G28, G40 and G4 sensors

[BrianSmith_S2]

Some may have read my fuel pump post and my dead S2 I replaced the G40 only to find the old one was fine

Now going back to basics and fault finding the wiring and sensors.

First check was to check the ECU was seeing a signal line from the sensors. I get continuity between pin 2 on the

G40 sensor to pin 8 on the ECU
G28 sensor to pin 49 on the ECU
G4 sensor to pin 47 on the ECU

Second check was resistence check on sensor G28 and G4

G28
0.97k between pins 1 & 2 (sensor disconnected from wiring)
0.95k between pins 1 & 2, 2 & 3

G4
0.97k between pins 1 & 2 (sensor disconnected from wiring)
0.95k between pins 1 & 2, 2 & 3

Third check with ignition on i checked voltages
G28
0.8V (0.9V when turning over) across pins 2 and 3, 0.1V across pins 1 and 3

G4
0.1V (0.1V when turning over) across pins 2 and 3, 0.1V across pins 1 and 3

Do these look ok?

I turned engine over today with fuel pump relay cover off and it does energise. I was told at ADI that someone had also done fuel pump test, i.e. fuel lines disconnected and ignition on and they saw fuel spray out, but it didnt pump when connected, i.e. it may have not worked under full load/pressure. Gonna buy back the fuel pump i returned to G&S last week as im running out of ideas

[S2central.net]

You can learn almost nothing from such readings - you need an oscilloscope to look at signals from G4, G28 & G40 to have any confidence in them.

The LED trick is pretty reliable for finding a dead G40 though.

Have you checked all the stupid things like fuses and earth points ?

Fuse 32 is for the coild for example, and another fuse (will look it up) is for the fuel pump.

It might be a security thing too - will PM you an idea to check that the ECU is getting powered up properly.

[BrianSmith_S2]

You can learn almost nothing from such readings - you need an oscilloscope to look at signals from G4, G28 & G40 to have any confidence in them.

The LED trick is pretty reliable for finding a dead G40 though.

Have you checked all the stupid things like fuses and earth points ?

Fuse 32 is for the coild for example, and another fuse (will look it up) is for the fuel pump.

It might be a security thing too - will PM you an idea to check that the ECU is getting powered up properly.

Thanks for your help, im going to try and borrow an oscilliscope later today

[S2central.net]

Make sure and get a two channel 'scope if at all possible - but dont forget those fuses !

[BrianSmith_S2]

Make sure and get a two channel 'scope if at all possible - but dont forget those fuses !

Yep, its a 2 channel scope, and ive been through all the fuses in the engine bay relay box to check for blown fuses.... i suppose i better also check im getting power to them as well.

[Fat Audi 80]

Good luck Brian, it sounds like a nightmare...

Cheers,

Steve.

[BrianSmith_S2]

Just connected scope to the three sensors and this is what i see when turning her over.

The G40 seems to be pulsing every 700 ms (approx) so the starter motor is achieving 170 rpm.
The G28 is pulsing every 2.6ms
The G4 is (the odd one) i calculate should be pulsing every 350ms but seem to be every 1000ms on runs 1, and 600ms on runs 2 and 3

If it were to idle at 800 rpm (13.3 rps) this would equate to .......
G40 pulse every 150ms
G28 pulse 0.55ms
G4 pulse 75ms

[S2central.net]

G40 looks OK

G28 - should have much higher amplitude - like 15 to 20V peak to peak - guessing you had a '10X probe' attached

G4 - similar comment about amplitude, but more worrying is that the pulse is just the wrong shape...

I will dig out a post for waves I captured before... You need to use both channels to verify that the G40 pulse is timed such that G4 from crank arrives in the middle of every 2nd G40 pulse.

[S2central.net]

Look at the G4 signal I captured at idle on my car - seen on others too...

http://www.s2forum.com/forum/showthread.php?t=30947

Your spikes look rather different - hope you can see that... might you have damaged the crank reference pin I wonder ?

[BrianSmith_S2]

The wave form on your example for the G28 was 4V peak to peak, about the same as mines.

Would rpm have an effect on voltage seen at the sensor as its an inductive type, i.e the faster the flywheel teeth pass by, the higher the voltage?

[S2central.net]

Here is a better pic of all three signals together...

http://www.s2forum.com/forum/attachment.php?attachmentid=27290&d=1276876987

BLUE is G4
YELLOW is G28
GREEN is G40

You can see how this trace was triggered with G4 inside the G40 window. Try and experiment with your scope to get the best captures... It is trickier at cranking speeds as opposed to idle obviously but you need to see that G4 signal accurately.

[S2central.net]

Brian - look at the most recent pic I attached... much clearer and everything at 1X scaling. Its about 20V pk to pk when measured on a live ECU.

As an inductive sensor - you simply get more pulses per second as the speed rises - there is 135 pulses every crank revolution - one for every flywheel tooth.

[BrianSmith_S2]

Look at the G4 signal I captured at idle on my car - seen on others too...

http://www.s2forum.com/forum/showthread.php?t=30947

Your spikes look rather different - hope you can see that... might you have damaged the crank reference pin I wonder ?

I hope not its still pulses + 1V then - 1 V but with a 600 ms spacing instead of the expected 350 ms (half that of the G40)

[S2central.net]

Look at the shape... mine makes crisp negative edge before an even sharper positive edge - just as the pin passes the sensor... Yours looks odd - maybe its a bad capture on the scope. You want to be set for AC trigger on a negative edge to capture it - try timebase circa 1ms or thereabouts.

[S2central.net]

Your traces have a range of 500mV - so you are either -

1 - just picking up noise and not triggering on G4 correctly
2 - crank reference pin is broken off
3 - G4 sensor is dead... you know it is same part as G29 sensor - so swap them over and see if badness moves to G28 signal with same sensor and hope that you can see G4 signal with the other sensor