Innovate LM2 Wideband Info

[VX L67 Getrag]

Not sure whether it's better or worse but with mine I cut the cig lighter plug off & put a blade terminal on the positive & ring terminal on the negative & feed it straight into the fuse box, pull out interior light fuse & use that & phillips head screw the earth near by!

I just like to make sure it powers up & ready to go before car is running, some cars loose power to cig when cranking!

[vsslpr]

i just used a fuse with a wire coming off it from jaycar (replaced the standard cig lighter fuse with it) and just plug directly into that with another 3A fast blow into the gauge. i used that same ground point you did but then changed it, im now earthing it to the PCM case- i did that to make sure the voltage to Pin A5 was the same as the gauge reference voltage in case this was skewing TP5 wideband readings

[VX L67 Getrag]

I don't think the PCM case is a good earth point, but others may know better!

[VL400]

You need to treat every piece of wire as a resistor. A larger diameter wire is low resistance and a smaller diameter wire a higher resistance. When powering a load there is current through both the positive and ground wires, with both wires having resistance a small amount of voltage will be dropped along its length. The reason lots of oddities in cars end up being ground problems is due to ground not being ground when large currents or switched devices are being powered.

So in this screen grab an LS1 coil is being charged and the voltage along the ground wire is being measured. Blue being the volt drop, while red is the signal from the PCM telling the coil to start charging and then fire...

5A automotive cable.png (9.44 KiB) Viewed 5733 times

At the peak of the coils current draw (which is actually around 6A, a multimeter measures the average power including coil off time and would prob read less than 1A) the ground wire near the coil is 400mV above a true ground voltage level. When the coil is not charging the blue line is 0V and the end of your cable is the same as ground.

If you increase the cable size the voltage drop is negligible, but this is a very large and short cable...

Larger cable size.png (9.12 KiB) Viewed 5733 times

The issue here is the cable sizes would become massive if you had one ground wire capable of not having any voltage drop at maximum load and tacked on all your devices to it. So the way around it is to have carefully selected ground points around the vehicle and then run a number of wires to this one point - often called star grounding. You then on the end of each of these ground wires choose carefully what is connected to each one. You would have all the noise sensitive low current devices (like oxygen sensor grounds, not the heater circuit) on one or a couple of wires and then other devices like ign coil and injector grounds on other wires. We dont care about a few hundred mV drop for the injectors or coil upsetting anything with them. The narrowband O2 sensor earth on the other hand, with a 0-1V swing if you tacked its earth wire on the end of the ground wire for the above coil a random error of 400mV while the coil is charging totally ruins any accuracy from the O2 sensor.


So tacking a wideband on to the end of existing ground wires for noise sensitive devices is a bad idea. The MTX-L uses a switching regulator, while it might only draw 1A when the heater is on measuring with a multimeter this is the average from a rapidly switching power supply (much like the coil above just faster but with less peak current). Pick a low current signal earth for the wideband earth and the PCM does not know that there is now a random voltage offset whenever the wideband heater switches on and off, it still thinks its a nice clean ground connection thats always 0V. Suddenly your CTS/MAP/TPS/narrowband O2s are going to be all over the place as it samples - sometimes it might be referenced to ground when the heater is off, other other times it will be something higher than ground.


Anyway, thats long enough and have not even touched on different load types causing more noise!

[vsslpr]

so the noise added by the wideband heater circuit can upset the readings of other sensors but how sensitive is the wideband to voltage offsets added by another noisy ground source? on a vs v6 behind the kick-panel under the steering wheel theres an eye terminal screwed onto a metal part of the body does anyone know what this earth point is for and would that be better than the PCM case? it sounds like ill be moving my wideband earth away from the PCM casing in case other sensor inputs are being affected.

[Holden202T]

just use the interior metal for the ground, ie. the metal part under the steering wheel that the fuse panel screws to or something similar.

[vsslpr]

so i took out my rh nb sensor and put the wideband in its place and the wideband is not at fault.. it reads very close to commanded afr when it is in this position! but interestingly i took some pics of both sensors:

wideband in its usual position just before the cat:

photo.JPG (2.61 MiB) Viewed 5539 times

narrowband:

image.jpeg (2.81 MiB) Viewed 5539 times

wideband after sitting in the narroband spot (upstream):

image2.jpeg (3.1 MiB) Viewed 5539 times