"Ding-Dong!"

Put on your detective cap on. Here is a strange one.

"I had to have my Jeep Cherokee jump-started as it sat in my garage. I took it to a service center and after checking out the electrical system, they determined the battery was defective. I bought a new battery."

 

The next night it was fine but two days later the battery was dead. I checked it out and noticed that the instrument cluster lights would stay on intermittently. I mentioned it to him but did not have time to work any more on the vehicle. He took it to the dealer.

 

The next day he called me again--dead battery. When I got to his place I checked the repair order. They replaced the Radio Frequency Hub Module and ordered 2 new key fobs plus a Common Body Controller. I noticed that the instrument cluster lights were on. I backed the Cherokee outside in the drive to have a little more room to work under the dash and noticed that the lights were out and didn't turn on.

 

I pulled the vehicle back into the garage and the lights would intermittently turn on.

 

What do you think? Probably the same thing I did. I went home, got my handheld spectrum analyzer (essential tool these days in the world of RF we live in) and found a 433 MHz signal in his garage.

 

Guess what frequency his system uses? 433.92 MHz!

 

"So, did you install anything wireless lately?" Was my question of choice.

 

"No--not lately," was the response.

 

"Well, what wireless systems do you have in your house?" I asked.

 

After going through his list, I found the culprit--a finicky doorbell button. For cost reasons, manufacturers use the 300 MHz - 433 MHz bands for their doorbell frequency range. In other words, he bought on the cheap.

 

"I've got to run. I hear the doorbell ringing."

 

Jack McGinnis

HiTech Investigations

"You've got to be kidding," Tom said as we stood over the engine compartment, staring down at a 2011 Duramax diesel engine, "the serpentine belt is causing the transmission not to shift properly?" He was referring to a truck that was towed in for a transmission issue. Being the great technician that he is, Tom verified  the concern. Step one in good investigative work.
      Step two involves a good visual inspection to look for the obvious: rodent-chewed wires (Honda makes a rodent-proof electrical tape), add-on accessories, disconnected/damaged connectors--you get the idea.
     "This vehicle went to an upfitter and had an aftermarket clutch pump installed, and the serpentine belt was replaced to accommodate the extra length," he explained.
     "Ahah!" We did some research--step three--and found a document from the manufacturer, explaining how certain aftermarket serpentine belts can generate copious amounts of electrostatic charge on their surface. This field can discharge and affect the transmission control unit which is mounted near the belt. According to the manufacturer, the TCM needs to be moved away from the voltage-inducing belt and fastened to the top of the underhood bussed electrical center (what I used to call the fuse box).
     "Unfortunately, the wiring harness isn't long enough to allow me to do that," Tom remarked. I can move it, but how much is enough? Is there any way that I can measure the charged field?"
     "You could make an electronic electroscope," I suggested.
     "I will be forever grateful if you explain that one," Tom came back.



 

 

     "This little device, consisting of a 9V battery, a clip, a Field-Effect Transistor and LED, can detect one volt and with ridiculously high sensitivity. You may want to install a 1 Meg resistor in the gate leg to protect it from voltage spikes."

     Tom determined that he did not move the transmission computer far enough away from the voltage-inducing belt. He finally fabricated an extension harness, mounted the transmission controller on the fuse box per manufacturer recommendation, and the problem was solved.

     "Wow! A serpentine belt stopped the truck from moving! I guess it pays to be thorough in your investigation, eh?"

     "Always," I responded.

     He looked at the small electronic electroscope. "Neat device," he said, "what else can it be used for?"

     "Field distortions," I said.

     "Meaning?"

     "Ghost hunting."

 

Jack McGinnis, HiTech Investigations Inc., signing off.



Get your notepad and smartphone out and check out a 1993 Cadillac Allante with Jack McGinnis of HiTech Investigations.

"Yep, they're still out there and the service technician is expected to diagnose and repair the old ones as well as the new. It doesn't matter--the diagnostic process is still the same," Jack McGinnis told me.
The owner of the Allante (actually, he owns one and his better half does, too) bought the Allante and a great price--unfortunately, someone hacked the electrical system, turning it into a tangle of useless wiring. The shop owner that worked on it, "J", did a fabulous job of retracing every wire and after a fair amount of time invested, repaired it--except for one annoying problem.

J: "It keeps flagging a trouble code--a P080--meaning the Throttle Position Sensor Learn hasn't happened. I use the on-board diagnostics to perform the relearn, the vehicle computer responds by saying the Learn Process is complete, but when I cycle the ignition key the code resets, saying it hasn't been learned."

Me: "The TP sensor is okay?"
J: "Yep--I checked the wiring and the sensor with an oscilloscope."
Me: "No add-on systems that shouldn't be on the car?"
J: "Nope."
Me: "did you try disconnecting the battery?"
J: "Yeah--I even shorted the battery cables together after disconnecting the battery."
Me: "Did you do your research?"
J: "Yeah--no technical service bulletins, no answers on the forums, no search engine results."
Me: "Sounds like you are thorough," McGinnis said, walking away as he looks through his note pad. "Oh--just one more thing, sir--did you check the engine ground strap?"
J: "I will. Hey--the engine ground is bad. Let me try using a jumper wire and then perform the TP Learn."
Me: "Did you try it?"
J: "Yeah--it's fixed! Why did a bad engine ground cause that to happen?"
Me: "I suppose that then the module does a TP Learn, it sweeps the TP sensor to reference low based upon ground. So, if the engine ground is missing, then the PCM uses a perceived ground and learns the TP value based on the perceived ground. Unfortunately, it discards the values at the next key-up because it has nothing to compare ground to. Perceived ground is not a memory ground."
J: I'm just glad to get the car out of my hair."
Me: "Yeah--me, too. Miller time?"
J: "You bet!"

Always remember to check those engine grounds.

The real Jack McGinnis resides in a world of technical data, of analyzing data packets and, as an automotive service technician, must have great detective skills. When someone has an issue with their vehicle, you must be thorough in asking questions (can you spell C-o-l-u-m-b-o?). Questions like, "Did anyone work on it before?" Or, "Are there any add-on devices, like remote start, or a GPS, or an insurance tracking device?"
"Oh--and, one more thing, sir--how long has it been acting up?" A good investigator needs to jot down all the observations so as not to forget any important details.

I recently worked on a 2017 truck with a problem in the charging system. The alternator, battery, battery cables, starter and powertrain controller were all replaced. I checked the system for trouble codes--none. I did a visual inspection--nothing unusual. Look for clues, right? The engine computer said the battery was charging fine. A digital meter at the battery told be the computer was lying to me. Why does a computer lie? It doesn't. Garbage in = garbage out, right?

I asked a very important question: Was there any other work performed on the truck?

A: "Yeah--the radio was replaced."
Me: "A new radio replaced? Why?"
A: "Because it quit working one day."

I used my inspection camera and snaked it up under the dash, looking for more clues. I found an extra ground wire attached to the radio. The original ground was bad. The technician working on the truck probably replaced the radio and when it didn't worked, checked and found a bad ground. Rather than find where the ground was bad, the technician ran a separate ground. Now there were two grounds at different potentials. I cut the original ground and the battery started charging.

With enough electrical noise, the engine Body controller "told" the engine computer that the system was charging normally, so the engine computer backed off the charge rate.

It's all about detective work. What would have happened if I didn't ask about the other work performed? Probably an unsolvable case.