Fixed! Chevrolet Lumina with loss of power on hills

Car model and year: 97 3.1 Lumina has 90K


a bit of a very slow coolant loss, and the pump was found leaking. Replacing that, about a month ago, it started having a loss of power between 20-40 mph. I took the car and found misfire codes on cylinders 2, 4, 5, and 6.

What I did:

I cleaned the MAS, replaced the spark plugs and wires, checked all the coil-packs with a spark tester (they checked out with strong spark), and checked the fuel pressure at the fuel tube that feeds the injectors . With that done and the problem persisting (especially when the car warmed up to operating temp), I had Cottman transmission (very reputable guys at this local shop) and they checked it out and found the same engine codes, but stated the transmission was working fine. During the next few weeks, another noted very slow coolant leak was detected without anything leaving the engine. During this time, possibly unrelated, I had problem starting the car and found the battery had developed a bad cell, but the alternator checked out, so I replaced the battery. Anyway, knowing the 3.1’s storied issue with the complicated factory head gaskets leaking coolant, I took the large task and replaced the head gaskets and had new valves and seats installed in the heads. it is much better, but still noted she felt there was still a lack of power on the hills.

Chevrolet Lumina scanner to use and test results:

Foxwell NT301scanner: not idea
I did get a Foxwell NT301 and saved some running data.

Here’s an example of one of the “printable” moments that the Foxwell recorded, there are well over 50 pages:
Calculated Load Value(%)14.9
Engine Coolant Temperature(øC)93
Short Term Fuel Trim -Bank 1(%)-6.3
Long Term Fuel Trim – Bank 1(%)-2.3
Intake Manifold Absolute Pressure(kPa)71.0
Engine RPM(rpm) 1893
Vehicle Speed Sensor(km/h)20
Ignition Timing Advanece for #1 Cylinder(ø)-1.5 Intake Air Temperature(øC)51
Air Flow Rate from Mass Air Flow Sensor(g/s)36.42 Absolute Throttle Position(%)43.9
Location of Oxygen SensorsB1S12–B2S—-
Oxygen Sensor Output Voltage Bank 1-Sensor 1(V)0.045 Short Term Fuel Trim Bank 1-Sensor 1(%)-3.9
Oxygen Sensor Output Voltage Bank 1-Sensor 2(V)0.105 Short Term Fuel Trim Bank 1-Sensor 2(%)N/A
OBD requirements to which vehicle is designedOBDII

Just test drove her again and got the error code:
Mass of Volume Air Flow A
Circuit Low

I replaced the MAF a few days ago. So I remeoved the new plastic one and replaced it with the original aluminum one (that’s clean as a whistle) and there was no change in the error code. Plus, I then tested it with my multi-meter, here are the results:
1) With the harness disconnected and the engine/ignition off:
All three wires (red, black and yellow) ohm-ed out as ground.
2) With the key in the ignition and it turned to run position without the engine on:
The power wire (red) had almost 13 volts, the ground (black) wire was grounded, and the signal wire had about 6 volts.
3) With the engine running and the harness attached:
I tested the Hertz with my TEST BENCH set to 20 Hertz and the signal wire (yellow) had a fairly consistent 2.14 hertz at idle (engine warm) and when the engine revved it smoothly went to just above 14 hertz and then evenly went back down to 2.14 when the idle was slowly released.

I ran the car with my Foxwell attached and the Foxwell has a (g/s) reading for the MAF during the events in real time. It seems it was registering 4-5 during idle and was at the 40-50 range when it was bogging down, but when it cleared and accelerated as expected, it was well into the high 60s and into the 70s.

With GM Tech 2
Finished (though I had to go to a shop with a Tech 2 and basically rent it with the mechanic).

Everything is factory spec, and it idles and revs without issue. No trouble codes. Everything is within spec. when we looked over all the info in his Tech 2 readings.

When driving (though markedly better and gets up to highway speed great), still it seems the engine, as it is accelerating: feels that it either adjusts the fuel/spark a bit, or the transmission shifts prematurely into a higher gear or something (that’s the only way I can describe it). It’s weird to feel the car do something without driver input. Whatever it is is keeps the car from continuing its acceleration for a few seconds. If I give it throttle during the event, it downshifts and accelerates great through the issue. If I don’t alter the accelerator, and keep it steady during the event, it waits a few seconds and then shifts to overdrive and feels normal.

Since I changed the fluid and filter when I first was alerted of the issue by my daughter during my maintenance work, and the old fluid was discolored (aged) and there was no metal dust or filings on the magnet, I’m wondering if I just need to allow her to drive it and change the fluid again in a few days and see if it flushes more of the old fluid that was trapped inside the torque converter?


So my friend has some awesome equipment to test Hz and other sensor feedback and we looked at the Knock Sensor and the O2 Sensors (much more sensitive than mine as well as had the exact ranges of signal that should be expected). The Knock Sensor and the O2 Sensors both looked slightly out of spec, enough to do a test drive with first the Knock Sensor disconnected and then a drive with both the Knock Sensor as well as the upstream O2 Sensor disconnected.

With the Knock Sensor disconnected the drag or delay in acceleration under load was gone, though it wasn’t completely smooth. With both disconnected, it drove really well and there was no drag or delay or alteration of power during driving. I replaced both the Knock Sensor as well as the upstream O2 Sensor and tested the car again. It drives as it should.

So, in the end, the car had these issues:
1) the #1 cylinder had a head gasket leak, fixed with head gaskets replaced
2) the MAF sensor was sending inaccurate info and had a damaged wire, replaced
3) the vacuum Transmission Modulator was broken, replaced
4) the upstream O2 Sensor was not within spec, replaced
5) the Knock Sensor was sending wildly inaccurate info, replaced

Interestingly, only item #2 above was determined by the PCM to be in error and would throw a Service Engine Soon light or throw a code, the PCM was silent on the other errors. I guess a ’97 GM PCM was not very sophisticated.

Hope it helps!

Be the first to comment

Leave a Reply

Your email address will not be published.