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.
Whenever one goes out and data logs, does one wonder what to do with the information? Or how to read and interpret it?
I've had ppl sending me incomprehensible Tuner Pro chart's freeze frames or raw giant excel .csv files. How do I figure out things?
I always like to have the .XDL raw file from TunerPro. From there, I use the monitor screen and display one data at the time to check my basics. It's like going to the physician for annual check-up. Here's a perfect example that happened this week. I was reviewing a 45 minute log and came upon a weird chart for battery voltage. I was looking for another unrelated problem but still checked my basics.
First, I noticed a drop in voltage that was unusual.
I combined different data with it and found that the drop in voltage was RPM related. Normally, a drop in voltage can be observed at idle or low rpm but not when revving at mid/high rpm.
Did a close-up and confirm the voltage drop with higher rpm
I would have never noticed this while driving along. Upon further investigation, this turned out to be an early sign of a failing alternator. I changed the alternator and everything went back smooth and flat!
I have many examples, where simply looking at one data at the time on the monitor, I have identified:
Failing MAF
Faulty thermostat
Faulty knock sensor
Faulty TPS/Throttle Body
Restrictive air box
Faulty O2 sensor
Improper O2 sensor location
Excessive knock retard
Faulty ECM
Corrupt data
Over rich/lean condition at WOT
And many tuning errors leading to excessive SA, improper OL timer, AFR, VE, IAC, etc..
My point is, make sure you review each data on the monitor first, before you start tuning. You will be surprised the amount of stuff you will observe that makes no sense.
In my case, I actually detected an early sign of alternator failure while I was looking for something completely unrelated. It's a form of prevention really. Better change it before it completely fails while 100miles away from home. Just like going to the doctor for a sore throat and they discover you have high blood pressure!!!
Also, when tuning, a steady voltage is important as well as minimal variation in CT and MAT.
Here's another interesting one. Just came in yesterday from a TGO member. Voltage spikes when over 3000 rpm. This is new compare to our baseline from previous logs. Spikes up to 15.7v systematically. Voltage has always been steady in 13s.
Intermittent bad MAF sensor on LB9 OEM. The MAF voltage was zero for close to a minute after start-up. It led to stalling upon cold start-up. ECM would still show a MAF reading in g/sec.
UPDATE:
After checking the whole MAF circuitry including some bench testing, no fault was found. After reinstalling everything, the MAF started working properly. There was no loose/corroded connections. It's one of them TPI'S gremlins!
Engine running rough chasing its idle upon cold start.
Cause: ECM not reseting after ignition off for 1 day. This caused the ecm to skip the Open Loop Start-up Enrichment and went directly to open loop AFR. Instead of 10.9 commanded AFR it ended-up being 14.03... way too lean.
Detected by confirming "ENG run time" did not start at zero which led to further investigation into open loop routine. Always confirm "ENG run time" starts at zero (basics)
Excellent observations and guidance. I like to simply turn ignition on, not run, and log the sensor values to make certain they are “nominal” or set properly, ie O2, TPS, IAC, etc.
Excellent observations and guidance. I like to simply turn ignition on, not run, and log the sensor values to make certain they are “nominal” or set properly, ie O2, TPS, IAC, etc.
Yes! Excellent advice.
I do that all the time. Check TPS range and MAP for Barometric Pressure (on speed density). I also like to see what’s going on during cranking.
I like this one. A bunch of corrupt data. What a nightmare to sort-out. You can tell that the Max & Min are all over the place. That's another quick trick to detect problems: Max&Min analysis on each data.
Air intake track restriction (on Speed density system). Detected by recording MAP before engine start-up and comparing with MAP at WOT. Again some sort of basic test I like to do. Recorded almost 2inhg of vacuum at WOT!!!! Established a baseline by running open element. Custom made a new air box with zero restriction. Felt the throttle response was way better.
Old air box set-up
MAP before start-up is 102.0kpa (confirm with Weather Network for my area)
MAP at WOT drops as low as 96.0kpa
Delta = 6.0kpa = 1.78inhg
Vacuum of 1.78inhg at WOT
Open element, no air box at all
MAP before start-up is 102.0kpa (confirm with Weather Network for my area)
MAP at WOT drops as low as 99.7kpa
Delta = 2.3kpa = 0.68inhg
Vacuum of 0.68inhg at WOT
New air box set-up
MAP before start-up is 102.0kpa (confirm with Weather Network for my area)
MAP at WOT drops as low as 99.7kpa
Delta = 2.3kpa = 0.68inhg
Vacuum of 0.68inhg at WOT
Intermittent bad MAF sensor on LB9 OEM. The MAF voltage was zero for close to a minute after start-up. It led to stalling upon cold start-up. ECM would still show a MAF reading in g/sec.
I read this whole post, AWESOME INFORMATION with graphic examples. Keep it up if you find more! Especially when you also explain what the fault of the problem was is very helpful.
Did you ever figure out what was causing the zero voltage on startup to the maf??
I read this whole post, AWESOME INFORMATION with graphic examples. Keep it up if you find more! Especially when you also explain what the fault of the problem was is very helpful.
Did you ever figure out what was causing the zero voltage on startup to the maf??
IAC going down to 1 step, and further in the log, to 0 when warm idling stationary. ECM could not keep commanded idle of 750 rpm.
Solved by adjusting TB.
I have seen this issue with the 58mm throttle body on a stock TPI engine. The IAC goes to 0, idle stays higher than commanded and throttle body is at lowest setting. Conclusion = TB is too big for engine!.
In the screen shot below, one can see some random corrupt data (SA). It happens sometimes when data logging and can be a pain in the a** if one is trying to do some analysis on MAX-MIN and averages.
This one was tricky. Again just by reviewing the basics, one data at the time, I discovered that the TPS voltage would start at a certain lower value, but settle a bit higher during the run. The initial voltage is the lower baseline value. It was 0.510v in this case. However, when the driver revved the engine a bit at idle while stationary, the TPS settled at 0.569v. Later the TPS would sometimes go back to 0.510v. It appeared it was doing it when the gas pedal was released from a mid position or slowly released. Further testing and investigation revealed at sticky TB.
There is so much for me to learn. I did not know you could do this kind of troubleshooting with datalogs You obviously know allot about datalogging and analysis. Keep up the good work.
Here's a dyno run during tuning of AFR and SA at WOT. Notice AFR is a bit lean because the exhaust had been beefed-up without retuning power enrichment AFR (PE). 4-6% more fuel was added after 4000rpm. It made a difference.
Notice how RPM bounces around? Could never find out the source of the problem. Suspected ICM failure, but a new ICM did not improve things. The problem was present with the original LB9 305 TPI and aggravated with the 383 TPI. Noting is noticeable while driving though. Engine performs well.
TPS = red
RPM = purple
SA = light blue
WB AFR = dark blue
MAP is not represented but held steady at 100kpa
I left out other data like CT, Voltage for example not to jam up the chart.
To get more details, more testing and investigation was done. I extracted the data and examined it with Excel to give more resolution. Here's the post:
I disabled HWY mode, but forgot that HWY mode has twofold
AFR + SA
Tuner has to disable both. In this case, I only disabled AFR and forgot SA. So, upon reviewing my basics, knowing I had maximum SA of 38* from tables, I was surprised to see 43.9*. Highway SA mode snuck up on me and added 5.9* according to table. Conditions had been met: CT over 193.5, Max LV8 of 125, minimum RPM of 1600 & delay timer of 30sec. Now since max SA was set at 41.8*, having 43.9* commanded SA yield 41.8* max.
Always check one data at the time try finding anomalies.
Here's a 36 minute log with SA never going over 38* and under 6*... like the "Tuner" requested.
Here's a complete review of a 383TPI during dyno pull. I used 3 charts to illustrate, otherwise it gets too crowded. I am looking for anomalies and data that deviates from tuner's input (commanded data).
1) FUEL & SA
RED = RPM: sweep from around 2.5k to 6k. Notice the bouncing pass 4k.
YELLOW = INJ DC% = (INJ PW / (120/(RPM*2)))/10: as expected, all normal
LIGHT BLUE = O2 sensor MV. A generally accepted conversion from MV to AFR at WOT is: 830mv = 13.0:1, 880 mv = 12.5:1, 930 mv = 12.0:1
GREEN = Commanded AFR in PE: as expected, all normal
DARK BLUE = SA-TDC = the actual SA the engine will experienced: as expected, all normal
PURPLE = WB AFR gauge: We ran a bit lean from 12.8 target on this one. Exhaust had been beefed-up without retuning power enrichment AFR (PE) or VE.
Knock retard is not illustrated since we did not run a knock sensor. However, I still monitor knock retard to make sure ECM is not pulling any SA for whatever reason. In this case, It confirmed that it was completely disabled as expected. It read 0* for the entire run.
2) AIR
RED = RPM: sweep from around 2.5k to 6k. Notice the bouncing pass 4k.
YELLOW = Vacuum InHg: as expected, all normal
LIGHT BLUE = MAT Deg.C: as expected, all normal
GREEN = Estimated airflow in CFM. This NOT air mass in g/sec or lbs/min. Formula is (CID*(RPM/2)/1728)*VE. In our case, it is adjusted to take into account that VE tables were artificially lowered to avoid going over 100%. The $8D - '7730 system does not allow that. Example: (383*(5850/2)/1728)*.664*32/25.98 = 530cfm.
DARK BLUE = IAC: as expected, all normal
DARK PURPLE = VE: as expected, all normal.
LIGHT PURPLE = MAP kpa: as expected, all normal.
3) OTHER
RED = RPM: sweep from around 2.5k to 6k. Notice the bouncing pass 4k.
YELLOW = speed MPH: as expected, all normal (see formula below)
LIGHT BLUE = INT&BLM = locked at 128: as expected, all normal
GREEN = Final INJ PW = includes Low PW adder & Voltage compensation: as expected, all normal
DARK BLUE = TPS%: as expected, all normal. Make sure you have your foot to the floor and throttle is in fact WIDE OPEN!
PURPLE = CT in DegC.: as expected around 185 Deg.F. (85Deg.C.)
I forgot to include voltage. But it was stable at 13.2V.
Formula to estimate speed:
Real measured tire radius in inches/168*RPM/Rear gear ratio/Transmission gear ratio
24.8/2/168*5850/3.89/1.00 = 111MPH
Graph shows 112MPH. Remember that the RPM, in our case, is inaccurate bouncing around +/- 100 rpm
So let's put 5900 instead and we get 112mph!!!
Rear gear in our case is 3.89 and we were in 4th gear on the T56 = 1 DO NOT use manufacturer advertised radius. Make your own measurement with level on top of tire to ground.
When dyno tuning or street tuning, you always want to record:
Ambient barometric pressure
Ambient air temperature & MAT
Humidity %
Engine CT
Battery or fuel pump voltage
Type of fuel used
Fuel pressure (ours was fixed at 43.5lbs)
These will make tuning logs more comparable with different elevation and ambient temperature. In some case, it will greatly help tuning MAT compensation. It also helps avoiding chasing one's tale when tuning, specifically VE using BLM values. I'll write something up about it in next post.
THIS IS A LONG ONE BUT VERY INFORMATIVE. It goes over post #22 & #23 as I am limited in character.
INFO#1
TGO member reached out to see if I could help with a '92 GMC k2500 454 TBI, 4l80e. Here's the issue:
"hello, i saw your thread about interpreting datalogs and was wondering if you wouldn’t mind taking a look at a data log from a truck i recently got. The issue im having is a massive off idle bog/hesitation while in closed loop. once warmed up and in open loop it isn’t bad but still slightly present. i’m fairly sure ive diagnosed the cause to be a worn out tbi creating a vacuum leak around the throttle shaft. my theory is that the ecm doesn’t compensate for the unmetered air while running in closed loop which creates the massive hesitation and stumble. once in open loop it can compensate. i would just like to be able to verify this with data but am having a hard time interpreting what im seeing in tunerpro"
I know there is misinformation in his statement and it will be addressed later. TGO member got open/closed loop confused and was thinking in terms of MAF instead of Speed Density system.
ECM: is a 16147060
MASK: $85
REPLY #1
Hi,
Without hearing, smelling, seeing, and touching the engine, here’s my 2 cents worth.
It seems like at the beginning of your run, the engine is not responding to gas pedal input, but by the end it gets better. But why? The MAP readings look normal when gas pedal is not touched, so I don’t suspect a vacuum leak. On a speed density system, a small vacuum leak would increase idle rpm. The TBI does not meter air so there is no “unmetered air”. The leak would just act as a second throttle body so to speak, adding air to the engine, decreasing vacuum, making the ecm add fuel, increasing rpm.
Here's the first event. I left a bunch of data out because it was all normal and it made the explanation too complicated.
Engine appears to idle normally with 20ingh of vacuum (MAP=31kpa – blue). No sign of vacuum leak. At around 32sec mark, you slowly increase throttle (green). The engine barely responds (red). You keep increasing throttle, engine does not respond, vacuum starts decreasing (MAP goes up). ECM adds fuel thinking engine is underload (not on graph). Engine bogs down with suspected too much fuel (no combustion) as you keep increasing throttle. More air, no combustion = massive decrease in vacuum (MAP at 93kpa), ecm adds mor fuel!!!! You let go of throttle, less air, some combustion, engine regains, vacuum goes back to normal, ecm decreases fuel.
Second event is interesting. Again, everything seems normal at idle (mind you, I can’t hear the engine!). You start increasing throttle again, but engine barely responds creating decrease in vacuum and ecm adding fuel. You give it a bit more throttle, but engine won’t rev. You let go of throttle, engine stays where it’s at, then another good throttle jab and engine finally moves up a bit. Again, I can’t hear the engine, but I suspect it was sputtering or running rough.
The only difference between event 1 and 2 is the target AFR (not on graph). First event it was 13.5 and second event it was between 14.5 and 15.5. This is normal as open loop is running its cycle (too long to explain). However, if there is a problem and the engine has a hard time making combustion for unknown reason, removing fuel might help, thus no bog down on the second event.
Third event is key. Again, everything seems normal at idle. You start increasing throttle, but engine does not respond at all for almost 2.5 secs!!! Definitely not normal with 5% throttle. The engine is not making good combustion. Then with throttle steady, engine finally responds a bit and goes down again. You jab the throttle and engine responds. You follow-up with 2 blips on throttle and it seems like engine responds better. Same target AFR as event #2. There is a fourth event that is the same pattern as this one.
It appears that the engine is not responding to the increase in added air with normal throttle operation. What is the cause? It seems like the problem is worst when AFR is richen. So, it seems less fuel helps. TBI systems are speed density. They estimate air volume and engine load using, among other things, the MAP readings (barometric pressure). If the engine does not make combustion with added air, the MAP readings increase leading the ecm to believe there is more load on the engine. The ecm adds more fuels like it’s programmed to do. Adding more fuel to an engine that does not make good combustion is not helping the situation. But the ecm does not know that.
Could it be an issue with the fuel system? Not impossible. However, the BLM/INT did not indicate a massive correction for too much fuel in closed loop. I don’t have much data though. Just something to keep in mind.
Could it still be a vacuum leak? Possible but not likely. I just don’t see it on the graph. But this kind of analysis as its limits, specially remotely.
I would definitely look at the ignition, compression, EGR or plugged exhaust. There’s gotta to be a reason why the engine is not burning fuel. Check your plugs see if they are fouled rich.
That’s just where I would start with little I know about your vehicle.
Hope this helps.
INFO #2 "first of all thank you for all of that information. you nailed it completely on describing what the engine was actually doing.
my theory for the throttle body vacuum leak is because at idle the shaft is relatively tight. once i open the throttle to around the 5% area there is a large amount of play in the shaft (potential vacuum leak). this is also where the trouble area is. my theory was that it wasn’t compensating for this added air entering combustion. as you saw, once the engine warms up, this event drastically improves. while driving the truck, it hesitates and bogs during this event to point where i can only idle it around, once its warm and in open loop, it drives absolutely fine. my theory was that while in closed loop it was running off its program with minimal sensor input and couldn’t compensate for unintentionally added air from the throttle vac leak. once it goes into open loop, it has a better chance at compensating for it. that was just my theory based on what i thought i understood of speed density TBI.
the only thing i haven’t replaced in the fuel system is the fuel injectors, their spray cone looks fine however. ignition has new cap, rotors, wires, plugs gapped at .040”. exhaust has no cat and a new muffler. the exhaust does have an odd scent to it that i can’t quite describe. not rich but also not sweet. the truck was abandoned for 15 years, maybe there’s a mouse nest way up since the exhaust manifolds?!
i’ve bench tested the egr and it held vacuum and moves freely, however, that isn’t the best test i’m sure. i was going to replace it but it turns out this egr is flat out unobtainable at parts stores. i can make a block off plate to see if the issue goes away and then possibly have it tuned out and delete it permanently. i am going to pull the plugs and test compression. also use the boroscope and see if coolant is puddling up in the cylinders.
again, thank you for your time, this has definitely gave me a ton of insight"
REPLY #2 first of all thank you for all of that information. you nailed it completely on describing what the engine was actually doing.
You’re welcome. It’s always a challenge!
Before you read, remember that when MAP reading goes up, vacuum goes down and vice versa. MAP reads barometric pressure in KPA.
my theory for the throttle body vacuum leak is because at idle the shaft is relatively tight. once i open the throttle to around the 5% area there is a large amount of play in the shaft (potential vacuum leak).
That would create more air going in, which would result in a “bigger throttle opening than normal like effect”. However, how much more air? Engine should have more air and just rev. Unless there is a massive amount of air going in at once.
my theory was that it wasn’t compensating for this added air entering combustion.
It is “compensating”. This is a speed density system not a MAF system. MAF systems will not compensate for vacuum leaks as some air is by passing the MAF sensor creating a lower than real air flow reading. On a speed density, more air = more fuel no matter where the air is coming from. Try unplugging some vacuum lines while idling and see what the engine does. I bet you, the rpm will go up.
as you saw, once the engine warms up, this event drastically improves. while driving the truck, it hesitates and bogs during this event to point where i can only idle it around, once its warm and in open loop, it drives absolutely fine.
First off, let me clarify: the cycle is start-up, open loop and closed loop. So, when the engine is warm and ecm is ready, it is in closed loop. That means the ecm uses the O2 sensor to adjust fuel delivery trying to target lambda (AFR of 14.7 for gasoline). In your case, during closed loop operation, when “it drives absolutely fine”, I don’t have any data to see what the ecm is doing as far as fuel correction. It is a possibility that it helps if the ecm is reducing fuel, but I don’t have any data, so at this point it’s pure speculation. You need to datalog when the engine is fine to see what the ecm is doing.
my theory was that while in closed loop open loop it was running off its program with minimal sensor input and couldn’t compensate for unintentionally added air from the throttle vac leak. once it goes into open loop closed loop, it has a better chance at compensating for it. that was just my theory based on what i thought i understood of speed density TBI.
Again, you inverted open loop and closed loop. While in open loop (no O2 sensor feedback) it is true that ecm is running on fixed calibration. That is why target AFR is richer. It’s part of warming-up process. ECM can not compensate if there is an issue that creates rich/lean condition. “Unintentionally added air from throttle vac leak” would be registered by the ecm as the MAP sensor would detect it. Engine would use that extra air and engine speed would go up. In event #3, I see throttle being opened 5% (TPS) with no response from engine (RPM) for 2.5 secs but a decrease in vacuum readings until engine rpm finally goes up increasing vacuum. Here’s the sequence of event for #3:
TPS = 0%, RPM= 775, MAP=36kpa, BPW=1.5msec = ALL NORMAL at idle
TPS increases to 1.6%, RPM increases = 900, MAP decreases to 33kpa, BPW =1.5msec = ALL NORMAL. You slightly opened throttle, more air went in, engine took it and increased RPM, vacuum went up fueling stayed the same.
TPS increases to 5%, RPM stable = 875, MAP increases to 44kpa, BPW = 2.0msec = what’s wrong here? Engine did not respond to more air flow. The vacuum went down and ecm added more fuel.
TPS stable at 5%, RPM decreases = 825, MAP increases to 58kpa, BPW = 2.9msec = what’s even more wrong here? Engine still not responding to more air flow from opened throttle. Vacuum is decreasing more. ECM is adding more fuel.
TPS stable at 5%, RPM has finally increased = 1150, MAP decreases to 30kpa, BPW = 1.3msec = what happened here? Engine reacted to more air after 1.5sec of 5% steady throttle, vacuum increased, ecm decreased fueling.
The issue is: A)does vacuum decreases due to vacuum leak resulting in ecm adding fuel and bogging down engine? B) Or is the engine not reacting to more air (bad combustion) resulting in decrease vacuum making ecm add more fuel aggravating situation? It’s true that in the early stage of open loop, the engine runs richer with no adjustment from ecm, so issue could be aggravated in both A-B scenarios with too much fuel. In closed loop (no data), if ecm is correcting fuel, it could have a positive impact on issue, but it’s speculative at this point.
Now let’s look at the end of your log. We’ll call it the last event. The throttle is opened to 5% again (red), the rpm (blue) immediately goes up with barely any vacuum decrease (green) and fuel added (light blue). ALL NORMAL. Then, throttle is increased even further at 9% with engine immediately reacting going up to 1550rpm with barely any vacuum decrease and fuel added. ALL NORMAL. The fueling conditions are the exact same as event#3 when looking at AFR/BLM/INT (not on graph but I verified them). That means ecm has nothing to do with difference between event#3 & last event. The difference though? The engine is getting warmer.
Some of this trouble shooting is tedious. But one as to go with the most likely possible cause. From what I know, I lean towards something wrong with combustion when engine is cold aggravated by the function of ecm. If engine does not use available air from lack of combustion, vacuum will decrease (MAP goes up), ecm will think engine is underload and add more fuel to an already bad combustion. In a MAF system this would not happen. However, can the bad throttle shaft play a role? I’m not sure. Should it be fixed? Why not. Will it fix the problem? Your guess is as good as mine? One thing is for sure, whatever the cause is, it triggers the ecm to add more fuel… it’s just the nature of speed density system.
INFO #3 "gotcha, it seems like i’m basing my theory off of the maf system i thought i was familiar with 10 years ago and was incorrect on map sensor function. the throttle body leak could be the potential cause of my high idle issue so it’s worth fixing on that front.
i pulled the plugs and it is clear that the combustion isn’t perfect. cylinder 1 looks especially bad. accidentally dropped #8 which is why it’s broken"
"i’m going to make an egr block off and run a compression test on all cylinders. cylinders looked rather clean and no sign of coolant seeping in. it seems that my issue is either compression or ignition at this point."
CONCLUSION: "awaiting the memcal with egr function disabled from @Tuned Performance to arrive in the mail, once it arrives i’ll be able to thoroughly test the truck in all conditions and give you a 100% conclusion.
the testing i did prior definitely seems like the egr was the cause. i would get a horrible stumble right off idle that disappeared once i removed the egr and blocked off the port. i could say the egr was the cause based off that test but id like to throw it in gear cold and try driving to really confirm because that’s where the issue was very present.
while i’ve been waiting for the new memcal, i disassembled my tbi unit, drilled out the throttle shaft bores, and installed bronze bushings. the throttle now opens smoothly with zero play. as we discussed last week, this most likely was not the cause so i don’t think it will skew any results i get from the egr removal. if anything im hoping this cures the high idle issue i’ve been having"
"can confirm egr was 100% the issue. i got the new memcal back and it has never ran smoother off idle. absolutely glides right up now. theoretically it was either the egr or its solenoid seeing as i removed both. thank you for all your help, i really appreciate it"
08-31-2023, 05:51 PM
