Skip to content
Shop
Explore

INDEPENDENT BMW ENGINE EXPERTISE

Rough run. VANOS.

N43 / N53 · NOXEM · DIAGNOSTICS

Select a technical image to view it in detail.

This topic applies not only to the N43/N53 engines but to other BMW petrol engines equipped with VANOS. It came up because of my friend Rob.

How the case started

Rob took the restoration of his car’s engine very seriously. First, the NOx system was restored using NOXEM. Then new adaptations were created, and the MSD80 software was updated to the newest available release. The engine improved after each step, but some unevenness remained. No related error messages were stored in the MSD80, and the cylinder misfire counters were close to 0.

The mystery was solved when the engine eventually switched to emergency mode and this fault was recorded:

2A99 crankshaft – exhaust camshaft, correlation

After the exhaust VANOS valve was replaced, the engine ran normally and the unevenness was gone.

This case led me to study the hidden causes of VANOS defects. The key word is “hidden”: ordinary defects (sensor, solenoid, mechanical problem) are recognised by self-diagnostics, and VANOS repair is simple with easily available parts.

Why VANOS faults can stay hidden

Information on the basic operating principle of VANOS is easy to find, for example: https://www.youtube.com/watch?v=hW_1CBoLUCQ

The system is simple. It is controlled only by a solenoid driven by the DME, plus a camshaft sensor that identifies the actual camshaft position. Even so, it has nuances that must be taken into account.

VANOS has no position sensors of its own that would immediately report the achieved camshaft position to the DME. The position is derived from the signal of the camshaft gearwheel as it turns. Even at 2000–4000 RPM, the pulse frequency is only a few tens of Hz, so the camshaft position information is refreshed relatively rarely. A short-term mismatch between requested and confirmed position is therefore possible. The DME does not “see” it, and even if it does, it cannot eliminate it completely, for example if the VANOS mechanism reacts to a pull or temporarily sticks. In engineering terms, the data update time is close to the system’s reaction time, which is not good practice.

The fault 2A99 (crankshaft – exhaust camshaft, correlation) means the following:

  • The data from the camshaft sensor (signal amplitude, shape, frequency) are within normal limits.
  • The calculated camshaft position changes, so the mechanism is not mechanically jammed.
  • The difference between requested and confirmed position has been outside the norm for a long time.

As usual, the manufacturer gives no information on the conditions under which this fault is stored. We therefore had to investigate whether a mismatch between requested and confirmed position is visible with diagnostic tools such as INPA. This is complicated by the fact that INPA live data is refreshed only a few times per second, so a short-term problem can be missed.

Test car: N53B30

From several suspicious vehicles with this possible problem, I chose one with an N53B30 engine:

  • The NOx system works correctly.
  • There are no error messages in the MSD memory.
  • The cylinder misfire counters are close to 0.

The only suspicious mode is 110–130 km/h while the NOx catalytic converter is being desulphurised. In this mode the engine runs unevenly. The unevenness is much “softer” than with misfires, but it can still be felt.

At idle, INPA live data looks correct in all menus, including VANOS. The VANOS menu shows a slightly increased offset adaptation of the exhaust camshaft (around 4–5 degrees), but the requested and confirmed positions of both camshafts match and are very stable.

The content of the VANOS menu:

1. Offset adaptation of the intake camshaft 2. Offset adaptation of the exhaust camshaft 3. Confirmed position of the intake camshaft 4. Confirmed position of the exhaust camshaft 5. Requested position of the intake camshaft 6. Requested position of the exhaust camshaft

Live data check with INPA, VANOS menu:

  • At the start of the video the engine idles. Requested and confirmed positions match precisely, and the confirmed positions are very stable.
  • In the middle part the accelerator pedal is pressed quickly. The confirmed positions of both camshafts follow the requested ones, so VANOS clearly works.
  • At the end the engine is held at around 3000 RPM. The confirmed position of the exhaust camshaft now looks suspicious: it does not stay in place but keeps changing. The requested value is 115.4 degrees, while the confirmed position moves between 113.1 and 118.4 degrees, a fluctuation of 5.3 degrees.

Live data under the problem conditions

In the problematic driving conditions (110–130 km/h, desulphurisation of the NOx catalytic converter), INPA shows:

  • In both banks lambda is maintained correctly, and the integrators change by only a few percent.
  • The misfire counters are 0 for all cylinders.
  • According to the air mass meter, the amount of air changes by ±10–15%.
  • The high-pressure pump pressure does not change. The requested torque (the problem occurs with cruise control on an even road) and the engine RPM do not change either.

The combination is interesting: the air mass varies while the mixture, load and RPM stay constant, which seems contradictory.

The VANOS position menu in these conditions:

The video shows:

  • The confirmed position of the intake camshaft is stable and matches the requested one.
  • The requested position of the exhaust camshaft is unchanged at 120.2 degrees, but the confirmed position changes from 115.2 to 123.6 degrees. It fluctuates by at least 8.4 degrees.

The resolution of the displayed position is 0.3–0.4 degrees, as the intake camshaft data show: its position moves by ±1 “step”. The exhaust camshaft position moves by as much as ±10 “steps”, so it is held about 10 times less precisely.

Conclusion

The exhaust camshaft VANOS unit has a problem. Because there are operating conditions in which the defect does not appear (the confirmed position stays stable), the cause is the solenoid, not increased angular play in the VANOS rings or the shaft drive.

The camshaft position problem also explains the seemingly contradictory data: changing air consumption with an unchanged mixture at unchanged engine load and RPM. When the camshaft phase changes, the engine’s “breathing” changes, meaning cylinder filling and the exhaust gas outflow from the combustion chamber. This briefly changes the intake manifold depression and therefore the measured amount of air.

A BMW dealer will not see such a problem, because:

  • The dealer is not officially allowed to use INPA.
  • The dealer will not normally take the customer’s car on a test drive to analyse live data and look for hidden defects.
  • ISTA D refreshes data on screen even more slowly, in my subjective impression. It offers no visual representation such as bars, and the displayed data are small and hard to read, so this defect is simply not noticeable in ISTA.

[[[KEEP2]]](https://bimmerprofs.com/wp-content/uploads/2018/01/bmw_vanos.png)

Part number of the solenoid for N53 engines: 11367585425

[[[KEEP3]]](https://bimmerprofs.com/wp-content/uploads/2018/01/vanos_valve.png)

Note: the intake and exhaust camshafts use identical solenoids. The same ones are also used in N52 engines.

Results after replacing the solenoid

This video shows the accuracy of the confirmed exhaust VANOS position at 2000–3000 RPM (in neutral) after the solenoid replacement:

This one shows the accuracy of the confirmed position in different driving conditions (60–120 km/h):

The jitter has dropped to ±1 “step”, and the unevenness has disappeared.

Observing VANOS PWM values

VANOS performance can also be observed conveniently in the VANOS test block. MSD80 loader 2.023 also lets you view the PWM values supplied to the VANOS valves (../F6/F1/F2). At idle, the PWM for both VANOS valves is 10%, which means the valves are deactivated.

[[[KEEP4]]](https://bimmerprofs.com/wp-content/uploads/2018/01/vanos_valves_2.png)

Above 1000 RPM the valves start to hold the required position.

Correct PWM values in steady operation (for example 90–100 km/h in 5th/6th gear) are 30–50%. A short-term fluctuation of ±2–3% is acceptable while the requested position is stable.

[[[KEEP5]]](https://bimmerprofs.com/wp-content/uploads/2018/01/vanos_valves_on.png)

Suppose that at the moment the requested VANOS position changes (the engine operating mode changes), the valve PWM briefly drops rapidly to 10% or rises rapidly to 90%. In that moment VANOS has trouble holding the requested camshaft position.

If the valve PWM shows increased jitter and regularly reaches 10% or 90%, the VANOS unit has trouble holding the required position. This can cause uneven engine running even when there are no VANOS-related error messages.

Related entries

Oil supply. N series engines

MSD80 diagnostics

Cold engine and problems of injectors

N53. Problem of cylinder No3

N43 and strange behavior

BEYOND THE FAULT CODE

A clearer picture.
A better next step.

Get in touch
Enlarged technical image
Open original image