Bouncing MAF Counts

[SecondhandSnake]

I've seen this posted a few times but haven't seen any resolutions.

Reading the datalogs, it shows maf_cts bouncing from 0-770, both engine on and engine off. At idle maf_v is between 0.5 and 1.3V, keeping in mind engine RPM is varying about 200 RPM. Engine off, maf_v is bouncing between 0.0 and 1.0V. The fact it is bouncing with the engine off tells me turbulence isn't the root cause. No fault codes.

I had checked the MAF wiring before when it didn't want to start (due to a bad cal.) I had 5V power, next to no resistance on the ground, reasonable resistance on the PCM ground, and would see in range voltage when you blow on the MAF. I'm guessing my DVM also didn't sample fast enough to see the constant fluctuation. I'm going to re-verify with the DVM, test with a different MAF, and test with MAF disconnected.

Hardware: HPX-E MAF (brand new), 3.5" housing, GT PCM, Xcal3 (everything logged on device then extracted), Livelink II. All firmware/software is updated to the latest version.

Any ideas?

 

[MalcolmV8]

I've had brand new MAF sensors that were bad and did weird things. Don't rule that out just because its brand new. I'd swap sensors if all wiring checks out and no loose/bad connections. I see minor fluctuation with key on/engine off presumably because that's not where the sensor was designed to work but not 0 to 1 volt.

 

[bigmoose]

I wouldn't rule out turbulence. The maf uses a hot wire system that determines airflow by the resistance of a heated wire. Therefore, when the engine is off I would expect to see maf readings change as the wire heats and then cools with convection. Whether a 1 volt swing is excessive I don't know.

 

[MalcolmV8]

I wouldn't rule out turbulence. The maf uses a hot wire system that determines airflow by the resistance of a heated wire. Therefore, when the engine is off I would expect to see maf readings change as the wire heats and then cools with convection. Whether a 1 volt swing is excessive I don't know.

True but when you do a key on engine off and data log it's minimal. Like 0.1 ~ 0.2 volt fluctuation as I recall.

OP what does your intake look like? Any pics of the setup and placement of the MAF?

 

[SecondhandSnake]

Over the weekend I tried several configurations. I updated software/firmware, tried different tunes, different MAFs, and even with the MAF unplugged. No matter what the maf_cts showed as bouncing between 0-770cts.

With an updated tune it idles fine. Didn't get a chance to datalog that yet, but if it can hold a steady idle with a slot MAF in a 3.5" housing with less than ideal geometry (merge pipe ~6" upstream of sensor) then variation cannot be that bad.

TL;DR: Weird bug, no associated variable, doesn't seem to impact running.

 

[MalcolmV8]

If MAF counts were bouncing between 0 and 770 counts the car wouldn't even run. Are you sure you're data logging the right thing? I've had a car with turbulence I was tuning a while ago that was bouncing between 155 and 195 or so at idle and it made the car sound like it was heavily cammed and surged like crazy. Drivability sucked at low loads too. We're only talking about 40 counts bouncing here, you're saying 770.

To fix the turbulence I put on an aluminum screen like this before the MAF sensor.

That made a huge difference. My data logs went from this

To this

 

[SecondhandSnake]

Yeah I agree; with the MAF configuration I'm using if the ECM actually believed that it would be jumping between no flow and around 96 lb/hr of air. No way anything would run with that kind of data.

I am logging MAF counts (maf_cts), not the 03-04 Cobra MAF counts since I have a GT with a GT PCM and wiring harness.

I was honestly contemplating a flow screen since previously the idle would vary 200 RPM, but I think that can be chalked up to the tune being too lean down low, leading to lots of misfire and stumbling. Now it's rock steady, which I'm surprised at with the large housing diameter and poor geometry.

If nothing else I wanted this information out there for when some other poor soul is seeing this and tearing their hair out.