Stuck n00b... Lean at WOT ~12.63 OTS map

[dale_gribble]

I have an 06 MS6, AP, AEM CAI, Cobb TIH + CPE TMIC. Otherwise stock, 33k miles. Here's a pic of my old 6 and the MS6 I recently acquired.



I installed the mods myself. The car runs outstanding. Even when I see KR registering, I don't hear anything and the car runs amazing with amazing power. Unfortunately, I don't have any pre-mod logs for comparison. Convenient, right?

Anyways, I started noticing a knock count in the KR real time data view and started logging as a result. I am running the OTS MSCAI v2 + TIH 91 tune. I tried the OTS MSCAI v2 + TIH + TMIC 91 tune and the KR was actually worse; same condition- AFR ~12.5 or so.

KRs are ~4.5 at WOT and it appears as though my AFR is lean for WOT @ ~12.63; My fuel pressure looks ok from what I have read.

I understand the MAF CAL, and I have the data to do it, but I don't think it is going to help my lean condition at the moment which appears to be my most vexing problem.

I checked and double checked my connections; it doesn't appear as though I have any leaks and the AFR looks ok otherwise. I am getting some very minor KR under partial throttle sometimes, but this sounds normal (<1 degree periodically).

Any help in the right direction... Maybe I just need a custom tune and these OTS just won't work for me.

Many thanks in advance.

[dale_gribble]

Based on what I read, I thought the ECU is supposed to dump more fuel when KR is detected which is why I am surprised my AFR isn't reflecting that.

[Black Metal]

Do a maf cal and learn the proper way to log. Get car to 2800ish rpm in desired gear, press log, wait for ap to say logging, go wot until 6500+ rpm, press to stop log, profit! Your fuel times are way off I'm surprised you don't have a CEL from it. The AEM and MSCAI likely have different diameter maf housings which is why you are seeing such high trims.

[dale_gribble]

Understand... I just am not sure if I should stay WOT to 6500 if I am knocking already down lower?

Also, will a MAF cal do anything to address a lean condition? Isn't that more of a normalizing calibration?

[Black Metal]

Originally Posted by dale_gribble Understand... I just am not sure if I should stay WOT to 6500 if I am knocking already down lower? Also, will a MAF cal do anything to address a lean condition? Isn't that more of a normalizing calibration?

Do the maf cal first. Your car thinks it is getting a different amount of air than it really is which is what is causing your lean problem. By calibrating your maf it will know how much fuel it needs to hit your requested afr.

[dale_gribble]

Amazing. I couldn't quite extract that from everything I was reading. Thanks.

[86AmishMs3]

Using logic you can deduce the issue. You are using an OTS tune for the mscai. I don't see the maf cal being off that much. You have a different issues. You're not hitting fuel targets...and your ltft's when in closed loop are fucking horribly high. You have a unmetered vacuum leak. Since you're still running lean at WOT that would suggest that it's not a pressurized leak becuase if it was you'd go rich.

So you have a leak somewhere vacuum side of the intake that is not metered...ie after the maf meter and before the turbo. Check the turbo inlet pipe.

Originally Posted by Black Metal Do the maf cal first. Your car thinks it is getting a different amount of air than it really is which is what is causing your lean problem. By calibrating your maf it will know how much fuel it needs to hit your requested afr.

This could be WRONG OP, and I say this because I don't want the OP doing something wrong before he looks at the basics...no disrespect @Black Metal. The AEM and MSCAI are the same internal diameter IIRC. See my explanation above. A maf cal will only disguise the issue.

[Black Metal]

Originally Posted by 86azms3 Using logic you can deduce the issue. You are using an OTS tune for the mscai. I don't see the maf cal being off that much. You have a different issues. You're not hitting fuel targets...and your ltft's when in closed loop are fucking horribly high. You have a unmetered vacuum leak. Since you're still running lean at WOT that would suggest that it's not a pressurized leak becuase if it was you'd go rich. So you have a leak somewhere vacuum side of the intake that is not metered...ie after the maf meter and before the turbo. Check the turbo inlet pipe. This could be WRONG OP, and I say this because I don't want the OP doing something wrong before he looks at the basics...no disrespect @Black Metal. The AEM and MSCAI are the same internal diameter IIRC. See my explanation above. A maf cal will only disguise the issue.

Hahaha thanks for the edit. No worries man I didn't honestly know if they are the same diameter or not. I agree that a leak is possible so check that first. I also wouldn't be running an ots map for different brand parts that aren't on my car lol.

[dale_gribble]

Originally Posted by 86azms3 Using logic you can deduce the issue. You are using an OTS tune for the mscai. I don't see the maf cal being off that much. You have a different issues. You're not hitting fuel targets...and your ltft's when in closed loop are fucking horribly high. You have a unmetered vacuum leak. Since you're still running lean at WOT that would suggest that it's not a pressurized leak becuase if it was you'd go rich.

Interesting. I will inspect this like crazy then.

Originally Posted by Black Metal Hahaha thanks for the edit. No worries man I didn't honestly know if they are the same diameter or not. I agree that a leak is possible so check that first. I also wouldn't be running an ots map for different brand parts that aren't on my car lol.

In my defense, COBB notes in the OTS map notes to use the MSCAI v2 map for the AEM CAI (AEM made the intake for Mazda).

Thanks guys.

[Black Metal]

Well that makes more sense with the map and I'd be willing to bet 86 is correct about the leak.

[TiGraySpeed6]

just stating cleanly-

1) correct any mechanical issues, i.e.: air leaks
2) MAF Cal to within +/-3 preferably, Cobb says within 8 is ok, but closer to 0 is better
3) WOT AFR fine tuning so actual matches target.
4) Now that air & fuel is true you can begin to work on boost

Steps 2 & 3 cannot be done accurately if there are mechanical issues. Remember, the AccessPORT is a force multiplier, for better or potentially worse if you're starting from misleading data due to air leaks or other mechanical issues that cause false readings.

[dale_gribble]

I just spent an hour carefully double checking every single connection to no avail... I reseated everything carefully from the turbo intake up including the intercooler connections. I don't hear or feel any leaks anywhere. The hoses are all seated and in good shape.

I reseated everything, reset the computer and drove around and the high LTFTs returned as before.

Maybe I should swap the stock intake pipe back in. Have the Cobb TIHs been known to leak at all?

[TiGraySpeed6]

How were you checking for leaks? You didn't specify, so feel compelled to ask as trying to spot leaks with engine running can/will often provide misleading or inconsistent results.

There are instructions posted for making a boost leak tester with very inexpensive items sourced from lowes/home depot/ace hardware or the like which work remarkably well.


edit to add: the cobb inlet is as good as most, rarely the cause of an issue unless caused by external sources

[dale_gribble]

I did the old 1) feel around all of the junctions, 2) listen for whistles, sucking, etc. and 3) spray a little carb cleaner around all of the main hoses/connections. I will say this. If it's there, it is subtle. I wouldn't expect it to leak enough to influence the AFR like that.

I wish I was more careful about modding the car. I should have baseline logged, installed one thing, baselined again and so on. My only option on that front is to demod and go piece by piece.

FWIW, the car runs perfectly, purrs like a kitten, doesn't hesitate or jump and has tons of power. Honestly, if I didn't have a data logger on there I would never even know I was having issues.

[btstarcher]

How's the clamp where the TIP connects to the turbo? Mine kept slipping off as I tightened the clamp; had to clean it off really good and scuff it up a bit so it would stay.

[dale_gribble]

It really feels like a rock solid connection. I am probably going to pull the COBB TIH and CPE TMIC and replace with the stockers to kind of start rolling back to get to a known good point.

[Black Metal]

have you cleaned your maf sensor?

[btstarcher]

Well there's not many places it could be....gotta be between the MAF and the turbo I believe. You plugged the MAF back in right? And the hose from the dump valve?

[dale_gribble]

I just wanted to take a moment to thank you guys for the help.

The MAF hasn't been cleaned, no. Is that worth doing?

Yes, all hoses and the MAF are reconnected. Through all of this the computer hasn't logged a single code so as far as ECU limits are concerned, this problem is somehow staying inbounds of the sensor/table ranges.

[Black Metal]

IMO its worth a shot a dirty maf won't read correctly.

[TiGraySpeed6]

You're using the term data logger- I'm gathering not an accessport?

You'll find the DISI in the Speed6 is much less flexible regarding bolting on mods, even normally simple things like an intake, without having the means to make adjustments or corrections to your tune. Which is where the accessport (AP) will come in.

Also, you'll find that even simple things like checking for air/vacuum leaks will be different, hence my recommendation to search for posts on making your own boost leak checker.

This might help as well, if you haven't gotten your hands on a copy yet-

Mazdaspeed 6 Technical Training

[dale_gribble]

I am going to do some reading here in my hole and crawl back out when I find anything out. Thanks a ton guys.

[86AmishMs3]

I would still make a boost leak tester..it comes in handy. Usually you can't visually see the leaks nor find them when the engine is running.

[dale_gribble]

Leak was found. You guys are pimps. It was the TIH. I repaired and went for a quick drive. 0 KR at WOT and AFR ~11.89. I didn't drive around enough for my LTFTs to readjust. I will report back tomorrow after I get some run and new logs.

I took everything apart again and saw that the upper coupler on the TIH squished out from under the clamp. The clamp was tight and the connection looked secure but it no longer had a hold on the metal coupler in the hose.

Actually paying for membership now. I came in as a n00b, you guys were totally awesome and helped a brother out. I am a car guy and very experienced at that, but MSF can be intimidating just based on the level of discussion here. MSF FTW.

[btstarcher]

Glad you got it sorted out!

[dale_gribble]

Ok, lot more information here. Not even sure where to start. Even though I found a loose connection at my TIH, it was not the source of my problems. I thought that it fixed the issue because I saw better AFR and lower KR. As it turns out, the better AFR and lower KR was because of a map I loaded where I added a global modifier to the MAF table of 1.03x at voltage >3v.

When I went back to the OTS MSCAI + TIH 91 Octane fuel map, I noticed that the AFR was lean again, the KR was back and the fuel trims were way off. This indicated that even though I had a loose connection, I still had something else going on with the tune or another variable.

So, I started comparing the OTS MSCAI2+TIH 91 v200 map against the OTS Stage 1 91 v200 map and the Base map, specifically the MAF Tables.

Low and behold, something started looking peculiar. The OTS MSCAI2+TIH 91 v200 map MAF tables g/s values are lower then the base tune for any given voltage. Here's an example:

Base tune g/s values (from 3.52-5V in .3-.4v increments):

115.65 119.92 124.26 128.70 133.25 137.90 142.65 147.52 152.49 157.58 162.78 168.10 173.51 179.04 184.70 190.49 196.41 202.47 208.67 215.01 221.49 228.13 234.92 241.87 248.98 256.26 263.77 271.43 279.25 287.23 295.38 303.68 312.15 320.78 329.58 338.56 347.70 357.02 366.52

OTS MSCAI2+TIH 91 v200 tune g/s values (from 3.52-5V in .3-.4v increments):

107.86 115.30 119.38 123.57 127.87 132.34 136.96 141.70 146.56 151.54 156.65 161.87 167.21 172.68 178.26 183.94 189.73 195.65 201.71 207.88 214.18 220.62 227.20 233.93 240.80 247.81 254.96 262.26 269.69 277.29 285.04 292.93 300.98 309.19 317.55 326.08 334.77 343.62 352.64

This completely contradicts what I have read in all of Cobb's documentation where a Calibration Intake Multiplier needs to be implemented to compensate for additional airflow. So, the OTS MSCAI2+TIH 91 v200 tune actually indicates that there is LESS airflow for a given MAF voltage, not more.

I checked other OTS tunes for different CAIs, for example their tune for the Stage 1 + cpCAI + TIH. These appeared to have a greater multiplier across the table as I expected.

126.27 130.93 135.67 140.52 145.48 150.55 155.75 161.06 166.49 172.04 177.72 183.53 189.44 195.48 201.66 207.98 214.44 221.06 227.82 234.74 241.82 249.07 256.48 264.07 271.83 279.78 287.98 296.35 304.89 313.60 322.49 331.56 340.80 350.23 359.84 369.64 379.62 389.79 400.16

So, since the OTS MSCAI2 + TIH v200 tune actually shifted the MAF table lower then stock, that explains why my fuel trims were so friggen high.

I took the base Stage 1 91 v200 tune and, using Cobb's worksheet came up with the Global Intake Multiplier for my intake and applied it to the MAF tables in this tune. I loaded this tune in my car and voila. Fuel trims ~ 1%, stupid power, 11.8:1 AFR @ WOT, and 0 KR.

Having said this, someone at Cobb should take a look at their OTS MSCAI maps.

[dale_gribble]

After checking, the INJEN maps were similar to the CP CAI maps in that the MAF tables were higher then stock.

[TiGraySpeed6]

@David@COBB

[Black Metal]

I was accidentally right lol. Glad you got it figured out.

[dale_gribble]

Couple logs today. This is an OTS 91 + TIH + IC with MAF tables that I calculated and populated. What a difference a day makes. Still not sure if I am right, but I am WAY BETTER.

[dead-zone]

It looks better than the first on

[dale_gribble]

Besides MAF cal, do you guys think I should screw with anything else?

[TiGraySpeed6]

Originally Posted by TiGraySpeed6 1) correct any mechanical issues, i.e.: air leaks 2) MAF Cal to within +/-3 preferably, Cobb says within 8 is ok, but closer to 0 is better 3) WOT AFR fine tuning so actual matches target. 4) Now that air & fuel is true you can begin to work on boost

Read the ATR Help file, at a minimum the tuning guide and the two addendum's that follow it
Read Abilors tuning guide
Read Buckers cheat sheet

Apply what you learn to your map, step by step, logging as you go to verify results.

With the above reference material, some study time & some common sense any fool can make an acceptable, decently performing and safe daily driver map.

You'll note that I said "map", not "tune".

If you need a tune, contact DIZZYTUNING (Phate or Bucker), FREEKTUNES (atvfreek) or the PERM TEAM (Djuosnteisn, Dano, Lex or Socks), in no particular order

[dale_gribble]

Thanks for the advice.

I have made good strides in the car's performance. My MAF cal is good and I am hitting my WOT AFR targets dead on (OTS tune = 11.4:1). I have been slowly upping the AFR target (now that I trust it) to 11:70:1. My car is hitting its HPFP targets no problem so fueling isn't an issue in my current configuration.

I am still getting phantom KR (between 0.35-1 at WOT) that appears randomly. It isn't repeatable and it doesn't sustain KR throughout a WOT run. Maybe 1-2 blips in a whole wot run.

I have learned a ton but the learning curve is steep. I am having a bit of trouble understanding the important relationship of wastegate and boost and how they relate to boost targets. I have pretty much left the boost/wastegate alone and have been playing with timing advance in very, very small increments.

When I compare my original logs to my new logs, I can definitely see what has changed and why so I am on the right path.

[dale_gribble]

UPDATE: 2 steps forward 1 step back. I thought I was getting a hang of this until I was stuck in stop and go and my TMIC heat soaked, bringing up my WOT BAT enough to start registering KR. My car was running perfect until the TMIC heat soaked and brought the BAT up ~20F.

For some reason, I can't tune in 3D; I think very 2-dimensionally. I now fully realize that the relationships between all of these metrics need to be tuned together.

I have commissioned Stratified Automotive Controls to hook me up with a tune before I blow my car up. I really want to study what they do and how they do it and try to figure things out working with a pro instead of ZZB.