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.
continue to work through the tune of my l31 hot cam TA and finally have pulled a data log. Clearly the EBL VE learns have been doing their thing but have noticed a bit of an odd phenomena with the WB appearing to oscillate by about 1 point for each row of the datalog capture. Enclosed is a small example data log extract showing what I mean. My wb gauge more consistently shows the lower (richer) reading.
given how regularly it oscillates, I'm inclined to believe that it isn't a real reading.
Newly purchased innovate LC2 and gauge. Settings haven't been touched, and have simply added the device to the ADC channels in the EBL system
I'm not so sure it's odd behavior... if you consider the following. The sampling rate on the data seems pretty fast... so within one second worth of data, you can get numerous cylinder firings.
Your data looks to be exclusively around 1000 rpm... At 1000 rpm, you get 500 cylinder firings per minute on a given cylinder on a four stroke engine. So for the side of the engine facing the WB, that's 4 cylinders firing. So you get 4 x 500 = 2000 firings per minute. ~33 cylinder firings per second.
It looks like you're getting about 17 line items per second of data? So purely on a time scale, each line item is approximately two cylinders firing?
Then you have the Hot Cam, which probably has a decent amount of overlap. And your data is all being taken at fairly low rpm where overlap can hurt the volumetric efficiency of the engine, and cause sporadic rich/lean conditions.
So I might expect some significant variation in the WB readings.
You also appear to be at 25C (~73F) on your coolant temp and in open loop, so a cold engine with a moderate overlap cam... yeah the WB readings actually don't seem that bad to me.
Ultm8Z, apols, the datalog I posted was a snippet of a 10mb file of a fairly varied drive and the WB observation continues pretty much throughout. A slightly larger extract enclosed. Car is set to run open-loop and the WB sensor is at the y-merge (so should be seeing the exhaust of all 8 cylinders).
dabomb6608 - thanks for posting that. Exactly, your wb doesn't seem to dart about a point like what I'm seeing with mine
not ruling out an actual firing issue, but given the wb gauge isn't jumping around, I'm starting to suspect it may be a grounding issue; the WB and ECU do not share the same direct ground. I messaged RBob about the best way to do that without risking frying the EBL unit, but haven't heard back.
absolutely happy to continue to hear thoughts; just scratching the surface with datalogs and currently overwhelmed with the data!
Looking at this larger data set, I think my observations still stand for the most part. In fact the WB is seeing the entire engine, then you have even more firings being sampled per line item in the data. And again, the engine seems to be at low rpms for all of this data (mostly all below 1500 rpm), where significant cam overlap can result in inefficient operation.
It'd certainly be good to do what you can to avoid electrical ground loops as much as possible. But the WB gauge jumping around or not could also be a function of the display update rate on the gauge vs the sampling rate of the EBL.
His o2 is mounted exactly where mine is. Without comparing specs, I would imagine my overlap is decently close to his as well. I am running a LC2 wideband as well.
His o2 is mounted exactly where mine is. Without comparing specs, I would imagine my overlap is decently close to his as well. I am running a LC2 wideband as well.
How does your wide band behave? Mine fluctuates a bit too... though only about 1/2 point either way... my cam is pretty mild though.
I think 100 ms gets you 3 cylinder firings at 1000 rpm if I did my math correctly?
If there's some sort of low rpm fuel distribution issue from side to side (like the so-called "split BLM" phenomeon), I could see some significant variation.
What happens at higher RPM's when the cam moves into more of it's efficient range of operation.
I think 100 ms gets you 3 cylinder firings at 1000 rpm if I did my math correctly?
If there's some sort of low rpm fuel distribution issue from side to side (like the so-called "split BLM" phenomeon), I could see some significant variation.
What happens at higher RPM's when the cam moves into more of it's efficient range of operation.
A consistent almost perfect 1 point move though?
Here's a similar chart from one of my datalogs at idle.
There is certainly up and down variances. But they are inconsistent and vary in how much split is between them.
Now that I am really looking at this and seeing these charts, that has got to be either an electrical issue or an issue with the setup of the LC2. It is way too consistent to be anything else.
Almost acting like it is pulling from a simulated narrow band output.
Last edited by dabomb6608; Sep 18, 2024 at 10:58 AM.
Interesting. Im pretty sure I’ve got the wb analogue output hooked up; see readings down to the 12s when I got into PE mode
but will double check over the weekend and attempt to get another log
On the LC2 there are two outputs. Both programmable within the Logworks program. Analog Output 1 is programmed to output between 0 V for an AFR of 7.35 (gasoline) and 5.0V for an AFR of 22.39. Analog Output 2 is 1.1 V for an AFR of 14 and .1 V for an AFR of 15. Not sure how you would have got readings down to 12 using only the 2nd output. Maybe if the wires are combined it goes haywire...who knows.
I think for my setup with the gauge I use (cheap prosport Evo gauge from old WB kit) I programmed Output 1 to be 0V=10.0AFR and 5V=20.0AFR. Probably left output 2 alone as I don't utilize it. I made sure to match settings in EBL.
Here's a similar chart from one of my datalogs at idle.
There is certainly up and down variances. But they are inconsistent and vary in how much split is between them.
Now that I am really looking at this and seeing these charts, that has got to be either an electrical issue or an issue with the setup of the LC2. It is way too consistent to be anything else.
Almost acting like it is pulling from a simulated narrow band output.
Not sure if you answered my last question...
"What happens at higher RPM's when the cam moves into more of it's efficient range of operation."
The general trending seems that there appears to be a nominal AFR that's being held, but like it's intermittently going richer by 1/2 point across the board, even at the higher RPMs...
The general trending seems that there appears to be a nominal AFR that's being held, but like it's intermittently going richer by 1/2 point across the board, even at the higher RPMs...
I'm not the OP. My results show a much different and inconsistent swing in the AFRs compared to GTA1990s.
just an iterative update that I'm still researching on how best to commonise the ground between the LC2 and EBL ECU.
In the meantime, I decoupled the wb gauge from the LC2 wideband controller WB feed wire (yellow) to the EBL ECU.
Data logged enclosed. Interestingly the swings are of less magnitude but still present; leading to me believe it is wiring.
I have a few busy weekends but will try and to keep this thread updated once I've managed to figure the ground out (if the ECU is grounded to the block, I'm thinking an additional strap between the block and battery, perhaps after getting the multimeter out))
just an update that I believe this is indeed a grounding issue.
>I replaced the ground wire to the LC2 on recalling it was a hard copper wire and also added an additional earth from the passenger side head (that the ECU grounds to) to the battery.
>as above decoupled the gauge and ECU wideband signal source. Sidebar, I reprogrammed the brown wire on the LC2 wideband to also provide a wideband signal so the gauge is getting its signal from the brown wire, and the ECU from the yellow wire. This was surprisingly easy to do given the horrors I'd heard about serial to USB adapters not being recognised - a cheapo amazon jobbie worked just fine
key in ignition off, I'm not seeing any 1/ 0.5 point spikes like I did. Unfortunately, and I believe unrelated (!), I've having a no start issue.
Will data log to confirm once I'm back up and running - but initial indications are positive
