My sensor grounds have been wrong for years!

Discuss which Sensors work best with these ECUs and share how you are using inputs and outputs
bazibbo
Posts: 63
Joined: Sun Nov 06, 2011 1:04 pm
ECU Model: S40 Pro
Location: East Yorkshire

My sensor grounds have been wrong for years!

Post by bazibbo »

I’ve just found a few wrong connections in my loom, and I’m a little concerned about rushing in and making things worse.

Normally, I’d avoid the pin ends unless I’ve got something simple to fit – shift light or tacho output – as I’m somewhat challenged electronically. I got brave and added an oil temp to pin 25, which is when I noticed the earths were all messed up. Before you all shout, as you will, ‘just fix it’ bear in mind that I could potentially mess things up. I’m wary of the pins as I can never get them far enough in without damaging the ‘wire’ where it enters the pin – where you’ve got to use long nose pliers to push the final millimetre, and the wire concertinas up.

So, should I leave it? It’s worked like this for three years or so?

Background – S40 dta pro, on a fuel injected, turboed aircooled VW beetle. It does run, though probably not fantastically and the wiring faults are:


Pin 29 - goes direct to battery -ve
Pin 28 - goes to crank sensor only

Pin 13 - links to air temp / water temp / TPS / Map sensor


Last thing of note is that my lambda isn’t grounded via the ECU sensor ground 29, mainly because of the rear engine layout and my AFR gauge / controller, Innovate MTX-L, being on the dash.

Whilst being wary of pulling the pins out, it does look like I can leave the pins in and adjust the wiring instead.

Do you think this could be causing issues?

Thanks guys, so, do I dive in?
Alex DTA
Posts: 1622
Joined: Fri Feb 28, 2014 1:03 pm
ECU Model: S40 Pro
Distributor: DTA
Firmware Version: 79

Re: My sensor grounds have been wrong for years!

Post by Alex DTA »

Just swap pins 29 and 13 around.

Use a paper clip to push them out (never pull them out using the wires) and then use the paper clip to seat the pins properly.
bazibbo
Posts: 63
Joined: Sun Nov 06, 2011 1:04 pm
ECU Model: S40 Pro
Location: East Yorkshire

Re: My sensor grounds have been wrong for years!

Post by bazibbo »

Wow that's an easy and elegant solution Alex :D

I was stressed, thinking that there would be loads of wires to cut and join, and the paperclip idea is great too - thanks.


Does anyone think that this could have been causing any issues previously?
Alex DTA
Posts: 1622
Joined: Fri Feb 28, 2014 1:03 pm
ECU Model: S40 Pro
Distributor: DTA
Firmware Version: 79

Re: My sensor grounds have been wrong for years!

Post by Alex DTA »

Probably would have caused strange issues. Signal ground and power ground being mixed up is never a good idea.
bazibbo
Posts: 63
Joined: Sun Nov 06, 2011 1:04 pm
ECU Model: S40 Pro
Location: East Yorkshire

Re: My sensor grounds have been wrong for years!

Post by bazibbo »

Thanks Alex, an easy 10 minute job.
Rob Stevens
Posts: 1247
Joined: Thu Jul 28, 2011 4:10 am
ECU Model: S100 Pro
Distributor: None-Coil On Plug :)

Re: My sensor grounds have been wrong for years!

Post by Rob Stevens »

Am I wrong in thinking that all the grounds are just common anyway inside the box?

You should really reference the Lambda to the sensor ground. you will certainly get oating voltages otherwise?
Alex DTA
Posts: 1622
Joined: Fri Feb 28, 2014 1:03 pm
ECU Model: S40 Pro
Distributor: DTA
Firmware Version: 79

Re: My sensor grounds have been wrong for years!

Post by Alex DTA »

Rob Stevens wrote:Am I wrong in thinking that all the grounds are just common anyway inside the box?
Very wrong. Sensor ground (5v) is completely insulated from power ground (12v).
stevieturbo
Posts: 3577
Joined: Tue Aug 02, 2011 12:08 pm
ECU Model: No ECU
Location: Norn Iron

Re: My sensor grounds have been wrong for years!

Post by stevieturbo »

Alex DTA wrote:
Rob Stevens wrote:Am I wrong in thinking that all the grounds are just common anyway inside the box?
Very wrong. Sensor ground (5v) is completely insulated from power ground (12v).
stating voltages with that is a little confusing.

And pretty sure if you were to test for continuity between sensor ground and power ground...you'd get a beep ?

Ultimately they must be connected together somewhere, at least they are on most ecu's including OEM.

But the "sensor" ground aspect is like a clean earth. From the ecu outwards there should never be anything else that can affect or influence it other than the sensors themselves
Which kind of makes hooking up external widebands awkward, as most dont actually provide a sensor ground to go with their analogue output, as they just rely on their own power ground for this.
Alex DTA
Posts: 1622
Joined: Fri Feb 28, 2014 1:03 pm
ECU Model: S40 Pro
Distributor: DTA
Firmware Version: 79

Re: My sensor grounds have been wrong for years!

Post by Alex DTA »

Fair point. I was trying (badly) to separate the two.
DTA Gareth
Posts: 37
Joined: Thu Feb 18, 2016 4:12 pm
ECU Model: No ECU

Re: My sensor grounds have been wrong for years!

Post by DTA Gareth »

i can give a bit of detail from my days as a calibration engineer at Rolls-Royce.
ground and signal ground will ultimately end up in the same place, correct.

however,

a sensor ground will be damped by using a capacitor.
Capacitors only pass an alternating current, not a Direct current, so in this context the capacitor offers an easy route for and AC noise to leave the system but keeps the DC signal which we used to read.
its known as de-coupling and commonly used in power supplies, i strongly suspect something very similar is used on the signal ground too.

apologies if i just stated something already known, else, here is a link to explain it far better than i can.
http://www.capacitorguide.com/coupling-and-decoupling/
Post Reply