G149 Modulating Piston Displacement Sensor

[Fix_Until_Broke]

I'm trying to record this sensor on my DAQ system and need to understand fundamentally how it works so I know how to both signal condition and interpret it.

On the surface, it looks like an LVDT, however that's kind of tough with only 3 wires I'm guessing that it's a DVRT, but wondered if anyone has played around with this before?.

My multimeter shows ~2.5vDC on all 3 terminals to ground so that confirms an AC supply & signal I think - guessing that it's 0-5v?

Between Pins 1 & 3 I get:
0.420vAC at idle no accessories on
0.520vAC at idle with all accessories on
This seems like what VCDS is recording/reporting (mine is out on loan right now otherwise I'd verify it myself)

Between Pins 1 & 2 I get:
0.205vAC at idle no accessories on
0.205vAC at idle no accessories on

Between Pins 2 & 3 I get:
0.162vAC at idle no accessories on
0.173vAC at idle no accessories on

Presumably as the core moves it induces a larger AC voltage across pins 1 & 3?

 

[Fix_Until_Broke]

Bump for anyone who might know something about this?

 

[oldpoopie]

Drivbiwire in 3..... 2..... 1......

 

[Digital Corpus]

Tell me which wires on a 10 mm harness they are and I'll hook my scope up to them and send you the capture. I've done this for the lift needle sensor: http://forums.tdiclub.com/showthread.php?p=4713198

 

[Fix_Until_Broke]

Pins 1, 2 & 3 as shown in the above diagram (in the yellow box)

 

[Digital Corpus]

...and 4 for ground. I've warmed the engine and have done some reving looking at pin 1, the source signal, and pins 2 & 3, the outputs. I'm tweaking the setup I have so I can trigger a capture on my scope since I'm doing this w/o an assistant. I'll have screen caps and details once I have captured something decent to show what I can see going on.

Anyhow, for preliminary signals:

• We're dealing with a 10 kHz signal across the board
• Pin 1 is a ~2.24-2.25 V-pp w/ ~2.64 V offset
• Pin 1 is static across the board regardless of IQ.
• Pin 2 has a max of 120 mV-pp and a min of ~30 mV-pp w/ ~2.64 V offset
• Pin 2 is a noisy mofo but also appears to have other frequencies mixed in
• Pin 2's amplitude grows as IQ decreases. An IQ of 8 yields the aforementioned max.
• Pin 3 has a min of ~1.29 V-pp and a max of ~3.6 V-pp w/ ~2.53 V offset
• Pin 3 is as clean as Pin 1.
• Pin 3's amplitude shrinks as IQ increases and grows as it decreases.

 

[Digital Corpus]

Here are 6 animations of the signals for you. There were taken at 5 ms intervals and I cut out the long delay between the 'Startup' and 'Idle' compilations. They were taken in one capture so if you just take the frame count difference and multiply it by 5 ms, you have the length of time in between.

It looks like the method I opted to use was slightly flakey because when the engine started up, my triggering jumped around a little because the ground clip was a little loose thus inverting the signal.

• The 'Startup' gifs is what happens when you turn the power on, but don't turn to the ignition.
• The 'Idle' gifs is what happens if you leave the car sit in the above state until you hear the relay click after a few seconds
• The 'Engine Start' gifs are turning the ignition fully on from the 'Idle' state up until the output has evened out. On a hot engine, the result is the same if you start from the 'Startup' state.
• Each compilation has a 100 ms pad before and after the major event, i.e. 20 frames.
• Pin 3 is inverted from Pin 1, not out of phase by 180°
• Pin 2 is out of phase by 90° afaik

Pin 3 Startup.gif - ~1.1 MB
Pin 3 Idle.gif - ~621 KB
Pin 3 Engine Start.gif - ~2.7 MB

Pin 2 Startup.gif - ~900 KB
Pin 2 Idle.gif - ~990 KB
Pin 2 Engine Start.gif - ~2.3 MB

You'll see two types of pulses in the Pin 2 Engine Start animation. One that is clearly defined appears to be injection events. One that is just sporadic noise is likely caused by the vibration of the engine inducing a voltage in the MLCC in the probe. The general noise that is present though isn't caused by my hookups.

Edit: If you want any specific captures in CSV form of any portions or parts of the aformentioned waveforms, let me know. The whiskers are soldered on and I can leave be for a spell. This is on an FA ECU, stock software, Sprint 520s, warmed up IQ of ~5.4.

 

[Fix_Until_Broke]

Thanks Digital Corpus - Those are great traces.

Could I trouble you to take some differential measurements between 1 & 3 please? Looking at the schematic, I'm guessing that the signal is a differential between one of the 3 wires on the sensor.

Is the yellow trace Pin 1 to ground?

 

[Digital Corpus]

Yes, Pin 1 is the yellow trace in each case, channel 1 in particular.

I don't have a differential probe, but I'll see if it won't harm anything by hooking my ground lead to Pin 1 and recording pin 3. Do you want the same outputs?

 

[Fix_Until_Broke]

DC - would you be able to try those differential measurements by chance?

 

[Digital Corpus]

I was waiting for a reply to my question of if you wanted the same type of readouts, though in the past 2 weeks I've been battling home network problems. Today it went down. I might have time after work tomorrow.

 

[Fix_Until_Broke]

Sorry for missing the question.

Differential 1-2, 2-3, 1-3 from startup, cranking, idling if you could please.

No hurry on my end, just following up.

 

[Digital Corpus]

I've not forgot. I hope to get to these tomorrow afternoon.

 

[Fix_Until_Broke]

cool - no rush on my end, thanks for giving this a try!