|
|
|
|
|
Welcome to Saab92x.com!
|
|
|
|
So I'm here now, still doing lots of studying, but for the moment just want to figure out why my IAM is 11 on a stock car (well, up pipe only). 
I'm going to scale my MAF [stock airbox] just as a starting point. I've been comparing my ROM to "A4TF800F - MY05 MT OSECU Stage 1 r8" but I'm going to hold off on making any changes until I do my timing belt next weekend. So exciting!
Offline
Offline
Welcome. 
If your IAM is 11 right now, that'd definitely be the first thing to iron out before you start changing things. Here's how to start with that:
1. Reset your ECU by disconnecting the negative battery terminal and pressing the brake pedal for ~15 seconds, or by downloading merchgod's learning view tool and using it to reset the ecu.
2. Start data logging, specifically trying to find what's causing your IAM to drop. Your IAM should start at 8 after you reset the ECU, so you have two options. One, you can just start datalogging from that point and see if you can identify under what conditions the car is knocking. This might take a while to happen though, because the timing will already be reduced, and depending on how you drive, it might be a while before the IAM climbs back up. The second option is to just go ahead and do vishnu's fast learning trick. This will bring your IAM up to 16. Then you can do some pulls and see when the car is knocking.
What brand of gas are you using? I was having a knock issue with my car back when it was stock, and it turned out to be because BP sells shitty gas. I switched brands and it went away.
Offline
Thanks kornfeld; saw you over on RomRaider earlier. Same kornfeld, I presume?
Here is the LV screenshot I took yesterday after resetting my ECU and going for a drive. I was on PetroCanada 91 ("top tier" brand, E10)
Earlier today I filled 3/4 tank of PetroCanada 94 and reset the ECU again when I got home. Now I'm just waiting for it to get late enough for there to be enough space for redline in 3rd.
I'd like to be able to run 91, as 94 isn't always available, and puts me at the mercy of one brand. We'll see later tonight if I settle at 16, but my future tune may be accomodating to 91.
How has your tune worked out? Are you [still?] using the OSECU tune I mentioned above?
Offline
Yep, same kornfeld here and on romraider.
Your maf scalings look good, I don't know if I'd bother changing that. You want all four of those numbers to be +/- 5% (and a bit outside that range isn't really a cause for alarm), and you're well within that range. If you want the practice though, it can't hurt to go through the motions.
The timing values in the individual cells are kind of meaningless right now though, because the IAM is low. You may actually have knock in more areas than just what is reflected in that learning view, but that knock is not showing up because the drop in IAM is hiding that issue. Based on what's there, though, I'm going to guess that your issue is at high load and high RPM, which is the most important range to make sure there are no issues.
Looking at that learning view table reminds me about the stupid ranges it has in its stock form (and in many maps you'll find online). Keep the following information in mind once you get to the point where you want to start loading maps:
You'll notice that the column headers are load ranges. A stock car won't ever go much above ~ 2.1 load, which makes the final column completely pointless, and the second to last column mostly pointless. These values should be adjusted so that the final column is 2.1+ or something like that if your car is otherwise stock. You'll get much better resolution this way.
You'll also notice that the row headers are RPM ranges. Your car will most likely never exceed 12,800 RPM, so the final RPM range is pointless. The second to last range is mostly pointless. These should be modified so that the final header is 6,000+ or something like that.
You can access and modify these ranges in RomRaider under the "Ignition Timing-Knock Control" category.
I completely abandoned the osecu map I downloaded. It turned out to not work well for me. It was fine at low rpm and load, and fine at high rpm and load, but ended up being crappy in the transition range. I basically went through the whole process of starting from scratch using the scooby newbie tuning guide on NASIOC.
Offline
Feeling better now:
I'll give 91 octane a chance again, maybe that 1/8 tank of old gas sitting in the bottom was enough to make the other 7/8 I filled crappy. Maybe all 8/8 are crappy. For now, 94 octane it is!
Thanks for the reminder about the columns and ranges. I've been doing a lot of reading of a lot of "How to"s and threads, and not all of it has stuck. Scoobie Noobie is a particularly big one. Did you end up following the guide fully? I don't mind going through it myself, but if there are any updated methods, please inform me.
I ended up going out for a big long drive before seeing your post, so I collected MAF data already:
So yeah, not much off here, but since I drove around all that time staying out of boost, I may as well apply it, right?
No adjustment greater than 1.4%, but it can't hurt. I have another long drive coming up so we'll see if the blue line is closer to 0 for that!
[All this shit is so cool!]
Offline
I agree...it still amazes me how cool all of this is. So much to learn.
And there won't be any harm in applying those updated maf values. But be aware that you will never get closer to the blue line than you already are. If you drive the car enough, it will always wander a bit.
As for your stock map, I say you finish out the tank of 94 and consciously try and run fully through each load range from low to high rpm. It can be easy to just datalog at low rpm and low load and then only datalog at WOT. It's much more valuable to methodically run through each load range individually.
Then refill with 91 octane and do the same thing. If your knock issue is no longer present, then good. If it does come back, try a different brand to see if it gets better or worse.
Offline
I need to do this baaaad for my SPT Intake, just havn't found a good time. What was your ambient air temp when you logged? Its been so damn hot and humid in Iowa iv been afraid to log at those temps figuring it would through stuff way off.
Offline
It was a turbo-whistlin' 65 deg. F last night. 
Offline
Well kf, I didn't run out my tank as methodically as you suggested, mainly because I forgot, but I have pretty much determined that the gas was to blame. I've been doing a bit of driving on gas from PetroCanada (Suncor) 91 and Shell 91, and both are pretty dismal. After a few tanks of each, my IAM never got above 15, spending way too much time creeping up from 9 (When allowed to adjust normally), and in some FLKC cells pulling over 4 deg. of advance! Also, if I drive on an empty tank, the IAM droppes back to 11. So in went the 94 and just like that no more issues. Well, some knock here and there, but never more than 2 deg. pulled.
So, out of interest, I loaded in the Stage 1 (rev. 8) ROM from osecuroms. 
It feels really good, but it's not entirely a good match. IAM settles at 13, the boost targets are incredibly ambitious, and the advanced timing makes me nervous should I ever get sub-par gas again and not have my laptop with me. Not to mention the initial IAM of 16; easier on the brakes than doing the quick learn procedure, but one safety measure gone.
As a result I've been tweaking up my stock rom, first just the formatting suggested earlier, then the CL/OL delay, and most recently, I've transferred over the boost target, max WGDC, and turbo dynamics maps from the osecurom rom, and turned them down to about 80%. The result feels almost as good as the Stage 1, with the boost coming on earlier and smoother. I still have to crunch through all my logs from today, but a quick look at LV after a 1.5 hour drive shows only two cells with one instance of knock each, which is less knock than what I was seeing on the same gas in similar heat with the stock ROM! I'm guessing the excessive CL/OL delay might have been the cause of some FLKC events.
It would be interesting to see if one cylinder is more prone to knocking. I hypothesise that if cylinder 3 (the one nearest the down pipe?) runs hottest, it would be more likely to knock than the others. The sensor/ECU doesn't care who knocks, it just pulls timing advance for everybody. So would a bit of fueling bias for the hottest cylinder reduce overall knocking almost as much as running all four richer? Is it possible, without specialized equipment or a bunch of sensors, to determine if one or two cylinders knock more than others? Perhaps by looking at the plugs or carbon deposits?
Offline
|
|
Site problems, bugs, or questions? Contact Administrator
All content copyright Saab92x.com 2005-2010
SAAB and the SAAB logos are trademarks of Saab Cars USA, Inc.
Your Privacy Choices