When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
1990 IROC 5.7
Over the winter I swapped in a TKX, Dakota box for the VSS, changed intake gaskets because of a rear china wall oil leak.
The car ran perfectly before this.
Now I have stumbling at part throttle while driving, idle inconsistency, and lean condition as per the spark plug.
Fuel pressure is 41lbs.
I changed the O2 sensor and idle went from 850 rpm to 1100 rpm
Timing is 6* with EST wire unplugged
I've been in contact with Brian from Tuned Performance and have sent him some Data Logs and he's not seeing what could be the issue from them.
Brian has been very helpful and very patient with me.
Some question I'm been thinking about.
Did I wire something wrong?
Did something go bad with the ignition system?
Test3 never gets into closed loop.... min temp for closed loop is 45C in the factory bin (assuming your previous tuner didn't change that parameter), and your data shows the coolant was 50C during that run, so you should have been in closed loop. Certainly could have been the O2 sensor.
But when you put the new O2 sensor on, it went into closed loop.
It could be that whatever the problem is now, it was also present before the work you did, except it takes closed loop operation for it to manifest.
Do all of your plugs look that stark white? On both sides of the engine?
In closed loop, I'm seeing a BLM of 118, which means the ECM is pulling fuel out for some reason. It's about an 8% reduction in fueling.
The ECM corrects only for the driver side of the engine (assuming your O2 sensor is on the driver side header). Which means the passenger side may be getting partially starved of fuel while the ECM compensates for what its seeing on the driver side.... which may also explain some of the hesitation.
In open loop, you're typically running richer and there's no correction by the ECM. So it could have been masking the issue previously.
I sent you 2 more files.
These are when I was out for a drive with the original O2 sensor.
I shouldn't have sent Test3 as I was learning how to use tunerpro and had just started the car for practice.
I'm confused why the idle changed with just the change of the O2 sensor.
Maybe a clue?
When i hooked up the cruise control it would not disengage with the electric switch on the pedal. It would disengage with the vacuum interrupt switch.
Could this have something to do with the Dakota box?
Can I drive the car without the Dakota box disconnected?
Thinking about disconnecting the Dakota box for test.
To my recollection, the only effect the VSS would have in fuel would be DFCO mode, and not at idle nut a cutoff RPM of around 1,100 stock.
I don't know what DFCO is?
If I was to disconnect the Dakota box and have no signal to the ICM, could I still drive it and be able to see if that has anything to do with the issues I'm having?
I'm thinking...
If it ran perfectly before, then it should have something to do with what I did during the TKX swap and intake gasket install?
If I was to disconnect the Dakota box and have no signal to the ICM, could I still drive it and be able to see if that has anything to do with the issues I'm having?
DFCO => Decel Fuel Cut Off
You're saying your ignition control module is running off the Dakota box?
Generally without the VSS hooked up, there's some low speed fueling parameters that are set up for low speed, low throttle operation... also the ECM won't know when you're at idle vs non-idle (idle is considered up to 15 mph). Cooling fan will never disengage... among other speed related parameters the ECM needs the VSS for.
I suppose you could try it and see if something improves. But if it doesn't or actually runs worse, then it may be due to lack of VSS.
Originally Posted by BigBlock73
I'm thinking...
If it ran perfectly before,
I'll be the judge of that
Let me look at your data tonight and see if can detect anything.
Though when you're driving around, can you see the vehicle speed in the Tunerpro data? How do you know the ECM is getting the right signal off the Dakota box?
So your 7-14-24 log shows a speed of 0. Was that because you weren't moving? Or is the VSS not functioning?
On your latest logs, your BLMs are all over the place, but largely down around 110-108.... so either the previous tune wasn't that great or something happened. But you're saying it ran perfectly like that?
"So your 7-14-24 log shows a speed of 0. Was that because you weren't moving? Or is the VSS not functioning?"
7-14-2024 log was done in the Garage for me to see what the new O2 sensor was doing.
With the new sensor it idled at 1100 rpm, with the old reinstalled it idled 850 rpm.
I still don't know why.
"You're saying your ignition control module is running off the Dakota box?"
The Dakota install was a large part of the wiring I changed along with clutch pedal neutral safety switch. Removing the auto shifter was part of that.
Thinking about how I can backtrack to see if I wired something wrong.
"On your latest logs, your BLMs are all over the place, but largely down around 110-108.... so either the previous tune wasn't that great or something happened. But you're saying it ran perfectly like that?"
It ran great all last summer before I did the swap.
I don't have any data logs from that time.
On your latest logs, your BLMs are all over the place, but largely down around 110-108.... so either the previous tune wasn't that great or something happened. But you're saying it ran perfectly like that?
What are BLM's and what does them jumping around mean?
After the swap and intake gasket change, before the first test drive,. I started the car and while it was running sparks started shooting out of the alt. I replaced it with a AC Delco new unit.
Could this have done some damage to the electronics or somehow affected the tune?
BLMs are an indicator from the ECM on how it's trimming the fuel, over the long term. ECM is always trimming fuel every millisecond, and you'll have very short rich/lean conditions that it continually corrects for... and that's normal.
But if you have a malfunction that's causing the ECM to continually add more fuel (or take it away) over a long period of time, this BLM value will start to shift up (rich) or down (lean). 128 is the ideal number, indicating that the ECM is seeing nominal conditions.
Lower than 128 doesn't mean you're running rich.. it means the ECM is pulling fuel out to avoid running rich. And above 128, it's the opposite.
108 is the minimum (i.e., it's out of adjustment range). Once you hit 108, and the ECM can't pull out anymore fuel, that means the engine really is now running rich.
A leaking or faulty injector, too high a fuel pressure, among other things... can cause the BLMs to drop as the ECM is seeing more fuel coming in than it's commanding.
Or, if the tune is not correct (i.e., VE tables are not correct for the application).
What does your chip look like? Is it just the factory memcal module by itself? Or is it one of those chip adapters with the memcal plugged into it and another chip on top? That'll impact what hardware you need to buy. For software, Tunerpro is free (although they wouldn't mind contributions). Tunercat is also good, but you have to buy it.
What's the p/n on the injectors? It should be molded into the plastic bodies? also a picture of them? If they weren't properly tuned for, that could be a contributor here.
It plugs into the side of the memcal with the PROM on it, then the other side plugs into the PROM reader.
And a prom reader, like an AutoProm or some other EEPROM reader. Moates is out of business, so the Autoprom is discontinued... though people are selling used ones from time to time.
I spoke with Southbay and they said the ones I have are correct for my 5.7.
I can send these back for cleaning at $180 or purchase another equivalent set for $220
I spoke with Southbay and they said the ones I have are correct for my 5.7.
I can send these back for cleaning at $180 or purchase another equivalent set for $220
It'd be curious to see all the driver side. If one cylinder is running rich, it'll cause the ECM to lean out the rest of the engine to compensate. So you might have one brown plug and 7 very white ones.
I'll pull all the drivers side plugs tomorrow, that's the easy side.
ULTM8Zthese will these allow me to write as well as read?
BoostedNW Memcal adapter header T48 TL866-3G High-speed Programmer EPROM/MCU/SPI/Nor/NAND Flash/EMMC/ IC tester
The 27S512 chips you might be able to find a little cheaper on ebay though. My suspicion is the chip on your memcal isn't electronically eraseable and I wouldn't even advise over-writing it even if you could. Instead, simply plug the entire module into the G1 adapter and then you can put the new chips on the adapter and it'll bypass the chip in the memcal.
I checked the fuel pressure while it was running and had 41 lbs. I left it attached and noticed the pressure dropped to 20 lbs. after a couple of hours
Good compression
good spark
Fuel pressure drops from 40-20 with fuel lines not pinched in 10 minutes.
Fuel pressure drops from 40-38.5 with fuel lines pinched in 10 minutes.
That pinched lines test pretty rules out leaking injectors (at least in terms of when the engine is off).... One injector could potentially be malfunctioning while running.
One thing you could also do is remove one injector plug at a time on the driver side and see what the BLM does with on injector not functioning. BLM should rise considerably each time since you're introducing unburnt oxygen into the system (remember, the O2 sensor bases it's reading on unburnt oxygen... it does not look for unburnt fuel). You should see similar behavior each time. If one cylinder produces very different results, that injector could be suspect.
I wouldn't suspect a bad plug wire or anything like that... incomplete combustion will result in unburnt oxygen and therefore the ECM detecting a lean condition, not rich. Each time I've had a bad plug wire on the passenger side, I had BLMs into the 150-160 range.
So I'm looking at your individual cylinder logs and what I'm noticing is this...
You're falling out of closed loop for long periods of time. It's making it difficult to detect anything because the ECM stops trimming the fuel. Not only does it mask the problem in the data, but it'll also make it difficult to detect running differences "by ear" when you remove an injector plug because in open loop the ECM does not react to any changes.
You said you're running headers.... are you running a 3 wire heated O2 sensor? Or the stock single wire? If the single wire, the O2 sensor is likely cooling off and causing the ECM to fall out of closed loop at idle speeds.
Or are you running the old sensor and it's going bad? In which case put the new one back in?
Tunerpro will display open or closed loop status but you kinda have to navigate to that. A simple way to tell is that when fully warmed up, if you see the INT return to 128, and/or the commanded AFR become something other than 14.73, then you know you're not in closed loop anymore.
Before sending another set of data, see if you can figure out the closed loop issue and ensure that it's remaining in closed loop during the test.
Ok, so if the new sensor does the same thing, then it's a problem with the sensor cooling off. Then it won't buy us anything to take any more data unforunately.
At that point, you're kind of at a cross roads...
1.) Convert to a 3-wire sensor. Delco AFS-74 is the preferred one. You'd need to run a switched ignition wire to the sensor. The ground side of the sensor can simply be routed to a bolt at the back of the cylinder head. If you're running stainless headers, I'd use a 4 wire, which has a dedicated sensor ground (I ended up doing this with my stainless headers).
2.) Take a chance and simply replace all the injectors and see what happens. If you're saying it ran perfectly before, then this issue with falling out of closed loop at idle wasn't really noticeable from the driver seat. If the new injectors solve the issue, then you can potentially ignore the O2 sensor problem (not ideal but maybe satisfactory).
Ok, mostly solved. It does drop out of closed loop for few seconds toward the end of the run. Long term, I'd probably look into a 3-wire sensor.
But for now it seems like it's good enough to keep troubleshooting. Go ahead and send over the one-log-per cylinder test again. But keep an eye on the commanded AFR and INT values... should remain at 14.73 and 128, respectively. Minor drop outs of a second or two is ok, but if it's dropping out for long periods of time, it'll make the data not as useful.
and again, keep an ear out for changes in engine behavior. Does one cylinder react wildly different than the others?
Ok, there's still the issue of falling out of closed loop....
The only cylinder where I can see a clear reaction by the ECM is #7.
#1 => You're in open loop for about half of the log. If you pulled the injector plug and put it back during that time period, then we didn't capture that event.
#3 => I didn't see any reaction for the first 3/4 of the log, after which point you fell out of closed loop again. (I'm assuming you did pull the injector plug before it fell out of closed loop)
#5 => Stayed in closed loop the entire time. No reaction. Again, I'm assuming you did pull the injector plug.
#7 => This is the only cylinder that does what I"d expect. INT (short term fuel trim) shoots up to 146 and drags the BLM (long term fuel trim) up to 140.
So here's a possible theory...
Injector #7 is bad
When you pull the injector plugs off 1, 3, and 5, the bad #7 injector is still overwhelming the oxygen sensor. My suspicion is that as soon as you pull the #7 injector plug, the ECM gets back to seeing somewhat normal results from the other cylinders.
140 BLM is not the ideal... the ECM having to trim the fuel richer to make up for a perceived lean condition. But I don't know what your BLMs were before this, so I can't say whether 140 was your nominal state (that is dependent on your VE tables in the tune).
You could potentially repeat #1 and see if it follows the results of #3 and #5. But you have to make sure the ECM is in closed loop during the time you remove the injector plug, otherwise the data is useless.
07-17-2024, 06:38 AM
