ECMlink Knock Under Low Boost SD

[Joshky9]

I just got speed density set up and dialed in for idle and cruise. I've been doing some wide open throttle pulls and keep getting knock. I'm only running 15 PSI of boost on 93 octane pump gas. I am using the EVO 8 (mod1) timing and fuel maps that are given on the ECM tuning wiki (v3configs [ECMTuning - wiki]). It seem like I am getting knock regardless of of A/f ratio. I've tried leaning it out to see if what I was seeing was rich knock and had little success there as well. I don't think I need to cut timing because it seems pretty conservative for such low boost. Any help would be greatly appreciated.

The pull stats at 217 secs.

 

[Calan]

http://www.dsmtuners.com/forums/ecm...efore-you-ask-ecmlink-dsmlink-log-advice.html

 

[Joshky9]

Apparently I did not know we had to go number by number for each of the PLEASE READ before asking for ECMlink/DSMlink log advice. We have checked all of these prior to posting the log so I do apologize for not posting this information along with the log. Current mods on my profile are accurate.

1) No boost or exhaust leaks which was checked right before posting this log.
2)Mechanical timing is dead on
3)Base timing is 5 degrees verified via timing light.
4)Plugs and wires are in good working order gap to .028 BPR7ES
5)Compression test is 180 in all 4 cylinders.
6)Basic throttle body adjustments are in check.
7) Base fuel pressure 43.5, FIC 650's
8)Compression Ratio on 2g pistons 1g big rods 8.5:1
9)We are currently in the process of replacing the knock sensor.
10) Wideband is properly calibrated.
11) 93 octane
12)Have reviewed all the ECMlink how-to videos
13)No DTC/CEL codes
14)Battery voltage is logged and appears to be good, fuel pump rewire.

Hope that takes care of that. Thanks again for any advice we may get.

Sincerely

Josh

 

[Calan]

Thanks for posting that info Josh.

Do you have any more logs that show this knock?

That knock is right in an area that is a real pain in the ass. It could be real, or it could be phantom knock. The problem is, you need a few logs to see if the knock is consistent, or if it has a bit of randomness to it. Although you are at low boost and RPM, the knock is happening in an area where you are transitioning to more load, so it's more critical to figure out if it's real knock or phantom.

Your AFR is around 12.3 when the knock hits, and you are pretty far into WOT at that point. BUT... right after that, the car is really rich, so it's a bit hard to say what effect the AFR is having on the knock (if it's real). That somewhat lean to really rich transition in just a few hundred RPM isn't helping matters, so one of the first things you need to do is get a more steady AFR between the point where the boost curve starts to level off, and red line. Although your boost never really hits a plateau in that log, you can see the curve flatten a bit around 3500rpm if you look closely. That is the point that I would want to hit my target WOT AFR. So in this log, you would want a steady AFR (11.3 - 11.5 or so maybe for starters?) from 3500rpms on up.

Once that's done...

One test you can do that sometimes points towards PK is to let completely off the gas as soon as you see the knock CEL, and then go immediately back to full WOT. If the knock goes away, it's probably PK; if it comes back as soon as you are back on the gas, get back off it immediately as it's probably real knock. This isn't fool proof, but it's a start. Set the knock cell to come on at 2-3 degrees, and set the stock boost gauge to show knock so you can get a better idea of how much you are seeing. 2-3 degrees for a second isn't a major deal; but if the gauge climbs and you start getting above 4-5*, you may be getting into a place you don't want to be. Just be sure to be quick on the gas pedal.

I absolutely hate seeing knock in that RPM/load area. It's easy to assume it's PK and just set the knock threshold to ignore anything below 3500rpm or so... but it's also in a critical area of load transition, that also happens to be one of the hardest areas to get the right AFR and timing dialed in.

 

[Joshky9]

Those areas in the log have also been puzzling us (my brother and myself). I also see that while not accurate my front O2 sensor seems a bit lean for WOT. I am wondering if we may not have calibrated the wideband correctly according to the Gofer method I reviewed last night. I think we just used the AEM WB R1 and it matched up relatively well so we assumed that it was accurate. I know what assuming things does with these cars. We also swapped out the knock sensor with a new one. Added another ground strap to the engine, and ensured proper torque to the knock sensor., and still have ended up with knock on 10lbs-13lbs of boost. Our future plans will be to get a smooth transition of AFR once we get into the boost, ensure our Wideband is calibrated correctly, and check the phantom knock method you describe. One questions I may have for you or gofer is why in his video for calibrating the wideband does he only calibrate it for as rich as 11:1 AFR, is that because he runs E85? Should I calibrate to as rich as 10:1? Thanks again for the reply we greatly appreciate it and have really invested a lot of time and money into this car and would love to get it running well.

 

[Calan]

One questions I may have for you or gofer is why in his video for calibrating the wideband does he only calibrate it for as rich as 11:1 AFR, is that because he runs E85? Should I calibrate to as rich as 10:1? Thanks again for the reply we greatly appreciate it and have really invested a lot of time and money into this car and would love to get it running well.

I'm not sure I understand what you are asking. Calibrating the wideband has nothing to do with what AFR you are targeting at any given time. When you calibrate the wideband, you are just ensuring that it reads the true AFR, over it's entire range.

As far as what AFR to shoot for on your tune, it really just depends on your setup. Some cars like a richer AFR, some can run more lean. I would start at around 11.4:1 or so on 93, and then go from there.

Regardless of what AFR you are trying to hit (based on the OpenLoopMaxOct table), you want the wideband and table value to match up, which is where getting the VE table and fuel parameters correct comes into play.

 

[Joshky9]

I'm referring to this thread http://www.dsmtuners.com/forums/articles-tuning-ecu/428702-ecmlink-linear-wb-calibration.html in Gofers video he describes the method he calibrates his AEM WB. I was not referring to anything regarding the AFR that I am targeting. I was referring to why he only calibrated it to 11:1 vs maybe 10:1. Just by looking at the video it appears that values richer than this will not be accurate but I may not understand the whole process very well but this was my interpretation. My thought is that the widedband may not be logging in link at similar values to the gauge. We will calibrate it and return with results. Thanks for the assistance Calan I really appreciate it.

 

[Calan]

Ahhh... ok.

Anything below 10.0 or so is meaningless anyway; you should never be running a gasoline AFR below 10.5:1 or so, even on crap gas.

He probably picked 11:1 because that would be about in the middle of the range you would want to be in on most pump gas.

 

[Joshky9]

Just to update. We have calibrated the Wideband O2 via the Gofer method. We have worked toward making a smoother more consistent AFR in the 11-11.4 range. We are still getting rather consistent knock. But our AFR is not matching up very well with AFR est.

We have even tried your method to determine if it is real knock or phantom knock and it comes on as soon as we get back on the gas.

We are still unable to get a full pull in 3rd gear to redline because we keep getting knock counts in the 4-6 range. I do notice that free reving in neutral we can occasionally get a few counts of knock if we rev rather aggressively. Could we have a combination of phantom knock and real knock?

I have reviewed multiple phantom knock posts and have tried to address many of the issues pointed out by others and we don't seem to have any of them going on. I would post a log but I feel that we have really not made much progress in our knock issue. We will continue working on it tomorrow. Thanks again for all of your insight and help.

 

[Joshky9]

So after we have done a lot of adjusting and trouble shooting for some phantom knock I think we have have some real knock because these FIC 650's just have never seemed to flow very well. Currently looking at these issues.

1. VE table in Speed density is trending toward too high of values. If you look at the adjustment table provided by your link tools. Does this mean that the Global fuel setting needs to be adjusted. It looks if I reach a load level and boost level that I want to run I will be way over the 100-105 range. ( Calan I have read in some of your prior posts that since this(global) is a relatively known value we should not adjust this in most cases)
2. To get airflow per rev to be in the .25 range we have had to alter the VE table significantly in the top left corner circled in red. What causes this low airflow reading? Is it related to global fuel setting being off or some type of issue with the MAP sensor or even a mechanical issue of sorts? It only has 1.5" of vac line and has its own private vacuum line directly tapped into the intake manifold.
3. I know the table needs smoothed out but I did not go too far into doing that if I was going to make some significant adjustments to global fuel setting.

Thanks again for the insight.

 

[Calan]

1. VE table in Speed density is trending toward too high of values. If you look at the adjustment table provided by your link tools. Does this mean that the Global fuel setting needs to be adjusted. It looks if I reach a load level and boost level that I want to run I will be way over the 100-105 range. ( Calan I have read in some of your prior posts that since this(global) is a relatively known value we should not adjust this in most cases)

The global fuel value is only "known" if all the factors that go into it are known, such as true injector flow rates, fuel properties, base pressure, etc. You always hope that the global fuel value is correct, but if it needs to be changed, so be it.

2. To get airflow per rev to be in the .25 range we have had to alter the VE table significantly in the top left corner circled in red. What causes this low airflow reading? Is it related to global fuel setting being off or some type of issue with the MAP sensor or even a mechanical issue of sorts? It only has 1.5" of vac line and has its own private vacuum line directly tapped into the intake manifold.

If you have to adjust low airflow values one way, and the upper airflow values the other, it usually involves inaccurate deadtime values. Since yours are both needing to be adjusted up, I would guess that global needs to be dropped.

 

[Joshky9]

I would guess that global needs to be dropped.

Do you mean that global needs to be increased IE less negative value?

 

[Calan]

Do you mean that global needs to be increased IE less negative value?

Yeah...closer to zero. My bad on the wording there.