Case Study
Surging, loping idle in 91 Sable 3.8 V6. Cause: Engine Control Module
Symptoms
Loping idle in gear, got worse as engine warmed up. Usually okay from cold start.
Shifting could tend to be a little harsh at times.
Engine would usually stall after WOT test.
DTC: 326, EGR ckt voltage lower than expected
Possible causes based on input from various sources Vacuum leaks Vacuum steady at 18 – 19 inches Fuel pressure Disconnecting pres reg vac source had no effect MAF sensor Cleaned sensor wires with magic cleaner- no change. Sensor steady at .6 volts, sometimes at .9 vdc, confusing but in specified range TPS Resistance measurements smooth and steady. Powered output smooth and steady EGR system Unplugged vac source- no change
EGR valve operated with vac tester (Mityvac)
Idle deteriorated as expected when EGR valve operated manually
EGR reg valve coil measured 33 ohms; cleaned filter- no change
EGR valve position sensor output in spec with manually applied vac PCV Not checked Air Filter Fairly new AC load Not a factor Idle Air Controller Cleaned and coil checked- no change Injectors Not checked, power at speed okay Ignition Not checked Cam Not checked Timing Chain Not checked Head gasket Not checked ECM
Testing (my inexperienced dead ends are not included)
MAF sensor seemed to be in range based on DVM measurement, .6 to .9 vdc. However, unplugging MAF sensor connector allowed a stable idle, was drivable with occasional stumble but no stall at idle.
Engine generally idled okay until it was shifted out of Park, then it started loping. Going back into Park did not return to a better idle but continued loping.
Time for the scope
MAF signal- unsteady and heavy spikes. Average DC about .6 vdc.
The oscillations would continue even after the engine had stalled. If key was turned off and then restarted, the oscillations were gone until triggered again by shifting.
All accessible power and ground circuits were checked, cleaned, etc.
I unplugged any accessible connectors. Idle improved and heavy spikes disappeared when the Axode2 connector to the transmission was disconnected. The transmission pressure control solenoid (EPC) was included in that connector.
On cold start, MAF and EPC solenoid, idle okay, EPC switching spikes showing up on MAF signal.
I wondered about the EPC solenoid chattering or having a failed snubber, but the instability was too uniform. Applying an external snubber diode only made the oscillations worse. Seemed that the ECM was the primary candidate.
Opened up the ECM and found 3 bad electrolytic capacitors. After changing those, the car is back to its old self.
EPS and MAF signal, .68 vdc, small spikes probably normal?
One of the replaced ECM caps was part of the reference voltage supply circuit for the TPS and EGR position sensor, and probably for the MAF sensor (but I can’t confirm). That may be the cause of the 326 code.
Without the use of the Pico scope, the car may have ended up with a new MAF sensor, TPS, EVR, EGR position sensor, EGR valve (?), etc. As it turned out the repair costs for components was nil.