V1 Pulsating Rough Idle (Idle Hunting), Seeking Assistance Please.

[phantom3277]

Hello,

I bought a 2006 cobalt ss 2L LSJ for my donor, which original drove fine, but would stall on idle.

I’m going to provide as much info as possible as I feel like I’ve done a fair bit of due diligence on this issue(s).

ZZP tuned ECU
2.8 Pulley

Donor idle when received (drove fine):

Donor idle before getting more violent

drive.google.com

Driving donor back went into limp mode for 5 minutes with P2138 code. APP1 and 2 Voltage discrepancy.

During Testing some electrical changes, the timing chain became quite loud (later found broken guide and broken primary tensioner):

Timing Chain Noise and Idle

drive.google.com

Replaced the Pedal and got no change. During testing the idle got worse, with the timing chain visibly loud and the idle getting worse I didn’t want to risk it and stopped testing. Decided to strip the donor and fix in the process.

During the build process, I:

- Replaced throttle body
- Replaced MAF
- Replaced Oil
- Replaced Primary Tensioner and Broken Guide
- Ran 6 new wires from ECM to pedal jumper
- Tested a different Pedal Jumper

When I started the car (now open headers by the way) the idle still was bad (up to before the goblin side of things I have a pretty detailed log of what I did in the attached files). I actually replaced the throttle body at this point and I regained pedal control during engine operation strangely (lost that right before stripping), and the idle did better than the other throttle body, but still was idle hunting and sucked:

New Throttle Body Test

drive.google.com

Here’s the original throttle body. You can hear even after shutting it off it shutters due to the throttle body valve oscillating so much during idle and doing it a time or more before it dies:

Poor Idle (Initial Goblin Startup)

drive.google.com

Here’s my thoughts:

- I need to replace the entire timing chain system and balance shaft system. Do you think it skipped a gear? Could this be the issue causing such a rough idle? FYI sadly so not have photos of chain at TDC, but they seemed approx 10 and 2 like the cloyes video said for the intake/exhaust marking.

- Directly wire the APP2 Sensor plug to the ECM with new wires. Maybe this part was damaged?

- The check engine light is still on, but no power to OBD2 plug so I need to fix that before I can confirm the code is the same (assuming it is or a similar). I think maybe I messed up the low speed cluster green wires for it. I need to dig into this as well.

- Bad tune or messed up ECU? It did get worse when I ran the one wiring test and originally thought I screwed it up more (see diagnostic log), but I’ve come to think it might’ve been the timing chain’s poor coincidence giving out. ZZP’s info on the tune matches up with what I see on the engine unless I overlooked something, but they also said my issue is always the throttle body, pedal, or wires. And I’ve tried all three. My other pedal was in rough shape that I tried (exposed to some water/elements), but it didn’t seem to change the result and appeared functional to the same degree as the other.

- Vacuum leak? I’ve attached a bunch of photos in case there’s something wrong, the previous owner had some interesting routing and perception on stuff in the not so good way, so maybe something is there. But the error code seems to point electrically?

Any and all help would be appreciated. I’m not unfamiliar to cars, but powertrain and electrical is not my primary knowledge in automotive (suspension and vehicle dynamics) and I’m still learning in these other areas.

 

[Ross]

You have a lot going on here.
I would check for vacuum leaks.
You need to get the OBDII port working, so you can check the current diagnostic trouble codes.
Checking the OBDII port for electrical issues.

 

[Rttoys]

Definitely need the DLC to do its thing so you can get data.

Other than that pedal code, I'm thinking a vac leak (edit) or whatever I have circled in the picture. One of our Houston guys had the same surge. after exhausting a lot efforts, I pinched off the brake booster vac line and it instantly ran normal. The booster uses a lot of vacuum, so it becomes a big problem when it goes bad. I’d start there and use the old school trick of carb or brake clean to spray on the intake gaskets and such to find leaks.

(Edit) Back to the pic. What’s going on here with the 2 map sensors?

 

[Desert Sasqwatch]

Why does it look like a missing 'something' in the ZZP boost controller, just above the top yellow circle from @Rttoys? There is a gasket(?) with a center port and 2 open screw holes. Does this get a fitting and vacuum line or is this just something that fell in there from somewhere else? As he pointed out the sensor that is plugged in in the lower circle is the 3 wire MAP, it has the IAT signal for the PCM - does this one go where the uncircled MAP 2 wire sensor is in the supercharger housing or someplace further upstream in the intake tubing?

 

[phantom3277]

Thank you for your replies.

I'll hopefully be back with the car on Friday, and will run some vacuum tests then. It's not possible a skipped tooth could cause an uneven vacuum from the timing chain being loose? When I pull out the entire timing chain and balance shaft chain (when I have time), I will verify all marks, but just thinking it through. The pedal issue seems it could just be its own thing based on these responses, but does a vacuum issue just suddenly go from idles semi conservatively rpm hunting to violent from one crank to the next - and would I not see codes for it?

To answer the questions and responses:

@Ross Very clever using a smoke machine to find leaks, I had not thought of that. ThiI will refer to your linked info for the OBD as well.

@Rttoys According to ZZP, that bottom yellow circle is a barometric pressure sensor and should be open to environment (LSJ Cobalt Sensors | Locations & Function - YouTube ). Agreed on the DLC, that's a priority, it's a shame there's no interface on the dash to show codes as a backup. The top circle is from the previous owner, I believe he just was using a sensor to plug up a T fitting since it seems he didn't have a 180 fitting (strange choice not sure). I have not tried checking the brake booster and will plug up and see if idle improves.

@Desert Sasqwatch The extra Evap Solenoid is what I believe you're referring to. I've removed the extra one (previous donor owner was too lazy to unscrew the bolt to give the new one a mount apparently). My goblin doesn't have EVAP I believe so the other side with two open ports has no hose or plug at the moment, but the donor ran the same as the goblin before stripping, so I am assuming it's the same. I believe my answer to Rttoys should answer the other questions, unless I'm not grasping an aspect of what you're saying.

 

[Rauq]

I'm unclear on where you're at with DTC's post-Goblin build. I'll circle back to that.

As others have said, make sure you don't have any vacuum leaks. Off the top of my head, not having EVAP means there will be a port on the throttle body that needs capping. I see three pressure sensors in your pic- one installed in the supercharger (SCIP), one tee'd into a line that I hope goes to the manifold reference barb below the fuel rail (MAP), and one that might be BARO? The sensor that's supposed to be in the manifold next to the coolant temp sensor also has IAT in it, so it needs to be in the manifold. I can't imagine why it would not be, unless there's a separate IAT sensor wired in somewhere, in which case, we're all shooting in the dark. In any case, I wonder if the spot where the MAP is supposed to go is also a vacuum leak now.

One thing I haven't seen folks mention much is, where's the O2 sensor and where's it at? Even if it's in good condition, if it's 2" from the end of open headers, it's probably not going to get a quality reading, and then an intake with no vacuum leaks or wiring issues will still hunt for idle.

On DTCs, if you're confident there's no vacuum leaks or sensor issues, diagnostics will be limited without being able to scan for codes or to datalog for sensor issues. Peeping codes or datalogs might even help you figure out whether you need to be looking at vacuum leaks or O2 sensor issues or anything else.