Wes J
Senior Member
I was watching a heavy equipment mechanic on YouTube he got very upset with me after I accused him of guessing and winging parts at a machine. Maybe I was out of line. Got me to thinking, how much testing do you do before you condemn a part, especially a complex part like a pump, valve, or computer?
For example, he was trying to fix an issue with a 2 speed track system on an excavator. The machine has a high speed mode to track around when lightly loaded. It seems to use some inputs to the computer to control a dump valve on the pump stroke to give more flow. It was not working correctly or intermittently working.
He did some testing of the control side and found some minor wiring issues. With those fixed, he says "it has to be the valve". Next video he does some more testing and finds a default mode where the machine works correctly. He says "it needs a computer".
Finally, some folks with experience on this machine asked him to test some of the sensors. But, these are solid state sensors like hall effect RPM sensors, not like the old variable resistance sensors you can test with an ohm meter. He has no oscilloscope to test them. Finally he swaps one of the sensors and proclaims the machine is fixed.
Now, he did the right thing by not shot gunning parts. I will say that. But I think that if it was not for the comments in the video that machine would have gotten a new valve, and new ECU, and several new sensors that it didn't need.
So, how much testing do you do before you buy parts? If you make the wrong call, who eats it? Do you test sensors before changing them out (with a scope or any other method)?
From my experience, some things are really hard to test. Unit fuel injectors, hydraulic motors, some electric motors, complex units like a gauge cluster or ECU, etc. I always to rule out every other part of the system before making the call.
So what is your experience with testing versus changing parts?
For example, he was trying to fix an issue with a 2 speed track system on an excavator. The machine has a high speed mode to track around when lightly loaded. It seems to use some inputs to the computer to control a dump valve on the pump stroke to give more flow. It was not working correctly or intermittently working.
He did some testing of the control side and found some minor wiring issues. With those fixed, he says "it has to be the valve". Next video he does some more testing and finds a default mode where the machine works correctly. He says "it needs a computer".
Finally, some folks with experience on this machine asked him to test some of the sensors. But, these are solid state sensors like hall effect RPM sensors, not like the old variable resistance sensors you can test with an ohm meter. He has no oscilloscope to test them. Finally he swaps one of the sensors and proclaims the machine is fixed.
Now, he did the right thing by not shot gunning parts. I will say that. But I think that if it was not for the comments in the video that machine would have gotten a new valve, and new ECU, and several new sensors that it didn't need.
So, how much testing do you do before you buy parts? If you make the wrong call, who eats it? Do you test sensors before changing them out (with a scope or any other method)?
From my experience, some things are really hard to test. Unit fuel injectors, hydraulic motors, some electric motors, complex units like a gauge cluster or ECU, etc. I always to rule out every other part of the system before making the call.
So what is your experience with testing versus changing parts?