Your diagnostic process

Hey guys. I've always kind of been all over the place in my diagnostic process and this year I'd like to add in an oscilloscope and with that I need to tidy up my process a bit.

I know the questions to ask when diagnosing but after that, where do you go? I gather my best thoughts on the issue and then plug into DLC. I read codes and see if anything stands out to me and. Clear codes and see what comes back (unless it's a pressing matter, then I get to checking). Voltages are my main go to first. A quick search online of common issues similar to mine and then... I "guess". They're good educated guesses and I don't throw parts at any issues, but the removal of parts to check is tiresome. I feel I waste a lot of time between looking for common issues and checking parts with my DMM whether it be benchtop or in car. I nail it eventually.

I'd like to get better at my trade and take a more diagnostic approach than my thrown together, homemade one.

Live2Fish88 wrote: I read codes and see if anything stands out to me and. Clear codes and see what comes back

Don't be so quick to clear codes. Clearing codes clears a lot more than just codes. It clears Freeze Frame data and Mode $06 data, too. Both of these can be very useful to gain direction.

The first thing I do is get an understanding of what I am dealing with. Codes are a good first source of direction (if there are any). Second, Freeze Frame data will help you determine under what conditions the code was set. It helps to know if the error occurred on start-up or, if it was a hot engine cruising down the road. Unless I am, totally, familiar with what a code is, and what it means for THIS vehicle, my next stop is Service Information. What is this code telling me? When does the test run? Under what condition(s) is the code set? What happens when the test fails and the code is set? Can you reproduce the fault? Is it intermittent? Does a test of the senses (Sight, sound, smell, touch, and sometimes taste :woohoo: :whistle:) reveal any irregularities? All of this information will help guide you to where you need to be.

A quick search online of common issues similar to mine

I'm not going to say "Don't do this." But, I will say don't put too much faith, or time, into it. It can lead you down a "rabbit hole". Read it, and put it in the back of your mind. Then, see if your research and testing lead you in that direction. It is, usually, a safe bet that YOUR situation is not, exactly, the same. Every diagnosis has it's own characteristics.

This is how I begin, almost, every diagnostic journey. You never know where you will end up.

I agree with Chad, don't be so quick to clear the codes
I like to do a full system scan and see if there's any related codes to my issue in other modules, then check freeze frame to see under what conditions the codes set so I can stand a better chance at re-creating the fault.
Then I jump into service information.
I probably spend more time learning and understanding the system and code set criteria than actually testing anything.
Understanding the system is paramount to a successful diagnosis.

Same with what Chad and Noah said. Understanding how the system works is so important. I've been burned in the past making assumptions to find that the condition was normal under the set circumstances.

Knowing what is supposed to happen can allow you to think outside of the box and "be the computer" and complete circuits with your test light, or other tools.

Being able to read diagrams is critical as well. Once I know how a system operates, I can use a diagram to determine the easiest access points for testing. You can also eliminate unnecessary tests understanding a diagram. If you see that a fuse provides power to xyz, and y is the only part not working, then the fuse has to be good because the other parts are getting power from it. A little deduction. It can help to print a diagram and highlight the circuits that work. Sometimes even just doing that leaves a small section left to test.

In a more complicated situation, I'll put a battery charger on the car, get a voltage reading at the battery and write that on the diagram at the battery, then write the voltage values of my tests on there as I get them so that I don't forget and have to redo any work, remembering and predicting how much voltage drop to expect based on how many connections and how far away from battery the tests are being performed are.

The other part of the diagram is know what values to expect in the test. Understanding circuit design and system operation can help you know if your test results are normal. Remembering to do your voltage tests with the circuit activated is key. Voltage drop is the best circuit test you can do, in my option. I like to do tests by keeping the component in place when it's not working to avoid accidentally fixing the problem when removing it.

If you are bench testing with the ohm meter, remember that a good resistance reading doesn't necessarily mean the part is good, but a bad reading can mean it is not.

The tough part to me is really knowing if a number is off, is it far enough off to be a problem or not? A classic over-thinker like me will spend too much time wondering about that, but I think only experience will get you through those issues more smoothly.



Nissan Tech

Some good info posted so far. But no one has mentioned diagnosing in the absence of fault codes ?
In this situation a lot of the above methods have very little relevance.
My advise is let the car tell you whats wrong.
First, what is the customers complaint. Can you replicate it. What are the symptoms.
Can you determine is it a fuel/air, ignition, mechanical or electric fault. Once you have done this, you can carry out basic system checks and tests to narrow down the possibilities, until you find the fault.
I know this sounds very simplistic, but it is a very basic outline of how I tackle these faults on a almost daily basis. As a full time diagnostic tech I only work for garages/workshops, no retail/public at all. (too much hassle) So consequently a lot of my jobs are of the more difficult no dtc present type.
Rob.

Live2Fish88 wrote: I'd like to get better at my trade and take a more diagnostic approach than my thrown together, homemade one.

My experience has been that most of the best diagnostic technicians I've known don't necessarily follow conventional process. :silly: They made it up for themselves based on what kicked their ass, and what keeps them out of trouble.

For me, known good vehicles are absolutely invaluable. Use those suspension noise or brake job vehicles that come through to capture known good data. I do it every day. Ever looked at the scan data from the Power Steering Control Module on a Ford? Or the Fuel Pump Control Module on a Chevy? Now's the time.