I didn’t feel a meaningful difference at WOT. If anything, near redline it felt like it pulled less hard. It wasn't burying the needle to 8k anymore, felt like it wanted to be shifted a bit early. However, part-throttle response was improved. I also saw the knock adaptation quickly start pulling timing. I came across a thread showing that the CSL setting can allow it to only pull 2 degrees of timing, but I have no idea which table that is. I’m not trying to take the risk of the DME wanting to pull timing but being unable to. If I could find that out I would be much more comfortable trying the canned file, for longer. The canned file had other CSL settings, for knock related things so I am just being precautious.
Originally posted by S54B32
View Post
Since I didn’t feel any meaningful difference at WOT, and I knew the knock sensors were desensitized, and the engine was pulling timing fairly quickly, even with relatively cool ambient temps. To me, that felt too aggressive.
This tune was used in a race car for many years, so it should be good. However, this isn’t a competition car—I drive it to tracks as far as 800 miles away.
I understand these might be tricks to help the engine hit a timing target, but unless I’ve verified that it’s not knocking on a dyno, I’d prefer to play it safe until I can get it on a dyno.
Here were the knock adaptation values after doing the same test loop and similar WOT/heavy part-throttle runs using the stock tune. So there’s clearly some room for improvement, and I’m using this as a tool to sort've gauge it.
I still don’t fully understand what this adaptation means, and based on other threads, it seems like others aren’t entirely sure either. But I figure it’s better than nothing, so I used it for data since the testing conditions were pretty consistent.
One thing that surprised me was how much better part-throttle felt, even though the canned map didn’t have many changes to the VANOS. It basically only advanced the intake cam about 15 degrees at high RPM. The exhaust cam was barely touched—only about a degree at high load and RPM.
However, it had a fairly aggressive part-throttle ignition map, and the torque management’s optimal ignition map was also edited.
So I chose to focus a lot of my time on the part-throttle ignition map. I increased timing, but more mildly than the canned tune, and kept the high-load/high-RPM areas close to the WOT timing values. I also added a degree of commanded timing at WOT, since knock adaptation showed the DME was trying to add timing. The DME should have no issue pulling timing if needed due to fuel quality, high ambient temps, etc. I did not change the torque management, optimal ignition map or the ignition minimum map. Just the part throttle map.
As for the VANOS, I retarded the cams toward the US values at high RPM and load, but tapered them similar to the Euro maps to promote some scavenging at lower loads. At higher loads, this should be less stressful on the engine. I’ve attached a document from M3Forum where people discussed adjusting cam timing. I also richened the WOT fueling, which should help keep EGTs down and reduce stress on the drivetrain.
Why didn’t I just use the Euro VANOS maps? Good question. I know what I have now is safe. The Euro map seems to play with VANOS a lot more, presumably because they weren’t limited by the restrictive USDM cats. With my much freer-flowing exhaust, using the USDM values should be much easier on the engine.
I’d really like to do some VANOS testing on a dyno or through controlled logging to monitor airflow and see what actually benefits my setup. If I had a more standard dual exhaust system from the headers back, I’d definitely lean toward using the Euro VANOS targets.
How is it? I feel like part-throttle response is definitely improved around town, which is exactly what I was after. WOT feels like the engine is less strained compared to the canned tune. Since it’s running richer and using similar VANOS values—with just mild advance at lower loads—it should be safe. Knock values also looked good.
I understand the way I’m doing this isn’t ideal. Like I said, I’d love to get on a dyno, rent one for the day, and really find out what this setup wants.
That said, I feel like I’ve made sensible changes—nothing drastic, but ones that should only improve things and be much easier on the engine. I’ve also kept the factory safety features intact until I can test more aggressive settings in a controlled environment. I’ll attach the tune file in case anyone wants to take a look and share their thoughts. It’s a 2701, so be sure to use that definition.
This DME has a lot more going on than the one in my E28, but it’s been fun learning and experimenting so far.
Leave a comment: