When Pinouts Pan Out

Jan. 1, 2020
The 2005 Chevy Cobalt that came to my shop on a car dolly had been sitting fallow for a couple years. It showed just more than 48,000 miles on the odometer when the owner left it with us, and we pushed it into the shop.

Pins can be mightier than words.

Motor Age GaragepinCobaltVicproblemfixing vehiclerepair shop trainingtechnician trainingautomotiveaftermarketMy uncle had a 1950 Model Ford that he had restored, but it sat for decades under a shed behind his house. When he decided to get it going again, just about everything was seized, dry rotted, locked up and deflated. That's what the Second Law of Thermodynamics does to machinery. Over time, anything – even a brick – that is made by man will slowly disintegrate into the elements from which it was assembled. We've all heard stories about brand new cars that, through whatever circumstance, were left sitting in nondescript warehouses or in boxcars on a secluded piece of railroad track, and if by some miracle they're still operative, we see that as the exception rather than the rule.
You might have read about the 1957 Plymouth Belvedere that was wrapped in a protective cosmoline-like sheath and buried in a big concrete vault back in 1957 so the lucky winner of the car could see in 2007 what things were like in the 1950s. The winner of the buried car would have several other prizes to look forward to besides the car. There was a $100 trust fund accruing interest until the year 2007, a 5 gallon can of gasoline and a jar of Oklahoma crude oil. Then in the glove box, there were 14 bobby pins, a ladies compact, a plastic rain cap, several combs, a tube of lipstick, pack of gum, facial tissues, $2.73 in bills and coins and a pack of cigarettes with matches — all items that might have been found in a woman's purse circa 1957. When the protective wrap was removed and the car was revealed, it had been submerged in water that had invaded its vault, and the whole thing was a putrid mess.
The 2005 Chevy Cobalt that came to my shop on a car dolly had only been sitting fallow for a couple of years, so it hadn't deteriorated quite that badly. It showed just more than 48,000 miles on the odometer when the owner left it with us, and we pushed it into the shop. The car originally had been dolled up and featured in Street Trenz magazine, but now the fancy wheels had been replaced with black-rimmed dry-rotted tires that were too small for the car, and there was some minor rat damage to the headlight wiring. Was there rodent damage anywhere else? It didn't look like it, but it's had to be sure.

Somebody said it needed a fuel pump, and it wouldn't be surprising if it did. Rotten fuel can do a lot of damage to a vehicle's fuel system, and whatever gas had been in the tank when this one was parked had a lot of time to rot. Oddly enough, a fuel sample we retrieved from the tank didn't smell bad at all!

First we needed to see if the fuel pump would even try to run, and we found that it wouldn't. We could reach for the scan tool, and a lot of guys will do that. But what I like to do first in a situation like this is to find the relay, yank it out, and then locate the pin that feeds power to the pump and check that pin for a ground, which should be present at that pin if the pump circuit is good. If there's no ground at that cavity, the pump is suspect.
Kick the gas tank and see if the light comes on. If there is a ground at the normally open relay pin and the fuel pressure rises when power is fed to that pin in the relay socket, then the relay is suspect, but make sure the common terminal and one of the coil terminals in that socket both have battery voltage with the key on. The PCM typically supplies the ground to the other coil terminal. If everything is there, the relay is bad. This works well if the relay is easily accessible. Admittedly, some Asian makes have their fuel pump relay integral to the EFI main relay and when they do, that relay can be sequestered away in an unsavory spot, and other measures will be in order.

On the Cobalt, the fuel pump relay is in the PDC. What came to light was that the person who had checked the car previously had mistakenly plugged the relay in so that two of its terminals were in cavities without terminals. This isn't ordinarily possible, but on the Cobalt's PDC, it's an easy mistake to make.

Plugging the relay into the right cavities was a peachy fix, but we weren't done, not by a long shot. When we connected the scan tool, we had no communication of any kind. The security light was on, and the little Chevy still wouldn't even spin over. Network diagnostics and a deeper search for rat damage would come next, but other jobs sidetracked us, and the Cobalt went on the back burner.

Air Bag Lights

Today's vehicles have smart forward airbags that deploy with less force when the crash isn't so bone-jarring. My 2007 F150 knows how much weight is in the passenger seat and turns the airbag off automatically when it believes someone lighter than a full grown person is sitting there. But that feature is so routinely ubiquitous on pickups that it's kind of boring even talking about it nowadays.
Some platforms have airbags in every pillar and all around the perimeter of the car's headliner. There are inflatable knee bolsters, but one hidden element of the airbag system most folks don't think about are those pretensioners that are built into the seat belts to lock the reel and/or tighten the belt a bit. They don't move much, so they aren't as dangerous or as scary as the airbags that can come blasting out of the steering wheel to punch you in the face like a giant boxing glove.
This 2006 Ford Crown Victoria was flashing a code 33 at key on, which directs the troubleshooter to check the driver side pretensioner (a passenger side failure would flash 34). The scan tool tossed us a B2292, which can point to either the driver or passenger pretensioners, which is quite interesting – usually the lamp flashout code isn't as accurate as the Bxxxx-code.

The pretensioner is supposed to lock the inertial reel and tighten the belt a bit, so it has a squib that is the baby brother of the explosive device used in the driver and passenger airbags, and it doesn't make much of a fuss, even when it lights off. Because I had almost nothing to lose (this was one of our campus cars), I committed the cardinal sin where airbag systems are concerned, but without negative consequences.

Disconnecting the Restraints Control Module (RCM), I found the pins that fed each of the two pretensioners, and then I measured the resistance of the pretensioner squibs right there at the airbag module connector with a digital multimeter. The worst-case scenario would be that I'd light off a seat belt squib and have to replace the assembly. The driver pretensioner read infinite resistance (not good) and the passenger squib read 2 ohms. When I disconnected the driver side pretensioner, examined the terminals in the connectors and jumpered the wires together, I did another measurement at the module connector – that time I was reading a dead short, so the intervening wiring was intact.

Let me say that I wouldn't have measured any of the other airbags like this, because there's so much at stake. The right way to troubleshoot is to insert a specially packaged 2 ohm resistor (airbag simulator) in the place of every airbag. That fools the module into thinking the airbags are present and removes the danger. Further, those steering wheel, curtain, seat and dash panel airbags are both too expensive and too dangerous to risk. They could, if deployed at the wrong moment, literally change your life forever or end it in short order.

I will also say, however, that, prior to this squib measuring exercise, I had measured the resistance of a driver airbag I was planning to deploy anyway. But I put the airbag in a large open space outside the shop and used about forty feet of speaker wire to keep my distance in case the bag lit off. I got 2 ohms of resistance and no explosion measuring the resistance. It was a grand and glorious exercise. But when I fed current from a battery down those same speaker wires from a battery, the bag deployed with explosive fury, jumped about ten feet into the air with a concussion that would put a twelve gauge shotgun blast to shame, and filled the service lot with smoke.

When the driver side pretensioner was replaced the airbag light blinked rapidly for a few seconds and everything returned to normal. Now it was time to re-engage that interesting little Cobalt.

Fire in the Hole!

The next time I reached in the window and turned the key on the Cobalt, the security light went off and the little car spun and cranked like it was brand new. We had good communication from everything except the Transmission Control Module and the Power Steering Control Module. The electric power steering was dead in the water and there was a message to that effect displayed under the tachometer. As for the TCM, the transmission wouldn't shift gears and the speedometer was a flatliner. Looking at the datastream we were able to access, we found that the Engine Controller was showing some fuel trim issues that turned out to be a problem with some split hoses – nothing surprising or mysterious about that.

Occasionally, even while the engine was running, the security light would engage and the communication would go away. When the security light was on, you couldn't even switch the car off with the key (it would simply keep running), and if the light came on and stayed on when the car was already off, there was no starting it either.

When the girl who would be driving the car asked about that come-and-go security problem, I postulated (off the top of my head) that the time in the barn might have caused issues that would go away once electrons had flowed through the PCM's brain for awhile. It was the best spin I could put on a dicey situation. While I didn't believe what I was telling her, it didn't cost anything to spew a little sunshine her way. But to my surprise, the longer the Cobalt ran, the more stable the network became to the point of normalcy, and at the time of this writing, that car has been in the wind for weeks with no network issues at all. Was I just lucky? Let's hope not.

The two consistent wrinkles we had while the car was in the shop were the TCM and the Power Steering issue. Well, it turned out there was a recall on the Power Steering, so we'd defer to the dealer on that one.

Figuring the transmission electronics might be a pretty easy troubleshoot, we disconnected the TCM's 49 pin shell and used the pinout chart to find the two wires that were supposed to lead to the speed sensor, which is hard wired into the TCM. The sensor was there and producing a signal, so we checked for power and ground. It was there, so we obtained a $100 used Transmission Control Module, producing datastream, speedo operation and, most importantly, transmission shifts.

The brake rotors were dreadfully rusty, and so before we turned the car over to the dealer, we snatched the rotors and machined them on the lathe, dressing the nearly new brake pads with a drill-driven rotor refinishing pad for re-seating.

After the Cobalt was taken to the dealership for power steering repairs everything was in order, and as stated earlier, she's been running the wheels off of that little car ever since.

SIDEBAR Don't Forget the Pin Numbers

When checking for hard fault problems in just about any system, it pays to find the connector pinout on the relay, inline or module connector, identify the pertinent pins and check with the appropriate tools to see what's available there. Further, it's a good idea to make sure there are no pushed back or damaged pins.

One GM instructor said many of the connectors on new GM cars won't survive being disconnected and reconnected more than about four times. Just remember when measuring that a DVOM won't tell you if enough current is available at that pin to pull a load, and so low impedance test lights still are a valuable tool to keep on hand.

If you find have good strong power and ground at the proper pins, and the right resistances can be measured at the input and output cavities, most of the electrical troubleshooting work is done.

2006 Crown Victoria 104,254 miles 4.6L Engine 4R70W Transmission Air Bag Light Flashing/Illuminated

Voice Your Opinion!

To join the conversation, and become an exclusive member of Vehicle Service Pros, create an account today!