I find it surprising that many involved in Engineering one way or the other have so little practical ability. The really skilled chaps (and some ladies too) take an idea and run with it, they make new items, they invent new ways, and turn out a useful item. Others do not seem to "get" the idea of the product they are trying to design, make, or mend. It is interesting to look at youtube where people have made videos of mending cars (and other things too) and the device has been to be repaired and come back still faulty or even completely broken. Take some of the electronic problems there. The start is a diagnostic computer that points to a part that is faulty, say a throttle position sensor. The dealer changes it but the fault is still there. He changes the throttle body (the other end of the chain) and the fault is still there. Then he is stuck, clears the fault record and returns the car to the customer (with a big bill) The trouble is that the customer finds that the car is still not right. He takes it to a friend who also has a car computer and he finds that the CAN bus has errors on it during driving occasionally. He looks at the wiring carefully (which is very hard as most of it is buried deep in the engine and dash, but still finds nothing. After a long time he finds that one of the windows doesn't work either, and module 157, the window controller number 3 has a faulty CAN transceiver chip which he changes and all is then well! What is the problem that was so difficult to find? It is that the original system design did not consider how to diagnose faults with the control bus on modules that were not central to the engine controls. It is that intermittent faults or very low rate faults in bus systems need complex automatic diagnostics to stand a chance of finding them. Every component needs a self-test routine to check operation and communication. Do they have them, no! Someone will always say "it is not economic for us (manufacturer) to do this". The garage mechanic is told "change modules until it works and then the customer pays". You will note that no one cares about the customer, service, or putting the then known good modules through the changing process back in the car. Engineering is being driven by the wrong goals. Brunel made provision for repairs to the chains holding up the Clifton Suspension Bridge without taking the bridge apart completely. It used to be automatic to fit diagnostics, in a sonar system I worked on once a long time ago, there were complete automatic diagnostic tests to every board in the system. If the light came on you swapped the board. Simple and very effective because this could be done in the middle of battle!

I find it surprising that many involved in Engineering one way or the other have so little practical ability. The really skilled chaps (and some ladies too) take an idea and run with it, they make new items, they invent new ways, and turn out a useful item. Others do not seem to "get" the idea of the product they are trying to design, make, or mend. It is interesting to look at youtube where people have made videos of mending cars (and other things too) and the device has been to be repaired and come back still faulty or even completely broken. Take some of the electronic problems there. The start is a diagnostic computer that points to a part that is faulty, say a throttle position sensor. The dealer changes it but the fault is still there. He changes the throttle body (the other end of the chain) and the fault is still there. Then he is stuck, clears the fault record and returns the car to the customer (with a big bill) The trouble is that the customer finds that the car is still not right. He takes it to a friend who also has a car computer and he finds that the CAN bus has errors on it during driving occasionally. He looks at the wiring carefully (which is very hard as most of it is buried deep in the engine and dash, but still finds nothing. After a long time he finds that one of the windows doesn't work either, and module 157, the window controller number 3 has a faulty CAN transceiver chip which he changes and all is then well! What is the problem that was so difficult to find? It is that the original system design did not consider how to diagnose faults with the control bus on modules that were not central to the engine controls. It is that intermittent faults or very low rate faults in bus systems need complex automatic diagnostics to stand a chance of finding them. Every component needs a self-test routine to check operation and communication. Do they have them, no! Someone will always say "it is not economic for us (manufacturer) to do this". The garage mechanic is told "change modules until it works and then the customer pays". You will note that no one cares about the customer, service, or putting the then known good modules through the changing process back in the car. Engineering is being driven by the wrong goals. Brunel made provision for repairs to the chains holding up the Clifton Suspension Bridge without taking the bridge apart completely. It used to be automatic to fit diagnostics, in a sonar system I worked on once a long time ago, there were complete automatic diagnostic tests to every board in the system. If the light came on you swapped the board. Simple and very effective because this could be done in the middle of battle!