Testing with noid light
I’m chasing a mis in the 94 setup. I think it’s random misfire of injectors.
I’m comparing tests on both cars to see differences. On the 94, if I pull the flashing noid light out of the injector harness with engine running and plug it right back in, it no longer flashes. It won’t flash again unless I turn the engine off and back on again. On the 96 I can pull a flashing noid light with engine running and plug it back in and it will begin flashing again immediately even without restart of the engine.
I understand the 94 is ODB1 and the 96 is OBD2. Does this explain the difference? Is it normal operation of the 94 PCM to cease providing ground pulse for an injector that has been unplugged? So I’m assuming the 94 (OBD1) requires an ignition off and restart before the pulse from the PCM will be restored, and the 96 (OBD2) will restore pulse as soon as the injector connection is restored, no restart required?
I’m chasing a mis in the 94 setup. I think it’s random misfire of injectors.
I’m comparing tests on both cars to see differences. On the 94, if I pull the flashing noid light out of the injector harness with engine running and plug it right back in, it no longer flashes. It won’t flash again unless I turn the engine off and back on again. On the 96 I can pull a flashing noid light with engine running and plug it back in and it will begin flashing again immediately even without restart of the engine.
I understand the 94 is ODB1 and the 96 is OBD2. Does this explain the difference? Is it normal operation of the 94 PCM to cease providing ground pulse for an injector that has been unplugged? So I’m assuming the 94 (OBD1) requires an ignition off and restart before the pulse from the PCM will be restored, and the 96 (OBD2) will restore pulse as soon as the injector connection is restored, no restart required?
Hmmm, any idea why the 94 might cease the injector pulse after unplugging the noid and not resume it until after ignition shut down and engine restart?
If you plug the harness back into its injector, does the injector fire? Can you hear a difference?
I have done the long screwdriver as a stethoscope method and #5 has a weak tick compared to the others during the missing, might even be hearing #7 tick bleeding over rather than a weak #5. And #5 is on the drivers side where the exhaust popping eventually happens.
So im pretty sure I at least have an issue with #5 injector but I still was wondering why the ground pulse from the PCM wont restore until after restart. This cease of ground pulse will happen on any of the cylinders I try it, so that interruption is surely based in the PCM, for whatever reason right or wrong.
I'll fiddle some more and post back
Here's something amazing. I unplugged the entire left bank (drivers side) consecutively and the engine still ran and idles. Obviously a bit rough but back at the tailpipe it was clear there was still "some" combustion happening even with all left bank injectors unplugged.
Matter of fact, I can even unplug #2 and #4, along with the entire left bank and it still didnt stall.!!! Only #6 and #8 plugged in, it didnt stall out!!! Ran really rough, but sustaining a struggling idle. Crazy huh?
I'm thinking I gotta have multiple leaking injectors.
I had replaced the injectors in the vette a few years back so I know they are good. I'm gonna swap them over and see what happens. The 94's injectors are the stock Delphi we know to be problematic, or so Ive heard.
If the result of the first test is "injector shuts off, injector resumes function normally" and the result of the second test is "no change, injector remains off," that points to a wiring problem somewhere. But the tests could also have more interesting, chin-scratching results. That's the next thing I'd try.
The change to newer injectors helped, especially after the PCM had some time to relearn. But something still isn't quite right. Smells too rich and its eating alot of fuel, even just in the garage idling and whatnot. This PCM had been sent out and tuned for the minor changes that were made during the swap and to eliminate VATS and other non-essential things. I wonder if the tuner set it a tad too rich. I sure wish I had a wideband on it then I could actually see whats going on with the A/F ratio. I probably wont bother the tuner anymore but I'll learn to modify the tables myself if needed.
My first successful endeavor using the ALDL was today using a free program called Scan9495. Anyway I captured two videos of engine data and the links are below. The first is 30 seconds of data at idle speed of 550 rpm. The second is similar length video of 1000 rpm no load. Do you see anything abnormal in these? Ignore the DTC, its AC circuit open, not relevant to anything I'm chasing.
The Best of Corvette for Corvette Enthusiasts
Switch to that. Highly recommended.Flashhack can also be used to easily retune these PCMs. No need to physically ship a PCM anymore, which is nice.
The thing I see automatically just by glancing at your screenshot is that your BLMs are way way way off. 143 means you're dumping tons of fuel. 138 means you're dumping a lot of fuel. Target is 128. So your O2s are reading a serious lean condition and the PCM is trying to compensate by dumping fuel in.
I highly recommend you perform the injector driver test, followed by getting a datalog with EEHack. You can then ZIP the .eedata file you save from EEHack, attach it to a post here, and I can then load that into my EEHack to analyze it if you want.
But yeah. Small moves.
When I talked to the tuner prior to sending the PCM there was a possibility I would be changing to 30lb injectors. I had a set of 30’s but discovered one of them was bad. I went with 24’s instead and I’m pretty sure I told him that. If the tuner actually programmed for 30’s, but I’m using 24’s, would that cause the o2’s to read lean and tell the PCM to fatten up, like we see in the data?
Also, what about placement of the o2’s? This LT1 complete combo came from a bone stock 94 Z28. So now in the wagon with long tube headers, the o2’s are in the collectors and farther from their normal location in the stock Z28 exhaust. Could that be a factor in o2 reading if the tuner did not compensate for it?
Will EEHack let me check what the tuner did and change if needed?
Last edited by JD1964; May 24, 2021 at 07:52 AM.
When I talked to the tuner prior to sending the PCM there was a possibility I would be changing to 30lb injectors. I had a set of 30’s but discovered one of them was bad. I went with 24’s instead and I’m pretty sure I told him that. If the tuner actually programmed for 30’s, but I’m using 24’s, would that cause the o2’s to read lean and tell the PCM to fatten up, like we see in the data?
Also, what about placement of the o2’s? This LT1 complete combo came from a bone stock 94 Z28. So now in the wagon with long tube headers, the o2’s are in the collectors and farther from their normal location in the stock Z28 exhaust. Could that be a factor in o2 reading if the tuner did not compensate for it?
Will EEHack let me check what the tuner did and change if needed?
EEHack can only read the datastream. If you want to find out what the tune is, you need to download it. The same developer has a program called Flashhack that will let you do exactly that. The tune can then be loaded into TunerPro and analyzed. To do that, TunerPro needs a definition file called an XDF, and--you guessed it--that same developer provides a free XDF for the 94-95 PCM.
EDIT: As for the O2 placement, I do not have real-world experience on what changing the placement will do practically speaking to your closed-loop trims, so I cannot say how much of an effect that will have in your specific example. I can say, however, by comparing the 95 manual Corvette BIN with the 95 manual Z28 BIN, that there are indeed some parameters regarding O2 sensor response times that are different. The problem is that they aren't well-defined in terms of, "if you moved the sensor, this is what you should set these exact parameters to." There is a parameter specifically commented in the code as relating to the delay between the sensor reading and the physical exhaust gas from a combustion event hitting the sensor, and indeed, the first value in the vable differs between the two cars. The parameter is at 0x26FB, and it's called Base O2 Integrator Delay Vs. CL Mode (O2 INT Delay - Airflow). The important value is 00, apparently, since the other values are exactly the same. On a Corvette, it's set to 0.4 seconds. On a Z28, it's set to 0.8 seconds. On its face, that would seem to suggest that the Z28 O2 sensors are further away from the exhaust ports than on a Corvette. I don't own a Z28, so I cannot confirm that.
Last edited by Nomake Wan; May 24, 2021 at 09:01 AM.
Diagnostic Aids: If an injector is disconnected and then reconnected while the engine is operating, the injector driver will be disabled for a predetermined time and a DTC 18 will be set. When the driver is disabled, an engine miss will be apparent.
DTC 18 does not turn on the SERVICE ENGINE SOON light.
Now that said, if you use EEHack you should be able to force the driver active again by clearing DTCs then flipping it off and on again in the injector control section of Controls.
But yeah. Again, sorry for not just RTFM! Looks like this is in fact totally normal!
The details I remember giving the tuner were as follows
Stock LT1 from 94 Z28 except for 1.6 full roller rockers and long tube headers (host vehicle 79 Malibu station wagon)
93 octane
24 lb inj
Delete VATS, smog, AC, EGR
moderate firm up to the 4L60E upshifts > stock stall speed
3.73 rear ratio with 26" tire
Last edited by JD1964; May 25, 2021 at 09:05 PM.
The details I remember giving the tuner were as follows
Stock LT1 from 94 Z28 except for 1.6 full roller rockers and long tube headers (host vehicle 79 Malibu station wagon)
93 octane - AFR targets and PE have been adjusted, but I'd have to run the interpolation math to see what final AFR they're actually targeting. I have no doubt that it'll be leaner than stock, but that's because stock PE is pig rich for a reason. Leaning it out is expected and encouraged. Burst Knock has been disabled. Maximum knock retard has been fixed at -9 degrees for normal mode (stock goes to -15) and -7 degrees in PE mode (stock goes to -12). Spark Advance has been changed across the board, which is to be expected. In general, they added 2 degrees to every cell over a 94 Z28. There are however manual adjustments to a few cells that are more/less. But the vast majority of the overall table is just +2 degrees.
24 lb inj - Interestingly they chose to shoot for a slightly larger injector than stock. The stock '94 Z28 uses 24lb injectors and has its flow rate set at 24.912 lb/hr. Yours is set to 25.377 lb/hr. The MAF and VE values were unchanged from a '94 Z28. This will cause a slightly leaner than stock condition if the injectors are the same as stock. The injector voltage offsets are still set to stock '94 Z28 values.
Delete VATS, smog, AC, EGR - This was mostly done. EGR is disabled, the AIR pump is disabled, and VATS was disabled. Technically VATS still has a single setting enabled, but it appears that this setting is redundant. Unless it's actually causing you grief I wouldn't touch it. A/C however was not deleted. All of the error checks regarding the A/C system remain intact. These are DTC 61, 66, 67, 68, 69, 70 and 71. They should be disabled if the entire A/C system has been removed so that the PCM isn't trying to test for its presence. Additionally the "A/C Clutch Disabled TPS%" should be set to 0.0, whereas currently it's at the stock value of 99.6. The Charcoal Canister Purge system was also left in, but that's because there are a ton of functions related to normal engine operation that are tied to it. Disabling the CCP system causes a shitload of funky crap to happen, so that's why they left it intact in the code.
moderate firm up to the 4L60E upshifts > stock stall speed - Okay, here's where things get spicy. Kickdown Mode has been set to RPM-only; normally it's a factor of speed and RPM. Instead, the speeds have all been set to lower values, so that the PCM will only shift based on RPM. The RPM has been set to 5450 for 2 and 3, and 5400 for 4. So at WOT, it will always upshift at 5450 RPM (except 4th gear where it'll shift at 5400). Maximum line pressure has been increased from 90 PSI to 125 PSI. Main Line Pressure tables have all been increased by about 33% across the board, but the real kicker is the line pressure per gear table which has been MASSIVELY increased. Stock values are between -16 and +6, new values are now between +15 and +90. That'll certainly firm things up! Similarly the line pressure offset vs TPS% has increased just as wildly......except in manual shift mode, where the table was left entirely stock. Odd. 1st gear overrev protection has been disabled (it won't shift you out of 1st if the shifter is in '1' and you're going too fast for 1st). Normal shift points have been adjusted down so that the transmission will shift slightly earlier than a stock 94 Z28.
3.73 rear ratio with 26" tire - This was adjusted in the speedometer section.
Before I try and change anything in this tune, I have a spare 8051 I salvaged from a 94 Caddy Fleetwood. The first thing I’ll try is turning off the VATS. I assume it’s still on anyway. As far the starter cranking in the Malibu, I’m using the stock Malibu circuit. When I plug the Fleetwood PCM into the car and turn the key, it fires up right away but cuts out after about a second. I assume it’s the VATS still enabled on that PCM. I should be able to easily turn that off with Flashhack and TunerPro RT, right?
Last edited by JD1964; May 26, 2021 at 11:35 AM.
Before I try and change anything in this tune, I have a spare 8051 I salvaged from a 94 Caddy Fleetwood. The first thing I’ll try is turning off the VATS. I assume it’s still on anyway. As far the starter cranking in the Malibu, I’m using the stock Malibu circuit. When I plug the Fleetwood PCM into the car and turn the key, it fires up right away but cuts out after about a second. I assume it’s the VATS still enabled on that PCM. I should be able to easily turn that off with Flashhack and TunerPro RT, right?







