Diagnosing a Toyota's power problem

Of the 7,139 languages spoken worldwide, it never ceases to amaze me that “drivability” is one language I can speak everywhere I go.

Just recently I had an interesting but difficult conversation with Rayo, a bright young tech from Guadalajara, Mexico. My lack of fluent Spanish and his broken English made conversing a bit of a challenge. Rayo was faced with a strange concern. A 2015 Toyota Land Cruiser, with a 5.7L V8 engine was experiencing what he described as a strange loss of power output. According to Rayo, the engine began to struggle under acceleration between 4,000-5,000 rpm.

The issue seems straight forward but I was faced with educating the young tech on how to obtain the appropriate data (and in the appropriate fashion) to allow for analysis. He was using Toyota’s factory scan tool for this era of vehicle, the Toyota TechStream.

Preliminary Data

After further discussion, he described the loss of power output to be more of an “engine cut-out.” I first viewed the data he sent me to get an idea of what he saw (Figure 1). As you can see, I configured the data PIDs in a customized list from the left of the screen capture in graphical format to see how they correlated with one another. I refer to this as action/reaction comparative measure. I do this to only see the data I need to see. This tells a drivability story. These include:

MAF
RPM
Throttle position
Lambda 1
Lambda 2
Total Trim 1
Total trim 2

If one multiplies the engine displacement (in liters) by a factor of 37, it will reflect expected grams per second MAF values at 5,000 rpm/full-throttle. According to the data, the engine appears to be breathing correctly under heavy acceleration (219 grams per second). This is good news because from over 2,500 miles away, I can tell for sure that the exhaust isn’t restricted, and the engine has no timing issues.

I can also see that the fuel system is up for the task. The lambda values on both banks indicate the engine is fueled correctly (nearly 20% richer than stoichiometry) and the PCM can carry out that task with no corrective factor (as indicated by nearly zero fuel trim on either bank).

This portion of the same capture shows the PCM exhibiting erroneous data. A common sign of trouble with the PCM entering a stare of default or reset.

However, something strange occurs about 20 seconds afterward (Figure 2). Although the operating conditions remained unchanged, the data indicates the lambda values (both banks) exceed 1.23 (over 23% lean of stoichiometry). I would’ve expected the fuel trim to attempt compensation but as can be seen, they both remain at zero. Viewing the data from the MAF as well as engine speed shows no variation at all. Any time you see straight lines on graphed data, it typically indicates a state of default, as in, the data isn’t true.

Weighing the Possibilities

Just a moment past the cursor (Red arrows), the vehicle comes back to life and the data begins to update again. This is the symptom Rayo has been experiencing and at this point, I'm growing suspicious of a computer issue. I believe the PCM is resetting for some reason. A few common causes for this scenario are:

High-energy discharge from an inductive device (like brain damage)
Loss of PCM voltage, ground or ignition feed
Loss of 5V reference circuit or (internal PCM fault or external circuit fault)

My advice at this point was to determine which of the three bullet points above was occurring. I advised Rayo to first measure the CKP/CMP signals on a digital storage oscilloscope along with their reference voltage and ground supply circuits. Unfortunately all I can show you is the picture Rayo took with his cell phone (Figure 3).

This lab scope capture demonstrates the CKP and CMP sensor signals failing. Definitely a cause of a PCM reset. However, it’s the 5V reference circuit feeding the sensors that is the root cause fault. This must be pursued to locate the cause of the PCM reset concern.

According to the scope capture, both the CKP and the CMP signals fail at the time the vehicle’s symptom presents. However, the 5V reference voltage that feeds both sensors (allowing them to operate) is dropping out simultaneously.

The Data Doesn’t Lie

With all the information in front of us, and the desired information not yet obtained, we are faced with deciding how to proceed. Here are some bullet points of what we know to be factual, and I will ask all of you, diligent readers, for your input: 