How's my LV?

[Spec B]

Also one more thing to note - your original LV showed that your highend maf trim was leaning your mixture by 3% on the highend. Although unlikely, it could be that the 3% leaning may be enough to cause knock..

 

Since your latest LV show 0 you are running 3% richer today than you were on your original LV.

[iNVAR]

I've been trying to go wot whenever possible, as you can see from the few bursts of 100% throttle in the logs. but it's hard during daylight commute hours in nyc. will try to do more on my days off for sure. Thanks for looking so far!

[Spec B]

cool...to get the highend fuel trims settled in you have to do 1/4 throttle (light cruise) in the higher-end of the RPM range. Also nice uphill 70-80 MPH 5th gear cruising. WOT runs do not allow fuel trim learning to take place..it obly happens during cruising.

[iNVAR]

cool...to get the highend fuel trims settled in you have to do 1/4 throttle (light cruise) in the higher-end of the RPM range. Also nice uphill 70-80 MPH 5th gear cruising. WOT runs do not allow fuel trim learning to take place..it obly happens during cruising.

edit: hmm, weird the message I typed in didn't submit.

[iNVAR]

cool...to get the highend fuel trims settled in you have to do 1/4 throttle (light cruise) in the higher-end of the RPM range. Also nice uphill 70-80 MPH 5th gear cruising. WOT runs do not allow fuel trim learning to take place..it obly happens during cruising.

Weird, my previous reply didn't show up.

 

Oh yeeeah, the closed/open loop thing, right?

 

I do drive on the highway and cruise some too. in fact, about 10 miles of that 13 mile commute is highway driving with light traffic on it, so I do cruise a lot. Strange that it's not adjusting the fuel trims. On the way home, I'll make a note to look at the airmass reading to see what numbers I'm getting and try to hit the correct range.

 

edit: Looking at my log, I do have some driving in that airflow range at around 25-40% throttle. Maybe just not enough mileage logged in it yet?

[subawang]

If you want to get some good variation, occasionally drop it to 4th and even 3rd gear on the highway and just randomly speed up and slow down.

[iNVAR]

Sounds good, will try that too. Thanks for the tip. Hope no one stays behind me, might piss them off!

[iNVAR]

And the saga continues...

 

GOOD NEWS! or bad news, rather. But at least I've made some progress. Found my knock, and here's what I got. When I left work, I pulled an LV and it was clean. Didn't bother making a screenshot. Temperature was approximately 85 outside.

 

I drove some highway mileage, the usual driving, and the log still looked clean, almost identical to the other LVs, with a minor change in the AF trims. 0 timing adjustments across the board. I was able to get out to my test strip.

 

I got 3 hard pulls but I did NOT reset the ECU prior to doing this since the LV table was completely clean. I checked this immediately before I started the pulls so I am sure it was all 0s.

 

1) The big log shows a knock event at line 8950 at engine load 1.17, 2237 RPM, 32.55% throttle. Should I make an adjustment here? The very first LV I posted in this thread shows the LV with -0.35 exactly in that range.

2) The first two WOT logs show a knock event at 6375 RPM, load 2.14 and 6391 2.15 load. Now, here's the strange thing.

 

The first log shows my Advance dropping from 8 to 7.5 briefly at 2.24, 6068 rpm but no knock, and as far as I know, the LV is CLEAN at this point. From looking at older logs, the advance should have remained at 8! What gives? Will the temperature compensation table cause the advance to reduce like that?

 

Second log shows advance being reduced from 8 to 6.5 at 2.34, 5755RPM. Even sooner than before but no knock event until the later range. Again, the LV is clean in the 2.34/5755 range as far as I know. What gives?

 

Third log shows huge advance reduction from 8 to 5.5 at 2.34, 5768RPM. Don't know why because there should be no learned knock at that range yet. Or maybe it's close enough since the range is 5800+ I believe? The knock table has been learned at this point, so knock sum doesn't increase.

 

4) Drove home some, posted the log just in case, but probably not very useful. But I pulled the LV and sure enough, there's learned knock in the same load/RPM I saw in the very first LV.

 

Thoughts on what to do? Also, should I put up my map so you guys can look at it? I'm thinking this is heatsoak related. Pretty hot out. The IAT was about 100, so you can only imagine what the actual temperature of the intake charge is... wish there was a sensor post-IC. It's an AVO TMIC. Oh yeah, totally forgot to mention my mods. AVO TMIC, AVO panel filter, Invidia catted DP, Invidia catless UP, SPT CBE, Magnaflow midpipe straight muffler.

 

Thanks.

 

Log I pulled for infamous who gave it a thumbs up, back in october 2010.

www.crimetank.com/misc/reference_log_clean.csv

 

1.3MB log driving home, shows 1 knock event at line 8950.

www.crimetank.com/misc/romraiderlog_20110818_152829_long_drive_home.csv

 

WOT pull 1 - funky reduction in advance (8 to 7.5), and 1 knock event

www.crimetank.com/misc/romraiderlog_20110818_155123_3rd_wot_1.csv

 

WOT pull 2 - funky reduction in advance (8 to 6.5), and 1 knock event

www.crimetank.com/misc/romraiderlog_20110818_155328_3rd_wot_2.csv

 

WOT pull3 - funky reduction in advance (8 to 5.5) and no knock event (knock table is learned)

www.crimetank.com/misc/romraiderlog_20110818_155452_3rd_wot_3.csv

 

Log driving home, nothing to really see here unless you're curious. Mixed city/highway about 3-4 miles.

www.crimetank.com/misc/romraiderlog_20110818_155541_drive_home.csv

 

LV attached.

[Spec B]

The reduced Advance is a result of your fine learning KC - what you see in the LV.

 

There are a few ways to deal with the knock - I think the issue is a combo of heat and load. You are hitting less load than you did in October (because of the heat) - less load at the same RPM usually equals more timing in that region of the map. The easiest way to deal with it is to lower the timing in the region of the map that you are knocking. Just pull 2 degrees in the area of the base timing map and smooth out the transition of the adjoining cells in that region of the map.

 

By smooth I mean do not have a big transition in timing between the cell you reduced and the adjoining cell.. So if you are lowering the timing from 18 to 16 (-2 degrees) adjust the cells that touch it accordingly..maybe reduce those by -1.5 degrees and exand the from there - reducing the cells that ajoin to the adjoining cells by -1 degree..hope that makes sense.

[iNVAR]

The reduced Advance is a result of your fine learning KC - what you see in the LV.

Yeah... I know it reduces the advance when there's FLKC, but the table was definitely empty when I made the first log, and I'm pretty sure it was still empty in the 2nd log as that knock event only occurred once (while making the 1st log.) And just in case I wasn't careful when I checked the LV before the WOT and maybe I slipped up, I went back to check the big log, which doesn't show any sort of knock in that RPM/load range, so it definitely wasn't learned.

 

Perhaps I will have to log FLKC and FBKC too...

 

There are a few ways to deal with the knock - I think the issue is a combo of heat and load. You are hitting less load than you did in October (because of the heat) - less load at the same RPM usually equals more timing in that region of the map. The easiest way to deal with it is to lower the timing in the region of the map that you are knocking. Just pull 2 degrees in the area of the base timing map and smooth out the transition of the adjoining cells in that region of the map.

 

By smooth I mean do not have a big transition in timing between the cell you reduced and the adjoining cell.. So if you are lowering the timing from 18 to 16 (-2 degrees) adjust the cells that touch it accordingly..maybe reduce those by -1.5 degrees and exand the from there - reducing the cells that ajoin to the adjoining cells by -1 degree..hope that makes sense.

Yes, it does make sense. Going to dig around the map and see what I can do. Thank you so much for the help so far!

[iNVAR]

Here's the original base timing. I think I have the right table, right?

 

I circled the two cells that I think I'm supposed to change based on the knock?

 

The one in the lower range wasn't on my LV yet, but I think it would've showed up eventually with enough driving. It was on the old LV and also on the log.

 

The one in the upper range already appears to have quite a bit of timing removed from it. It's down at 15.86 while surrounding cells are significantly higher... And now that I look at the LVs from revisions I went through last year with Infamous, that was the same cell... maybe he didn't actually remove enough timing from the region and it was just the cooler weather that stopped the knock.

 

What should I do? Looks sort of messy here with that 15.86 in the middle of nowhere.

[iNVAR]

Revised map. Does this look okay? The low load ranges I made changes to because I saw in the original LV knocks in those ranges, and when I took a closer look at the long driving maps, I saw a few knock events in those ranges too. Reduced it by a tiny bit.

 

How's everything look?

 

edit: Another thing I noticed from looking at the logs. I don't appear to be hitting my boost targets.... At 4200, I should be at 18 lbs but I'm only at 16.8. At 5200 I should be at 17 but I'm only at 15.6. HMM... I mean I realize it's hot so boost will be reduced, but by that much?

[Spec B]

I would not pull timing in the low RPM (<3000K) and I would make 4800RPM @1.05 load 38ish degrees and 5200rpm @1.05 load 39ish. The highend needs some work. You want the columns top to bottom to increase steadily after 3000 RPM (as the RPMs rise) and you want the timing in the rows to steadily decrease left to right (as the load goes up). That means you have to widen the area of the reduction....take your time.

[Spec B]

As long as you have a fairly smooth transition between the cells - you will be ok. It does not have to be perfect..

[iNVAR]

Hrm, okay, back to the drawing board. Wonder why Infamous stuck that 15.86 in there then in his final map for me? :confused::confused:

 

And holy crap, if I make it steadily increase from top to bottom, that's a lot of pulled timing in the cells above it. Isn't this going to really take out some of the oomph from my car? Noticed in my log that at that knock event, I'm missing almost 2PSI of boost.

 

Also, any reason for not pulling timing at low RPM?

 

Thanks, you've been an immense help so far!

[Spec B]

I typically write off any knock below 2500K as noise...

 

I didn't notice that dip in the original map..I suspect it was a mistake.

 

You should start by pulling enough to make the knocking stop and then slowly add in timing wherever you can. You will lose a little bit of power - not alot. At the end of the day keeping the columns to smoothly increase going down is important.

[iNVAR]

I don't think it was a mistake... Here are a few LVs from last year before Infamous did the final revision that I'm currently on. I think he aggressively removed timing from that spot... but I don't know why he didn't smooth things out.

 

Attached a smoother map. Look okay to you?

[Spec B]

Looks good to me.

 

I can't see how changing that one cell in the map makes sense - I still think it's a mistake. he would have at least changed the ones following it to something lower. Infamous is a much more experienced tuner than me - maybe it is not an error.

[iNVAR]

I'll flash that map on tomorrow, but it seems to me like Infamous did it on purpose. I checked several older revisions of the maps including the very, very first base map he ever sent me and they all have that dip. He definitely would have noticed it after about 10 different revisions over 2 tunes. Hmmm... Now I'm wondering if I should be messing with that area so much...

[sjwelna]

The bottom line is that if you are seeing real knock events, you need to do something about it. Easiest way is to combat it is to just pull a couple degrees of timing. Will you lose some power? Possibly, but the motor will be much healthier for it. In the end, if the computer is learning to pull 2* of timing out at a spot in the map after experiencing detonation and you pulling that timing out up front so it never detonates still gets you to the same amount of timing. Which would you prefer? I know I'd rather just never let my motor see those detonation events to begin with. I'd rather have 280whp with the motor never seeing knock versus 320whp with occasional detonation.

 

As far as blending timing maps, I always seem to feel like I'm icing a cake blending and smoothing out the high spots. When you smooth out the transitions in the tables it makes it much easier for the ECU to interpolate between cells when things are changing rapidly.

 

The new timing map looks good to me, I'd flash it, get IAM back up to 1.000 and do another 3rd gear pull and see how it responds. This is half the fun, making small changes and then logging how the car reacts to it.

 

-Steve

[iNVAR]

Thank you Steve. You're right, I'd much prefer knock free than more power. It's my daily driver, not a track car. I was just worried that all that timing removed in the surrounding area would rob the car unnecessarily of power, but meh, better safe than sorry I guess.

 

I went out today, temperature about 80-85F, sunny, and did about 20 WOT pulls... took damn near 2 hours but I after every 3 pulls, I'd check the logs and make adjustments. As you can see from some of the logs, the IAT was extremely high, and as I understand it, that's pre-IC. I can't even imagine what the post-IC heatsoaked temperatures are actually like!

 

What do you guys think of the new base timing?

 

On a related note, I looked at the knock correction advance table as well and am curious. By the end of my pulls, the intake temperature was so high that engine load at that RPM range was low enough to drop to the column to the left, where Infamous had setup my DA to only 3.87 and 3.16 in some cells. I've attached a shot of the DA table as well.

 

So, I have a question here. If DA is what's added on top of your base timing to = total timing, why not just have the DA table always at 8, and have base timing reduced accordingly so that total timing is still the same? Wouldn't this provide the engine with an extra layer of safety and allow the ECU to pull timing to a larger degree if necessary?

 

Files attached:

2 LVs (first taken right after doing pulls with the new map we worked out above... didn't include more LVs because they all looked the same during the tuning. Knock in that same cell. Second taken after everything's been revised)

5 logs (first taken after the map we worked above, and another taken after every revision of the map where I pull some timing. I did at least 3-4 pulls for each revision.)

3 images (the final map, the final map compared to the original Infamous map I had, and the DA table)

 

Thanks all!

romraiderlog_20110819_142550.csv

romraiderlog_20110819_143540.csv

romraiderlog_20110819_144848.csv

romraiderlog_20110819_150247.csv

romraiderlog_20110819_151800.csv

[iNVAR]

Ugh, this is pretty disgusting. Imported the CSV into Airboy's spreadsheet, and this is what the dyno plot looks like. Yikes.

 

plot1: from Oct 2010 on original infamous tune.

plot2: from yesterday's WOT logs, still the infamous tune.

plot3: today, after I put on my final revision.

[Spec B]

 

What do you guys think of the new base timing?

 

So, I have a question here. If DA is what's added on top of your base timing to = total timing, why not just have the DA table always at 8, and have base timing reduced accordingly so that total timing is still the same? Wouldn't this provide the engine with an extra layer of safety and allow the ECU to pull timing to a larger degree if necessary?

 

 

 

Timing looks really good to me - smooth. Logs are nice and clean now...as the weather gets cooler you may need to adjust the timing in the higher load cells (the ones to the right), if knock pops up. Peak Maf g/sec looked like it dropped only by about 5, so power is only very slightly down. Well worth it for having a safe map.

 

You can chnage everything to 8 and adjust accordingly on the base map side. That is how I have my map setup. But i would not do it in the cells below 2000 RPM - because if IAM ever drops to 0.5, your timing will be negative near idle. Not exactly sure what would happen in that case.

[Spec B]

Ugh, this is pretty disgusting. Imported the CSV into Airboy's spreadsheet, and this is what the dyno plot looks like. Yikes.

 

plot1: from Oct 2010 on original infamous tune.

plot2: from yesterday's WOT logs, still the infamous tune.

plot3: today, after I put on my final revision.

 

It's mainly the heat difference. My winter logs show 30-40 more WHP...same tune. I have noticed that the difference between a 70 degree day and a 90 degree day is also pretty huge.

[Spec B]

Wait..you made MORE power with the new map?