1G 1g DSM should ECU case be grounded?

[Badge023]

Been digging through wiring diagrams and was having problem with TPS throwing Code 14. I dug through diagrams and saw that on page 33 or so that the TPS ground reports to terminal 24 on the ECU. If the ECU itself is not grounded, will the ECU report the ground.

IIRC the ECU is not mounted and just laying in the compartment under the radio. Does the ECU need to be mounted to the body to report the completed circuit? Specifically Pin 24?. I'm currently replacing all the connectors that are seem to be made out of Doritos to get good contacts.
.

 

[97egl]

The ECU is mounted to the steel radio frame by 3 screws and the radio frame is mounted to the body so if it needed to be grounded it would be. But I have ran multiple cars with the ECU laying on the passenger floor boards while working on them and doing diagnostics and its never been an issue so I do not believe they need to be grounded.

 

[19Eclipse90]

There are multiple grounds out of the ECU harness that connect to the frame. As above, the ECU case does not need to be grounded.

 

[steve]

Like Brian says, the case doesn't need to be grounded, it doesn't hurt but it's not required since the wiring harness provides the ground connections.

Fault code 14 can be from several things. The ECU checks to make sure that the TPS voltage is greater than 0.4 something volts when the throttle is closed (Idle Position Switch closed). That the TPS voltage isn't some large value when the throttle is closed. That the change of the TPS voltage doesn't have large spikes or dropouts in normal use.

If your having other faults with analog sensors it's possible that the ECU's internal sensor ground is damaged. We used to see this a lot when people messed up using a universal O2 sensor and put 12v onto the sensor ground wire.

If your ECU hasn't had it's capacitors changed (or whoever did it didn't do it right, common with autopart store replacements) their can be corrosion damage to the ECU's internals that can also cause various fault codes.

 

[Badge023]

Thanks for all the feedback!
On the scope closed throttle voltage is .63 which from what I've been reading is acceptable. New quality TPS sensor, so the signal is clean on the scope. Idle switch is adjusted correctly. Going to test according to the FSM (attached) and see what I get after soldering in the new connector. The original was crumbling away.

I figured I would ask after seeing the ground symbol on pin 24 and kind of remembering the ECU was just dangling in the area under the radio. I'll pop the cover on it and take a look at the PCB and hope the previous owner didn't let the smoke out of the box. Code 14 is the only code I'm getting consistently. I fixed the CTS code with a new connector.

The car came with two other "spare" ECUs. Any recommendations on a good rebuilder if I need one?

 

[steve]

The car came with two other "spare" ECUs. Any recommendations on a good rebuilder if I need one?

The only ones left that I trust is ECMTuning, the same guys that make ECMLink.

 

[Badge023]

Thanks Steve! They were actually the one that came to mind since they seem to be predominate in the DSM world.

 

[Badge023]

Well, I found the answer to my question. I was seeing a nice TPS sweep on the scope. The problem is that it isn't making it's way back to the ECU.

I replaced the connector on the TPS, even ordered up the right color wires to keep aesthetics, next step was to check continuity back to the ECU. So apparently someone cut the green/white signal wire because it seems to be broken somewhere in the harness. It looks like they tried to bypass the break with black primary wire.

The same random ass black primary wire that I found dangling in the engine compartment not connected to anything. At least now I know where the mystery wire was connected. I still have another mystery wire that looks like speaker wire running across the firewall, but I think that was hooked to an old car alarm. I've never sunk this much money into a car that has the book value of a coffee and a doughnut. Sentiment...

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[Badge023]

It was more trouble to remove the extra wire, than to fix the broken one. All codes gone, purring at 750 rpm. Throttle body rebuild worth every cent. CAS sensor connector, injector connectors and set the timing. I think I’m done under the hood for a bit. Just in time to focus on the rat nest under the dash of old car alarm wires, relays and half ass splices.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[steve]

Loosen up that throttle cable, there is supposed to be a little slack (a couple of mm) in the cable when the throttle is closed. Having it tight like this has caused high idle issues in the past.

 

[Badge023]

Will do, Steve. Thanks! Also, a place named LSWiring is a great source for the right colored wires.