Background:
Amarula Electronics created the 5-R system as an approach to fault-finding electronic products. Whether the device in question is a simple device with few components, or a full embedded system running an operating system such as Linux or Android, this method provides a proven route to ensure success without wasting precious time and money. Historically, we have worked with many projects where customers required our assistance in order to solve an unknown fault.

These faults could arise from many causes, Including Short-Circuits, and even intermittent faults that only become apparent due to for example, a sudden rise in temperature. The list can be endless, so the method of finding the fault can vary depending on the source. This page will not provide instruction on how to fix a PCB fault. We will discuss some of the finer points of these methods in our future blogs.

There are many times when attempting to fault-find, that the brain automatically leads us away from the actual area of the problem which can be a costly mistake. This is why it is important to have a plan to follow which can prevent us from straying from the path to success.

It is often difficult to provide an estimation of time to the customer, as we are looking for a fault, based on a symptom which can vary widely. So, let’s jump right in and look at our proven method which has helped us succeed time after time.

The System:

Research
Knowing the symptoms of the reported fault is half of the battle when fault finding. Imagine taking your car to the Garage for a fault and the Mechanic asks: “So, what’s the problem?”

If your answer is anything other than specific, the Mechanic will undoubtedly come back with something not too dissimilar to: “Oooh, Well, It might take us a while to figure out what’s going on”. At this point you start clutching your wallet/purse and begin to worry how much this is going to cost.

Electronics Fault Finding is not much different from an estimation basis. If the customer can give you an idea of the Issue, you already have an Idea of how long it might take and how much it would cost.
Be sure to get as much information from the customer as possible from the outset.

Here are some examples of just some of the questions you should be asking:

  • How many faulty units can you provide for testing?
  • Is the Fault repeatable?
  • Does the device power on?
  • Does the device get warm/hot?
  • Is the fault apparent on some devices, or all devices that have been assembled?
  • What are the symptoms of the fault? (Not booting, No power, Freezing, No sound…. etc)

Once you have a good feeling for the symptoms, you can begin to decide on how to reproduce the fault accurately.

Reproduce
The second and key aspect of this system is to reproduce,reproduce and reproduce again in order to be sure that the fault is indeed a fault and not a bug or external influence. If the fault is reportedly on every device this section should be relatively simply to reproduce. However, if the fault is only on 10% of devices, this section can be more time consuming, especially if the fault is intermittent.

Resolve
Probably the most time consuming section of the system is to come up with the resolution…We will not go into this now as there are too many possible faults you will come across, each of which may have multiple resolutions. The key is finding the most stable, cost-effective solution, that is repeatable, which leads us to the next section of the plan:

Repeat
Performing a fix on just one device is not enough to ensure stability moving forward, The more products that you can repeat the solution with the better. This will prevent customers coming back further down the line with complaints that the fix has only worked on for example 70% of devices. We have seen this on occasion, where the customer only provided one sample of the product, which we were able to resolve. However, due to tolerances in the components used, the fix applied was later seen to work on not all of the units produced. This led the customer to supplying more units and spending more money and time on the same issue. So again, I stress that it’s extremely important to ask from the outset: How many faulty units can you provide for testing?

Release
Now is the time to provide the customer with clear and concise instructions on how to resolve the issue. Performing a modification in your lab can be quite different to how the customer would perform the same modification in their production environment. Always be sure to understand the customers build process and tailor the solution to fit this process whilst minimizing any additional time taken to carry out the build.

I hope this helps you to control your fault finding process and become much more productive as we have using this 5R system.

Happy Fault Finding!
Amarula Electronics