Lean during WOT

Hello all,@Redline
a few weeks back I did a compression test with the help of @VTMongoose to help diagnose my engine and unfortunately found low compression on 2 cylinders. Results were as follows:
1: 180
2: 155
3:160
4:185

a leak down test has yet to be performed to further diagnose the issue but gut feeling and community wisdom is pointing towards worn piston rings. This has prompted the need to run a more conservative tune to extend the life of the engine before an engine rebuild is feasibility since the car has been on the Cobb OTS stage 2 map. And custom tuning is good practice in general so I opted to go with @Jason@DizzyTuning for my tuning needs. my logs can be found attached. Due to my lean condition during the WOT log, there is reasonable evidence that there is an intake leak allowing un-metered air in. I will be doing a full intake tract inspection for cracks/tears and replacing hose clamps hopefully before the weekend to determine if this is true root cause.

Any help in this matter is appreciated as this is my daily and won't be able to afford engine rebuild until early next year.

 

Hope it isn't cracked ringlands causing your low compression. If so, you should be able to hear sort of a hissing sound coming out of the oil filler hole when you do the test. Listen for where the air is escaping. Oil filler hole (rings/ringlands issues letting much more air into the crankcase)? Through the intake (intake valve issues)? Through the tailpipes (exhaust valve issues)? Also, pay attention to your leakdown %ages/cylinder.

Regarding the leanness: did you run previous WOT logs that didn't show the same leanness at WOT, or is this the first set of logs?

If it is a mechanical issue, I can only think of two possible causes: 1) unmetered air entering the system, or 2) an exhaust leak, particularly before the wideband 02, like a crack on the DP to turbo flange. For 1, make sure you tighten down all of your intake couplers really well, you already said you cleaned your MAF, which is good, and you may also need a new MAF o-ring. For 2, visually inspect your DP, especially for cracks pre-O2 sensor.

If it's a tuning issue, @Jason@DizzyTuning will get you sorted quickly, I'm sure, by adjusting the WOT portion of your MAF cal. Here's some good info about this stuff from a completely random forum o.0
http://www.mazdaspeedforums.org/forum/f426/understanding-fuel-trims-using-them-detect-leaks-138722/

 

Thanks for the thorough response man, I really appreciate it. I will have to redo the compression test to determine where the air is escaping and respond back then. Unfortunately I don't have much free time until the weekend of 10th and 11th so it may be awhile until you get a response from me.

this is my first set of logs

intake and exhaust inspection to be performed this week

thank you for the link, very helpful in understanding whats going on. thinking I will head over to home depot tomorrow and make the boost leak tester NATOR style.

 

Just point of clarification: you'll have to do a LEAKDOWN test to determine percentage of leakdown, and where the air is escaping. This will tell you a lot more than just a compression test. Good luck! I wonder if @VTMongoose has a leakdown tester/source of compressed air? Hopefully someone local does, at least regarding a compressor. You could probably get a decent leakdown tester cheap from Harbor Freight anyways.

 

I believe @VTMongoose does and we were supposed to do it last weekend but life got in the way... went to a concert instead. Weekend of the 10th and 11th is happening. I've had an air compressor on my do buy list for awhile now so going to rape harbor freight soon.

 

Unfortunately, I'll be at midwest meet that weekend. You are welcome to borrow my tester though.

As for your problem with going lean at WOT, I must tag @Jason@DizzyTuning here to alert him that you have a Cobb SF intake (perhaps this was not discussed) and he needs to use that MAF cal with your tune instead of the stock intake MAF cal, which he's currently using. That is the main problem right now. Once the correct MAF cal has been applied, it will be easier to chase down whatever leaks he's seeing in your logs (I didn't look at them in much detail honestly).

 

My understanding is that they have the same inside diameter and are therefore compatible, albeit not optimized, running the same MAF cal. Getting it dialed in with certainly improve things. Having done a lot of self-tuning, seeing WOT AFRs in the mid to high 12s would not be consistent with running a stock intake vs. a COBB short-ram on the same MAF cal. That's a major AFR difference on two intakes that are virtually identical in ID. It'd be helpful to know what the O/L AFR target is. But if it's ~11.6 or so and you're seeing 12.5+, I definitely think something more than a MAF cal issue is at hand.

I hope that's not the case, but the degree of practical experience I have in this type of thing leads me to believe it very well may be. *fingers crossed*

What do you think, @Jason@DizzyTuning : would stock airbox vs. COBB short ram being the only variable be responsible for such a large WOT AFR shift? Or do you suspect something mechanical, like a intake vacuum leak or pre-wideband exhaust leak?

One guiding principle is always best applied in scenarios like this IMHO: it is ALWAYS better to rule out a mechanical issue than to try to change the tune not knowing what exactly is occurring. If you're not sure the cause, always eliminate mechanical ones first. I'll definitely defer to a pro on this, though...

 

The Cobb SF and other SRI's do indeed have a slightly larger diameter than the stock MAF housing.

Here's the stock MAF cal at the voltages he's going lean at, as provided by Cobb. I have personally used this MAF cal on stock intakes and it works perfectly:



Here's the Cobb SF MAF cal at the voltages he's going lean at, as provided by Cobb. I have personally used this MAF cal on Cobb SF intakes and it also works as intended:



Here's a snip of brainbot's log showing g/s values that correspond roughly to the stock intake MAF cal at their respective MAF voltages (notice that they are a little higher because the ECU is adding fuel using WOT trimming):



Additionally, I calculated the difference between the logged AFR and what I'm assuming Jason is targeting (11.5:1). It's about a 10% difference. So the ECU is probably using whatever compensation it can to help keep the AFR's in check, but there's still an ~18% difference between the MAF curves at each MAF voltage, and the ECU is fueling accordingly.

Perhaps instead of prefacing your statements with the fact that you "have done a lot of self-tuning", you should preface them with some reading and perhaps browsing around in ATR. I am definitely guilty of being lazy myself from time to time, but in this case the data is easily accessible and pretty clear cut.

 

Once again, @Redline is spot on. The stock sized maf housing calibration compared to the Cobb maf housing may cause a .1-.2 afr deviation, but not as large as we are seeing.

We just started troubleshooting this issue and the first thing I have informed Brian to check is the complete intake tract for any leaks, as this is the most common source of a lean issue. Once that is all checked and known to be leak free, we will move on to the next area of concern.

This is all too common, and I am fully confident we will get this issue sorted out.

 

Sorry for ruffling @VTMongoose 's feathers. I didn't think the difference between the Cobb SRI and stock was as significant as it is, but ATR shows ~325 grams/sec at 5 volts for stock airbox and 383 grams/second at 5 volts for the COBB. That is ~15% difference, and since @brainbot1 is driving a Genwon which doesn't have WOT fuel trims, could that be the issue, Jason?

And let's all bear in mind it's not about having a perfect memory, but trying to help someone. I certainly have no problem saying I'm wrong/misremembered

 

Really? There's an 18% difference between the two calibrations I referenced above at the voltages of interest. You're suggesting that if I take a known good MAF cal on a Gen1 and just shift it 18% downward, I'm only going to see a 1-2% lambda observed difference? Man that's awesome to hear! That means we can all just run E85 mixes without even bothering with those silly injector scalars and such.

 

See my post above. Pretty sure it's a Genwon vs. Genpu thing. What's not a big issue WOT for a Pu can be a big issue for a Won.

 

That is true. However OP has a Gen1.

Sent from my A0001 using Tapatalk

 

Exactly. I believe Jason's statement was presuming a Pu. Gotta cut the guy some slack; he tunes a lot of folks. Remembering everything about every car off the top of your head is pretty much impossible, you know? I actually had a map update sent to me from FREEKTUNE that was locked to an APv3 with a completely different S/N. I have an APv2b o.0

It happens...

 

I have no problem with Jason using a slightly wrong MAF cal and never got up his ass about it... Mistakes happen and no damage has been done here. Now if he's going to say it wasn't a mistake, then I have a problem.

Sent from my A0001 using Tapatalk

 

I appreciate the responses that have gone on in this thread. Regardless of the MAF cal, it wouldn't cause the lean condition I am experiencing and that still needs to get sorted. Once it is sorted, Jason will provide me with a new map.

 

I'll have to take a look at this in the morning, cause I may be confusing two Brian's. Lol.

I'm also confused at the map that is posted in the OP, because I don't name my base map's "base map".

Regardless, we will get this issue resolved.

 

I saved it as a different file name, didn't want to blast my info out there

Sent from my SM-G930T using Tapatalk

 

flashed the new map last night and did a few datalogs, AFR's are looking much better. Big thanks to @VTMongoose for identifying the issue and @Jason@DizzyTuning for quickly sending me the new map

 

Good to hear a simple MAF cal got you fixed up. Come to think of it, you could see in your LTFTs how the amount figures were off was going up more and more by breakpoint range. MAF curves also happen to be exponential by nature, so that makes perfect sense in hindsight.