9.3v from ECU to starter solenoid signal wire

[ryanthecarguy]

I have a DF goblin kit car that I bought 3rd hand and uses a 5-speed 2008 Chevy Cobalt SS TC as it's donor.

I had to engine swap with a new third gen due to porous block from my LNF. Now the car doesn't crank using the key. Will crank when jumping the starter relay. After cleaning a bunch of corrosion, and tracing the circuit backwards with several continuity tests, I find that there is only 9.3v coming from the ECU to the signal wire mind you, when the key is turned to run. Turning to start has no change. I scraped off a little insulation from the purple and white wire right next to the X2 plug to be absolutely sure. Am I in need of a new ECU or am I missing something?

I should mention that the clutch position sensor is zip tied to simulate the clutch being pushed in.

 

[Desert Sasqwatch]

The clutch position switch needs to be active in the ECU - among other signals such as from the LAN, throttle pedal position, and ignition switch - for it to have the correct set of signals to trigger a start enable. The signal wire with the 9.3V you're referring to is the starter wire (purple)? Do you have any OBD codes?

 

[ryanthecarguy]

I'm referring to the purple wire with a white stripe that runs to the fuse box from the ECU. Not the solid purple wire that runs to the solenoid from the fuse box. I had codes but none that had to do with the starting system if I remember correctly. I went over it with my dad and he also thought none of the codes should matter in this situation. I can run out and get the codes again after I shower if that helps. Would the ECU spit out low voltage if any of those starting criteria were not met?

 

[Desert Sasqwatch]

The clutch switch is a variable signal, not an on/off switch, and is 'calibrated' to the clutch pedal position. Zip tying it to simulate the clutch pedal being pressed may or may not match the ECU 'learned' position for the clutch pedal. This is what I was getting at for the needed signals in the ECU to enable a starter signal. I don't have any of the schematics with me, I'm not at home, but the required inputs to the ECU need to be present before a valid start signal will be output to the fusebox starter relay.

Did the engine start and run previously or did you purchase before it was at that point from the builder(s)?

 

[ryanthecarguy]

Ahhhh I see what you mean. The engine ran perfectly fine on purchase for 3 months back in July 2022. Coolant was getting in the oil so it still ran fine but the replacement was necessary regardless. Projects been on hold for all this time. Attached is the full scan I just did with my bluedriver obd scanner. I scanned my sister's car before mine and I guess the vehicle tag is just wrong at the top of the page. My sister's car is an automatic and is having transmission issues.

 

[Gtstorey]

What is your battery voltage? What are you using for ground when you get a measurement of 9.3 volts? I think that should typically be 12v but if the relay that is closing has good connections all the way to ground, 9v should pull the relay in. I’ve tried it with a 9v battery before.

That signal voltage won’t vary because of anything like the clutch switch since all it does is close the relay. If it is zero volts that goes to 9v when you turn the key, the ECM is trying to start the car.
My guess is you still have corrosion and bad connections. Maybe just a weak battery.

 

[ryanthecarguy]

Battery voltage is 12.25. When measuring, I touch to engine block directly or that frame stud where all those grounds are relocated to. I've confirmed that the relay has a good ground. I actually ran a new ground because I wasn't getting one initially. I also replaced the relay with one I got from a gm dealership.

Using the continuity mode, I've tested the connection where the relay plugs in, through the board to where it comes out of the board. No resistance.

The plastic block the board plugs into, through the purple and white wire to pin 57 of the X2 ECU connection. No resistance.

Which all makes sense as the reading on the fuse box where the relay plugs in and the purple and white wire as close to the ECU as possible have the same reading.

 

[Gtstorey]

Continuity test only means a little since the amperage for that is very small. See if that 9v will light up an incandescent test light across the signal connections at the relay. Not a led test light. Or something like a brake light bulb.

I would also confirm your voltage measurement directly to the negative terminal of the battery. Also check your voltage feeds to the ECM to confirm full battery voltage at that point.
And confirm the 9v comes and goes based upon key in the cranking position to eliminate phantom voltage in the ECM.

 

[ryanthecarguy]

Okay I will try all that. A few questions though.

I did try testing the voltage going to the ECM. At least I think. I tried measuring the battery voltage at pin 56, and the ignition pins at 3,5, and 6. I got nothing.

My technique must be wrong or something since the car does come alive a bit when the key is turned, brake lights, throttle body flap, etc.

I'm using about 4 inches of wire with the ends stripped. Seperating just a few strands to shove in the X2 pin ports, enough so I don't have to hold it. Then put the black lead to either chassis or engine and the red to the other end of that stripped wire.

Lastly, the 9v comes even when the car is turned to run, not start. It stays the same when cranking but while testing that purple and white wire I'm realizing now that I had the relay out so I will check for voltage drop with it in.

 

[ah.b.normal]

I hate to mention it, but bad ground? (high resistance at ground point will test as less than supply voltage.) Re-check that purple/ white stripe wire using the battery negative terminal as your ground? These computers spew hate/confusion due to weak/dirty grounds!

 

[Gtstorey]

Don’t forget that the ECM is what actually controls cranking so the voltage will maintain for a short period of time after releasing the key. But if it has a voltage when just turning the ignition switch on then you probably have something else going on. Is the voltage showing the correct direction when measured? If not it could be a bad ground causing a backfeed to ground trough the ecm.
You need to check for voltage to ECM by backprobing while the connector is plugged in.

But does still sound like a missing or bad ground.