Non Valkyrie -- Chevy Van Air Bag Fault Light

[vanagon40]

I have a 2015 Chevrolet Express 3500 Van. The Air Bag light came on. The scanner shows a malfunction of the front impact sensor. I tried disconnect and reconnect, but no luck. So I bought a replacement sensor--genuine GM part at AutoZone for $240. When I got the sensor, I unhooked the old sensor and connected the new sensor and the fault cleared. But at the time, I needed a special socket to physically remove the old sensor and install the new one, so I reattached the old sensor and put the new one back in the box (I had held the new sensor in place with a small clamp).

So yesterday the socket finally arrived. I removed the old sensor and installed the new sensor. BUT, I still have the same fault code. It will not clear with the scanner. I tried some contact cleaner and reconnecting, but no luck. I tried leaving the battery disconnected for over 12 hours, still no luck.

Any idea why the new sensor worked two weeks ago, but not today?

[GiG]

First idea:
The sensors are Piezoelectric materials. They are subject to electro magnetic disruptions (static discharge). Maybe the new sensor was damaged by ESD or pressure from having been clamped in place. Retailer will deny refund, but you can bench test the sensor to find if it works.  *
Maybe get another from a junk yard...
Good luck, let us know what you find 

 *Testing ProceduresVoltage Generation Check:
Connect the piezo element - sensor leads to a multimeter (DC voltage mode) and apply physical pressure or taps to verify output generation (typically 1V to 4V depending on force).. Use rubber mallet.

[Pluggy]

Sometimes a diagnostic system can show a "false failure". Consider disconnecting the battery and waiting a few minutes. Reconnect and the diagnostic system will take another look at the sensor. Doesn't cost anything to try it.

[Ken aka Oil Burner]

Be VERY careful testing any supplemental restraint system with a digital multimeter. The internal 9v battery can set off airbags; seen it in person at the shop I used to work at. A guy was under the dash testing things at the airbag control module and set off the driver's, passenger's and side curtain airbags with his meter leads. VERY expensive oops. I'm pretty sure he lost some hearing that day, too. It was damn loud to me, and I was 6 bays away.

Also, most (not all) SRS connectors are built with grounding lugs that apply whenever a connector is disconnected, to avoid the possibility of static charge sending voltage down the circuit inadvertently.

Look in the connector housing of the new sensor and make sure none of the pins got bent over from misalignment when plugging in the second time. Sounds simple, but seen that also. It doesn't take a whole lot of force to bend over a thin terminal pin. And I agree with GiG about the clamp. Depending on the force, you could have internally damaged it by clamping it in place, even for the short time you did.

Actual airbag diagnostic tools use known resistance simulators that plug in to connectors to take place of sensors, seat belt pretensioners, etc., to isolate problems to the component vs the circuit/module. They do this because using a multimeter is too risky, as mentioned before.

[GiG]

Good advice, Ken!   
It's been wet & humid lately in Indiana, so an ESD event might be slightly less likely now...
Tends a little more towards the clamping scenario because the DTC
Says "internal damage"   
 Vanagon40 will be completely safe performing the unpowered bench test I outlined, no explosions possible...

[Ken aka Oil Burner]

Correct. Testing the sensor itself, while disconnected from the system: 

Testing harness resistance back to SRS ECU with meter: potential no bueno...