next post
<QUOTE>Have yet to hook up wideband O2 sensor.</QUOTE> <QUOTE>At the same time the wide band O2 says it is going to nearly lean cutoff, approx 16-17:1 AFR, then back less that 14:1.</QUOTE>
Sorry, I am confused here. What exactly did you mean? Is the wideband hooked up or not? The reason I ask is that there is considerable debate about using a wideband in place of the narrowband O2 sensor as an input to the ECM. I looked at your data log, and your crosscounts are acting a bit strange. At times you get crosscounts, and at other times it stops. Also something to consider is the distance from the head to where the O2 sensor is attached. If the stock location is like my truck, it is about 4 inches from the head. If it is installed in the collector on a long tube header, the time difference between making a fueling adjustment and checking the O2 sensor to make sure it happened may be a problem. Just trying to throw out a few things that haven't been mentioned. Once you get your rich problem corrected, you will need to address your timing problem. Disable knock counts (being careful not to get too lean while tuning). There was a rather lengthy discussion on this forum about the difference between TP and actual timing that EagleMark did some work with, it might be worth your time to look that up.
Last edited by jim_in_dorris; 03-30-2015 at 04:33 AM.
Square body stepsides forever!!!
Well, the PCM never went into closed loop during that log. Did you disconnect the narrowband O2 sensor when you connected the wideband? Another thing that just came to mind.. Do you have any exhaust system past the oxygen sensor? If the sensor is getting air other than exhaust air on it, the readings will be all over the place. On this log, they werent and the O2 voltage just flickering around .440 makes me think you didn't have it connected.
Brian
Brian, go back and look at his other log. You may be on to something, in that log, the O2 sensor mostly stayed above .660 and cross counts only happen occasionally. on most O2 sensors, the crossover is about .450 volts
Square body stepsides forever!!!
Jim,
The log test9.xdl had good O2 activity. Just mostly on the rich side and BLM bottomed out at 90. Crosscounts were active too.
Last edited by sturgillbd; 03-30-2015 at 04:45 AM. Reason: Added crosscount activiy
Brian: I disconnected the NB O2 sensor and ran it that way. The wideband is still connected and working properly. There is an exhaust system downstream of the NB and WB O2 sensors. They are both near the junction of the exhaust manifold and approx 6 inches further down the header pipe. The two join each other on the RH side and then go through a muffler so as not to get the neighbors upset. Quite sure there is only exhaust passing by the O2 sensors as the exhaust is completely enclosed. Have read that VE tuning should be done in open loop mode.
Still getting cyclic BPW at idle along with Wideband AFR going high 16-17:1 then corrects itself just before dying and with larger BPW and retarded timing approx 10 degrees ATDC. The timing and BPW go together during this cyclic issue.
Also lengthened the TPS wiring and rerouted, relocated MAP Sensor such that its wiring is away from spark plug wires. Need to run it again and will connect the NB O2 sensor and keep the WB O2 sensor in the other pipe side. Just hoping the cyclic issue goes away. We'll see.
Comments appreciated.
The VE table not being close will cause the surge. It was so rich on the other bin, I pulled quite a bit of fuel from it. It has to have the narrowband O2 sensor connected to make fueling corrections. It can be tuned open loop with just the Wideband data but wideband data isnt in your log. You might consider using the emulator portion of your APU1. Wouldn't have to keep pulling and burning chips. You can make adjustments and see the results immediately.
Crosscounts are when the O2 sensor voltage goes above and below aprroximately .450 volts. At approximately .450 the fuel mixture is stoichometric or at 14.7:1 AFR for straight gasoline. If you open the data dash while the engine is running and you are datalogging, you can see the crosscounts as they happen. You can also see the BLM etc.
Many thanks again. Pulling fuel from the VE tables has helped. At first startup, it was running very well. Next I will attempt to connect the Innovate LM-1 AFR DC voltage out into the APU1 autoprom. Then do another data recording to see the Wideband O2 data correlation with the remainder of the parameters. Then get setup to emulate. Can one emulate, record and data trace at the same time or is the data trace not necessary? Need to get a lot better with TunerPro ins and outs.
Yes you can emulate ,view data and record at the same time.
87 4Runner, 15" spring lift, 3" body, chevy vortec 355, 5.29 gears, 38.5x15.5x15" Boggers, 280hr, 16168625 running $0D
93 S10, 36x12.5x15 TSL's, custom turbo headers, 266HR cam, p&p vortec heads, $0D Marine MPFI with 8psi boost.
05 Silverado, 2' lift, 4" exhaust, Bully Dog programmer,
Bookmarks