[inobu]

I have a 2007 335i that had DME water damage. I was able to clear all the codes and get it running again. This thing was a mess. I had to re-solder a pin, run new jumpers between boards and clean up all the corrosion.

The last issues is the Barometric Pressure sensor.

I was able to trace everything back to the Ifineon TC1796 MCU. I replaced the Baro sensor MPXHZ16115A. In the interim I put the old one on a breadboard and it tested good (wasted $31).

Before I change out the DME I want to make sure that there's nothing else to do. I have hard time believing this is a defective port on the chip based on the fact that everything else is working.

I isolated the Baro Sensor signal going to the TC1796 and when I run Messwertblock ADC the DME Ambient Pressure Sensor is showing 5 volts.

This throws the P2229 code.

To further isolate the problem I placed jumper into the via on the PCB and ran them out of the DME so I could take real time measurements while the DME was dunning the car.

The input to the TC1796 was 3.06v but the DME showed 5.00v

I lifted the signal leg of the Baro Sensor and the Messwertblock ADC the DME Ambient Pressure Sensor started showing 1.85v.

When I took a voltage reading the DME Ambient Pressure Sensor on the ADC went to 0v.

When I was tapped on the input to the TC1796 I could see the 3.06v from the Baro Sensor and it would go to 0v when the DME went to sleep.

Does anyone have any insight to this issue or have a internal schematic of the DME MSD80.

Thanks,

Inobu

[inobu]

The issue maybe a corruption problem more so than a hardware fault.

The solution may be in rebooting or reinitialize the TC1796 MCU. There is a Power On reset and a Hardware Reset. I'm hoping that a Hardware Reset will
cause the MCU to reset the registers that will clear the problem.

There is a pin called "HDRST" on A23 I/O A2 H

Hardware Reset Input / Reset Indication Output (open drain pad with input spike-filter.)

The other pin of interest is "BYPASS" A24 I A1 PLL Bypass Select Input

This input has to be held stable between to power-on resets. With BYPASS = 1 the spike filters in the HDRST, PORST, and NMI inputs are switched off

I'm speculating that the "ByPass" will just perform the reset out right.

This is what I'm thinking.

Hard Reset is a class 2 reset while a soft reset is a class 3 reset. The class 2 automatically initiates a class 3. The main difference is the state of the processor and the peripherals. In the case of class 2 the processor is reset and the peripherals are reset. This is most obvious in looking at the peripheral registers and seeing the ‘reset’ values indicated in the TRM. With the class 3 there is more subtlety with the state of the M4 (debug PPB ******* and the peripherals are not reset.

Hopefully someone will chime in.....

Inobu