[Revelate]

I did a forum search for oscillation issues and saw a wide variety of issues that most people blame on a tune. While a lot of the issues are likely caused by aftermarket tunes, I have discovered that the factory ECU programming can be the issue as well.

Car
2010 135i 6MT

Symptoms
Between ~3800RPM and ~5000RPM, I experience a power oscillation that causes the car to lightly buck during WOT acceleration.

Power mods in order of installation (vehicle new as of Jan 2010)
1) Dinan Stage 2 ECU - May 2010
2) Berk Street Performance exhaust - December 2010

Background
1/10 through 5/10 - The stock car ran as expected from January through May.

5/10 through 12/10 - After the Dinan Stage 2 ECU installation, I immediately noticed a power oscillation when accelerating hard (WOT) at various RPMs, primarily between 3000-4500RPM. I could feel and hear it in the car, but at the time it didn't seem to noticably affect acceleration, it was more of an annoyance.

12/10 through 6/11 - After the Berk exhaust installation, the oscillation became front and center as an audible surging, and it became more pronounced and obvious in the seat as well.

6/11 - I get the car dyno'd and see the following results:



The two pulls show variations of the same issue.

The first run (red) gets a bad start all the way from 3000RPM and develops a significant bounce all the way to 5000RPMs. A/F ratio is smooth (though inaccurately rich) and the boost is fairly steady.

The second run (blue) starts off better, but at 3800RPM begins a very bad oscillation that peaks between 4200-4500RPM then mostly disappears. There's also a sudden dip in boost between 4200-4500RPM that recovers quickly.

At this point I know something is wrong, and the swings in power are significant (almost 50HP from 4200 to 4300RPM).

I brought the car to the dealership in August for the HPFP recall, and they reflashed the Dinan ECU as well. We hoped that between these two changes, the problem would go away. It didn't.

I brought the car back in October and gave the tech a ride to reproduce the problem. They acknowleged the symptom, and the suggestion was to flash back to stock and see if it goes away, which I agreed with. After picking up the car and trying a few hard pulls, WOT from any RPM was smooth all the way to redline. With the stock ECU, the problem appeared to be gone... or so I thought. For my own sanity, I decided to get a stock ECU dyno as a troubleshooting/diagnostic effort for Dinan:



There is still odd behavior in the 4200-4400RPM range with a strange spike that I don't feel or hear on the street or the dyno. But it's there.

I took these results back to the dealership, and they shared it with Dinan. Dinan responds by saying they are aware of "about 25 Dinan tuned N54 cars" that experience the exact same behavior. They supposedly have narrowed it down to the throttle plate oscillating (even on a bone stock car), but I didn't get an explanation if it was a mechanical or electronic deficiency (the throttle body on our cars is electronically controlled). When the Dinan tune is added, the problem is simply exacerbated.

I imagine this would be the case for any tune, not just Dinan. I see forum posts about similar issues and everyone blames the tuner, but if the information I'm getting is correct, it's actually a BMW problem, at least in my case.

I'm told that Dinan is working with BMW engineers for a fix, and I will be pinging my SA every couple months to get updates if available.

[Potty_Pants]

thanks for the info... hope there is a fix, my car is subject to the same problem.

please note... doing an italian tune up sustained 4k-5k for a few minutes appears to eliminate the pulsating/surging.

strange how they are leaning towards a throttle plate (hardware) causing this problem.

who knows, I'll be happy when its fixed.

ppp

[JamesM3M5]

You could use a BT tool to log actual throttle position to see whether that's the reason the boost is wavering. Or it could be a faulty vacuum pump or boost control solenoid? The system uses vacuum to close the wastegates and generate boost - opposite of a typical wastegate/boost controller.

Test for boost leaks?

[335i E92]

Happened to me, fix was a new set of diverter valves!! Only cost around £50 and super quick installation.

Tried the boost control valve 1st which didn't fix, give it a shot!!

[bryce]

[speculation]

what's interesting is the rpm range. that's at or near the VANOS change-over event. i'd be curious to see if this is related to VANOS at all

[/speculation]

[Revelate]

Boost and throttle position have already been ruled out. I'm told that the plate is physically moving, but the throttle position is not.

If you look at the power curve, the oscillation peaks are higher, and the valleys lower, than the curve, which means the system is "holding" more power when it needs to. The boost is too steady to be directly responsible.

I thought about replacing the DVs just because, but since this is all a warranty issue right now (including the tune, DVs and everything else), I'm letting BMW and Dinan attempt to figure it out before I try anything on my own.

[JamesM3M5]

Originally Posted by Revelate

Boost and throttle position have already been ruled out. I'm told that the plate is physically moving, but the throttle position is not.

If you look at the power curve, the oscillation peaks are higher, and the valleys lower, than the curve, which means the system is "holding" more power when it needs to. The boost is too steady to be directly responsible.

I thought about replacing the DVs just because, but since this is all a warranty issue right now (including the tune, DVs and everything else), I'm letting BMW and Dinan attempt to figure it out before I try anything on my own.

How can the throttle plate move without seeing it in the actual throttle position parameter change? If that's the case, there is something wrong with your throttle. You said Dinan and BMW suspect the throttle plate is oscillating, and that could be seen in a log of actual throttle position.

[Revelate]

Originally Posted by JamesM3M5

How can the throttle plate move without seeing it in the actual throttle position parameter change? If that's the case, there is something wrong with your throttle. You said Dinan and BMW suspect the throttle plate is oscillating, and that could be seen in a log of actual throttle position.

I won't claim to know how much communication the ECU has with the throttle body (bidirectional) for requested vs actual. I'm just passing along when the BMW dealership is telling me, which they are being told from Dinan, so it's entirely possible that something is being lost in translation.

I understand it that the ECU requests a throttle plate position, but the plate itself is oscillating on its own and not being "reported" to the ECU. Either it's a bug in the programming, or a mechanical/harmonic issue with the throttle body.

[timo1980]

Update?

[Revelate]

Originally Posted by timo1980

Update?

Update: Dinan blames BMW programming and says they won't try to correct for it... too much R&D to fix someone else's problem.

After getting nowhere with Dinan, I purchased a Cobb AP in April and flashed it right over the Dinan software. Oscillation gone! The Cobb throttle response is a little jumpy compared to Dinan, but it's not annoying. Reported this information back to Dinan and got the response I expected (paraphrasing) - "We don't make Cobb software, we don't know why it works and ours doesn't. It's still BMW's problem and we won't try to correct it."

My local dealership has been great through all this. They even agreed to do a complete car software upgrade (every module and the ECU) and reflash the Dinan software in an attempt to get rid of the problem. Unfortunately that didn't work, so I installed the Cobb software again.

I then tried to get a refund on the Dinan software for being faulty (since Dinan offers no other way to sell or transfer their software after purchase) and they would only agree to a $500 refund. I laughed. I'll keep the tune and cause as much grief with Dinan as possible. Which reminds me, it's time to call Dinan and bitch.

[ragingpanda]

In before Dinan troops storm the thread lol.

Congrats on seeing the light.

[shortseller]

Originally Posted by ragingpanda

In before Dinan troops storm the thread lol.

Congrats on seeing the light.

Ok, here we come, guys are you ready....what! Not coming, don't care, fine!

Sorry no troop storm today, they just don't care.

I guess I'm one of the others(outside the group of 25), that are having no problems with my Dinan flash. I feel very bad for revelate and I'm glad to hear he solved his problem with a different flash, and IMO it's pretty shitty that Dinan won't compensate him for the issue, especially after he tried so hard to solve it within Dinan patronage.

Btw, take a look around, no one has a perfect flash if you read the forum long enough. So save your enlightenment for real more serious Dinan bashing.