I was reading a bit today and saw this article from Haltech:
https://www.haltech.com/news-events/...dos-and-donts/
and this post on HPAcademy, specifically the one by Chris250
https://www.hpacademy.com/forum/gene...round-question
Two sections stood out to me:
Quote:
|
This second example is one that I used to use when I interviewed support engineers to work for us, but I guess now I'll have to think of another question to evaluate people. It's actually a mistake made by Mazda in the factory NA6 MX5 / Miata loom; they fixed it on the next model (the NA8). The sensor ground is externally grounded to the engine, as well as to the ECU. The ECU is grounded to the engine also. We know that as the injector duty cycle increases, the average ground current of the ECU will also increase, and therefore so will the voltage drop between the ECU and the engine. The ECU ground will be sitting at a slightly higher voltage than the engine ground, so any sensors connected to the engine ground instead of to sensor ground on the ECU, will read a lower voltage. In the case of coolant temperature, it means that as injector duty cycle increases, the voltage seen by the ECU on the coolant temperature input reduces and the ECU believes the engine is at a higher temperture. This is fairly easy to spot in logs; you can look for noise on inputs such as coolant temperature, throttle position etc, but it causes problems for obvious reasons.
|
Quote:
|
Sensors, if they have the ground isolated from the sensor body (for example TPS, pressure, temperature sensors), they must be grounded to the sensor ground on the ECU, not the engine ground. The reason for this is as given in the second example above; if you ground the sensor to the engine ground, then its output is going to be offset by the ground current times the ground resistance. This is why ECUs have dedicated sensor ground wires.
|
Sooo, I unplugged the ECU and found that the sensor ground still had continuity. Started pulling grounds and found that the sensor ground wire was being grounded through the two eye rings on the valve cover.
I plugged the ECU back in without those two eye rings grounded and the ECU wouldn't turn on. When I grounded either of the two valve cover eye rings and the ECU powered on.
Seems like progress is being made, I guess.