Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Missing ground?

Status
Not open for further replies.

Uluz2a6

New Member
2005 Cobalt 2.2

Just removed, rebuilt, and replaced engine. Primed it and all was ready to start but all I get is cranking.

Injector Harness has voltage (5V and 12V) and ground. However no injector pulse. I believe that it is triggered by the crankshaft position sensor through the ECM so I next verified voltage at sensor. I was surprised to find 2.5V on both sides of the connector instead of 5V on one side.

I am more comfortable doing engine mechanical than electrical but I believe having the voltage split like that indicates a missing ground on that circuit. That correct?

I verified continuity from connector all the way to the ECM pin. Also verified wires are in perfect condition.

Does anyone know what would cause the ECM to provide power to both sides of that circuit? Where do I look? Am I on the right track?

I examined the wiring harness extensively looking for any issue. Engine replacement went well and had no pinched wires or any issues. All ground wires are connected and clean. I see none missing.



Thanks in advance for any ideas.
 
I once had a weird problem on a car, when trying to start it didn't crank but a spot light came on instead. Turns out the earth strap to the engine was loose and the only path to earth was through the spot light. Sounds like whatever is earthing the engine is high impedance, check the earth strap.

Mike.
 
If it's an active three-wire sensor, that implies a short circuit, with the two equal length wires acting as resistors and causing half supply at the junction.

However if it is a two-wire inductive sensor, half supply is not unusual - one side is likely a "bias" voltage and the other one will pulse high/low as a projection or gap on the crank pulley passes it.
 
All ground/earth straps are firmly in place.

It is a 2 wire sensor that should (according to the pinout I read) only have a 5V signal feed on one side, the other should not have voltage.

If my info is incorrect, please advise. I freely admit that electronics baffle me.
 
It is a 2 wire sensor that should (according to the pinout I read) only have a 5V signal feed on one side, the other should not have voltage.
Two wire ones work on the same principle as a guitar pickup - just a coil of wire and a magnet in or near it. Iron moving closer or further produces a voltage, but no voltage across them with no movement.

The "0V" for such sensors in a 5V system is typically half supply, so the used signal relative to that stays within the zero to 5V range.

7433505.jpg
 
With the sensor disconnected I have the 2.5V on both sides. The test procedure i
Was using showed that I should only have 5V on one side and if voltage present on both that meant issue is "up stream" from that point and to correct that prior to continuing.

I understand how the Hall effect works (vaguely) but it cannot in this case because it's seeing constant voltage on both sides of the sensor that doesn't change with RPM.

The ECM is not even getting an RPM signal from the sensor. I feel the ECM is splitting the 5V signal feed across both because of an open ground to it or in it. I just don't have the electrical know-how to track down my error
 
Yes the ECM was unplugged multiple times to trace wires and verify continuity from pinout to sensors.

Battery disconnected each time as well as all the care taken to properly reinstall each time.
 
Put your meter on AC and measure across the sensor - the DC level will not change, but it should be producing an alternating high&low cycle at each edge of the pulley pegs or holes it goes over.

That should read as a small AC voltage.
 
crank position isn't the only thing needed to get it to run, it also needs the cam shaft position signal. That tells it when to actually fire the injectors. Since the crank makes 2 revolutions to the cams one revolution(4 stroke engine) it relies on the cam sensor to signal both injectors and ingnition. Doing it by the crank sensor alone would give two pulses with every rotation of the crank.
 
This engine has no camshaft position sensor. Only the crankshaft position sensor.

This issue (as I see it) is before that circuit. Something is causing the ECM to provide constant voltage to both sides of that connector.

My limited electrical knowledge leads me to believe this is caused when a ground is missing on the circuit, therefore allowing the value (5v) to split across the circuit, hence 2.5V and 2.5V.

That being said, I have no idea what would cause the ECM to open a ground internally (of that is where it is occurring) or what other circuit to the ECM would cause it.
 
I can do the AC test but it doesn't change because the 2.5V on both sides remains constant even while cranking.
 
Had to read a little about 2005 cobalt 2.2, never worked on one. Seems your correct about the lack of a cam sensor. Some how they read voltages on the 3-4 coil pack to trigger the injector timing. Are you getting spark? Did you clear all codes in the ECM?
 
It's battery was disconnected for 2 days while engine was removed and replaced so all history in ECM was erased. No codes are present. It will not be able to show a code until the engine actually fires up.

As far as spark goes, I don't know. When we discovered no fuel and the odd voltage on the crank sensor circuit we focused on that.

The ground on the back of the cylinder head for the fuel pump and ignition coils is intact and has continuity.

I have 12V 5V and ground on correct locations on both the ignition coil and injector harness.

I could verify spark if that would give you a direction to send me in.
 
I could verify spark if that would give you a direction to send me in.

Well one of the cobalt forums said the 3-4 coil pac is where the injector pulse is signaled from, instead of using a cam sensor, so to me if you have no spark you have no injector pulse. Just thought of this, did you check fuel pressure or fuel presence on the fuel rail? Will it run if you spray starting fluid in the air intake? Hard to diagnose things from a distance.
 
I didn't try spraying starter fluid in the intake because it's a fresh engine and if it catches, I really want to break in properly and stay running up to temp.

The fuel pressure at the rail is perfect, just no injector pulse. I will verify spark in morning. Thanks again.
 
By that I mean the AC value doesn't fluctuate either while cranking
After a bit of research, you should get a reading of around 200 millivolts AC with the meter connected directly across the sensor, at cranking speed.
(Note, not from either wire to ground).

You will get little or no output if the gap between the sensor and the pulley/wheel it senses it too large.
 
With harness disconnected I get the AC voltage within spec, but if I back probe sensor connector while plugged in I get no signal on AC setting. I assume that is due to 2.5V on both sides of connector
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top