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.
The member I was helping earlier with seemingly false knock have re-engaged after a little hiatus...
Coming at it with fresh eyes this time, I happened to notice that Tunerpro is reporting that the ECM is not detecting knock despite pulling out up to 20deg of advance.
For example, when I export the data, the column entitled "Knock Detected" never shows "yes", despite the ECM pulling out that massive amount advance.
How would the ECM not report that it's detecting knock, but at the same time be pulling timing and logging knock counts?
The thing that's got us up in the air is that we're getting it consistently between ~1500 and ~3000 rpm and high map. Right where you'd expect a TPI to experience knock.
We tried octane boost and to my surprise the knocking went away completely on the first run. It worked so well I was suspicious. To be safe, I had him make another run and the knocking returned in full force.
We've also pulled out (I think it was) 15 deg of advance and didn't put a dent in the knock retard.
so we have data points that suggest false knock and data points that suggest real knock...
He's been all over the car looking for loose hardware, exhaust leaks, etc, and other possibilities for artificial noise sources fooling the knock sensor, but to no avail.
Is there any way a faulty distributor could be causing this? Prior to it being replaced by another shop before all this stuff started to happen, I don't 'think' the issue was there before.....but I could be wrong since this has been going on for so long.
yeah we tried that... I temporarily disabled the knock retard entirely for diagnostic purposes and he reported that it ran really strong and without evidence of knocking.
however that's not a condition I'd like to leave him with.
That may be the case, was a knock detected flag added to the SuperAUJP data stream? And you are using a regular $8D BIN?
A knock detected bit was added to S_AUJP v5, plus the ability to determine the type of knock that would set the bit ("slow" or "fast"). The default S_AUJP v5 BIN sets the Knock Detected bit if "slow" knocks are detected and will report the slow knock count. $8d requires 255 "fast" knocks to be detected before one "slow" knock is reported.
Might try changing the XDF items below to report "fast" knocks. These are the exact counts detected and reported by the sensor. Why change? Because the $8d code retards timing when "fast" knocks are detected, NOT "slow knocks"; and if you are flagging/reporting "slow" knocks, you can see timing retarded without a knock flagged or counted. Likely not 20 degrees, but making the changes below may shed more light on the situation:
ALDL Word #43
00c1 = "slow" knock count (default)
00c2 = "fast"
OPT 2
b6 Not set = flag a "slow" knock (default)
b6 Set = flag "fast" knock
I'll run the revised xdf on Monday. The car is still at the garage getting a new distrubutor put in.
So to help me understand.....running the old version xdf was pulling timing due to fast knock counts, but not recording the slow counts? And one slow count equals 255 fast counts?
So if the ECM was pulling timing due to fast counts, running the revised xdf will only pull timing when a slow count is recorded?
I'll run the revised xdf on Monday. The car is still at the garage getting a new distrubutor put in.
So to help me understand.....running the old version xdf was pulling timing due to fast knock counts, but not recording the slow counts? And one slow count equals 255 fast counts?
So if the ECM was pulling timing due to fast counts, running the revised xdf will only pull timing when a slow count is recorded?
To clarify, timing is always pulled in $8d based on "fast" knocks regardless of what is reported. By reporting fast knocks rather than slow, you are able to see the exact connection between knocks detected and timing pulled. But as previously stated, if 20 degrees are truly being pulled, it's hard to believe it's due to knock counts. Removing 20 degrees while accelerating under a load would be like putting on the brakes. Have never seen that amount of timing pulled. Something else is the likely contributor.
If you log with fast knocks and post a .csv file, I'll be glad to take a look at it.
I was helping a fellow LT5 owner tune his ZR last year. Car was running well, but the datalogs showed no knock and we increased timing to “take advantage” of that. However, the timing was going beyond the point of where I felt comfortable especially since the tables were more agressive than the SA tables I as using for my ZR. I suggested he swap KS which he did. Lo and behold, KNOCK. As RBob, would say, if u suspect knock, start looking at the plugs for evidence despite what the datalog shows.
I was experiencing knock at mid rpm/ mid kPa levels (1800-2400rpm/45-70kPa). Turns out I had at least one bad injector and several marginal ones (thank u General Disorder), and oil seepage into a couple of cylinders. Funny what u find when u start pulling on a thread and when following the logging data to the physical evidence.
My next step was to have him use some local high-octane racing fuel that's available nearby. After a couple of runs with it, I think that'll definitively show whether we have a real or false knocking issue.
If the ECM is still maxing out the knock retard with 100+ octane fuel, then it has to be false knock. But if it goes away consistently after multiple runs, then we know it's real.
So he put in 110 octane fuel and is STILL getting massive amounts of knock retard kicking in... Here's a picture of a couple of his plugs. It seems typical for the pictures he sent me. Look at that history plot of the knock retard... how the hell do you end up with much retard on 110 octane fuel with ~9:1 compression?
His BLM's look fantastic and didn't have to do anything unusual to get there... to me the plugs look nice, but I'm not an expert on reading plugs. Light tan color, seems like fueling is just right, no?
Plot 1 is the maximum values for the run... plot 2 is the average...
Mike......the plots that you are showing.....are those from the straight up 110 octane run, or is it the 3rd run where 110 and 93 octane are mixed?
For Everyone reading.........I was running no less than a high 90's octane fuel and the knock retard is still insane. This is a stock motor with the original cam. It was rebuilt due to slight oil consumption, but the cylinders couldn't have been honed out all that much resulting in crazy increased compression.
Running a .bin that eliminates the knock retard and the engine runs awesome......no complaints. I do hear some light pinging/knocking, but nothing crazy. I don't hear the same pinging/knock when running .bins that still allow for knock retard......or at least I don't think I do.
When running a .bin that eliminates the knock retard, wouldn't I be seeing a rather large drop in knock count when compared to running .bins that still use knock retard? Isn't the whole idea of pulling timing out meant to reduce the knock counts?
It doesn't seem to be running lean based on BLM's really close to 128. It's got the proper plugs in it. It doesn't have a vacuum leak. I'm not seeing any codes being thrown. Timing is set at about 10 BTDC, but even when it's been set to 4 BTDC, the knock retard was still present.
If an injector was bad, the engine wouldn't be running great when the knock retard bin was being run, so it's most likely not an injector.
If the EGR were bad, wouldn't I be throwing a code?
There is a slight rattle coming from the right side exhaust where the cat used to be, but that isn't something that happens all the time. Possibly when the high MAF/low RPM zone is hit, maybe it's modal frequency is setting the exhaust rattle off?
This 'search' has been going on now for well over a year, and I still haven't found the source of the supposed knock.
The history table below is from 10-3-2018 when I had Jeff run octane boost as a quick test to see if the knock was real... Somehow, we got zero knock for the entire run. Never have been able to duplicate this since that time...
A knock detected bit was added to S_AUJP v5, plus the ability to determine the type of knock that would set the bit ("slow" or "fast"). The default S_AUJP v5 BIN sets the Knock Detected bit if "slow" knocks are detected and will report the slow knock count. $8d requires 255 "fast" knocks to be detected before one "slow" knock is reported.
Might try changing the XDF items below to report "fast" knocks. These are the exact counts detected and reported by the sensor. Why change? Because the $8d code retards timing when "fast" knocks are detected, NOT "slow knocks"; and if you are flagging/reporting "slow" knocks, you can see timing retarded without a knock flagged or counted. Likely not 20 degrees, but making the changes below may shed more light on the situation:
ALDL Word #43
00c1 = "slow" knock count (default)
00c2 = "fast"
OPT 2
b6 Not set = flag a "slow" knock (default)
b6 Set = flag "fast" knock
HTH
So I went back and made this mod to the xdf file and Jeff made a run with it. Interesting... for all the knock retard being administered, there is no evidence of fast knocks being detected. I've attached the .csv file. The column for "knock detected" doesn't have any entries for "Had Knock"....
My plugs are looking like that these days. My BLMs are great. Still get a bit of knock on a small increments of acceleration or decel that matter. Did a datalog yesterday, but need to have a look at it. Car feels great and very responsive. My MPG gauge shows very good mileage near 30mpg at 75-80mph cruise. Sometimes u just scratch ur head.
Originally Posted by ULTM8Z
So he put in 110 octane fuel and is STILL getting massive amounts of knock retard kicking in... Here's a picture of a couple of his plugs. It seems typical for the pictures he sent me. Look at that history plot of the knock retard... how the hell do you end up with much retard on 110 octane fuel with ~9:1 compression?
His BLM's look fantastic and didn't have to do anything unusual to get there... to me the plugs look nice, but I'm not an expert on reading plugs. Light tan color, seems like fueling is just right, no?
Plot 1 is the maximum values for the run... plot 2 is the average...
Could this be piston slap ?
Jeff had mentioned to me the engine was honed out to address oil burning.
was it honed to much to cause a loose fit of the pistons ?
Mike, I think we had this issue or something similar with my car a couple years ago. I could never figure out where the ECM was picking up knock from. I think we left off on the possibility that the head gaskets I used may be thinner than the OEM ones and it was driving up the static compression ratio a bit.
I used Chevy Performance 10105117 head gaskets on an iron headed stock 350 engine. Not sure if this helps you guys or not. Im still scratching my head as well. I would like to get my car dialed in a bit more and get rid of the knock that is being detected.
Brad, yep I remember that. But on yours, I recall the knock retard was within reason for what I suspected was the culprit (i.e., the thinner head gaskets).
On Jeff's car, I've NEVER EVER encountered knocking like this... 9:1 engine (I'll even give you 9.5:1 with possibility of thinner gaskets) that the engine rebuilder swears up and down was rebuilt to factory specs....
20 deg of retard running 110 octane racing gas with stock timing?
I did get your e-mail, and I can take another look at your data. I may have you run your car with the modified XDF file as well to see if the fast knocks are being detected. That's one thing new that I learned on this go-around that I didn't know before.
I'm also having Jeff back probe the ECM connectors to ensure we don't have any wiring issues. Matter of fact, I just solved another member's issues by having him backprobe the ECM connectors which revealed a bad ground (was causing intermittent BLM spikes up to 160 and back again)...
I'm anxious to hear Elky's analysis of this situation now that we find Jeff's ECM pulling timing out like crazy but ostensibly without any fast knocks being detected.
BTW, glad your car is running good... hopefully we can figure out your knocking as well...
Could this be piston slap ?
Jeff had mentioned to me the engine was honed out to address oil burning.
was it honed to much to cause a loose fit of the pistons ?
I don't know... this whole thing to smells like some sort intermittent electrical issue.
I still hearken back to that one time last year where I had Jeff put octane boost in the car and for some odd reason the knocking COMPLETELY went away. I had him then re-run the car without any other changes to verify it wasn't a fluke.... and lo and behold the knocking came back. A significant mechanical issue (like what you're describing) wouldn't behave like that IMO...
03-23-2019, 06:44 PM
