Fault Finding

To fault find or not?

Fault finding is an inductive process if the fault is not apparent during observation. Fault finding is no longer a linear process. If it costs $400 to clearly identify every fault or soon to be realised fault and it costs $600 to replace the oven; then no one will repair anything. We repair many ovens every day and most are completed for less than $260.00. So, for most customers it makes good sense to change to obvious fault rather than pay for diagnostics. Having said that, only the customer knows if their oven has been subject to a variety of issues over time. As a technician we can judge from what you describe. If you go to a mechanic and ask for your tyre to be replaced and then find the motor dies a few weeks later, when you knew full well the engine had been an issue from some time; then the mechanic can’t be blamed for changing the tyre.  However, if you start the conversation with context that includes the motor issues then the mechanic will provide repair options for you to consider.

Having said that above; some faults are not so readily diagnosed and can take time to determine what the cost might run into. We have to take a sensible approach and aim for an economical outcome. So, if you want absolute diagnosis, please make that clear at the outset and expect to pay for it separate from any fixed price offering.

Good inductive reasoning is very reliant on you the customer. If you tell us the oven is not heating, we know the probable outcome will be a new element. Many if not most elements display visual cues that takes seconds to verify that it’s faulty. Much like your car Tyre. No diagnosis necessary for a flat Tyre. On the other hand, if you tell us the oven is getting too hot, we’re not going to be checking your elements! So, if you know the element is cracked and falling apart but still working tell us, because otherwise will be likely conclude the thermostat as the culprit and leave the element to its fate.

The worst outcome is your observations simply aren’t correct or are vague. This means we must spend more time going through a possibly unnecessary tests. Time wasted equals money wasted! The more you can be sure about what you’ve observed and clearly communicate them, the quicker/cheaper we can resolve your issues.

Need a new Control Board; read on!

Risks associated with fitting electronic parts

So, we’ve had quick look at your oven/stove and advised it is likely a “Control Board” fault. “Likely” being the operative word! Because, no one in our industry does diagnostics on a faulty control board. To do this takes schematics and component specifications and expensive test equipment that is better suited to a lab than a van.

Further, the suppliers of the electronic parts will not refund any electronic parts once removed from its packaging. And for good reason, PCB’s are electrically sensitive, and can be damaged by inline electrical parts, or static electricity etc. Suppliers have no way of knowing what happens to the PCB once removed from its package, so they can’t or won’t take these items back.

Deductive reasoning

Not to get to technical; but for most oven/stove faults, we use deductive reasoning. Which basically means we can test certain conditions of a device which will give us a high probability that the device is faulty or not. For instance, a fan motor with an open circuit will not work unless the circuit is closed. It’s straight forward, we measure the resistance and we look for a specific result.

Simple example (Wikipedia)

An example of a deductive argument:

  1. All men are mortal.
  2. Socrates is a man.
  3. Therefore, Socrates is mortal.

The first premise states that all objects classified as “men” have the attribute “mortal”. The second premise states that “Socrates” is classified as a “man” – a member of the set “men”. The conclusion then states that “Socrates” must be “mortal” because he inherits this attribute from his classification as a “man”.

Inductive Reasoning

Inductive Reasoning is basically the opposite of Deductive Reasoning. Inductive reasoning starts with a conclusion and deductive reasoning starts with a premise. Inductive reasoning moves from specific instances into a generalized conclusion, while deductive reasoning moves from generalized principles that are known to be true to a true and specific conclusion.

We use this logic to deal with an unrepresented fault. In other words where we can’t specifically test the PCB for a fault, we test the rest. If the rest is okay; then the PCB is the most likely the culprit. Most likely being the operative word!

But to prove a printed circuit board (PCB) is faulty means we must spend the time verifying the other electrical devices like the are correct. To do this we rely on resistance readings, micro-ohm reading, thermal scans, temperature readings etc. This can mean having to isolate each device and verify the correct resistance readings or other readings. Sadly, resistance is only one element of ohms law. Which means it doesn’t provide absolute positive proof that the device is functioning correctly. Same with a thermal reading, a low differential reading might imply no cause for alarm. However, using a different cooking function might change the current being drawn is just enough for what is known electrically as the pinch effect to kick in, which can then dramatically vary the differential reading without changing the temperature range. Variables are the crucial factor in determining an unrepresented fault – inductive fault finding.

Economics

Unfortunately, the cost of the diagnosis time can be as expensive as the repair itself and generally prohibits an economical outcome.

Example: (not real values)

Call out

 

$100

Diagnosis Labour

1 1/2 hour

$165

Electronic Board

 

$220

Installation Labour

1/2 hour

$55

 

 

 

Total 

 

$540

Given the cost of the machine when new was say $650.oo, it is difficult to justify proceeding with the above repair.

There are many factors that would drive any final decision, such as age, condition, brand (original value), new price etc. For instance, if the machine was just 2 years old and a power outage had caused the PCB to be damaged then it might be sensible to repair the machine. But if the machine was 7 years old and was not in good nick then repairing the machine would not make much sense.

So, the reason this page exists is to save our technician from having to spend 15 minutes trying to explain these details so that you can make an informed decision. And to advise you (the customer) that if you do decide to go ahead and replace any electronic part such as the PCB, we won’t be held liable for the outcome. And once we remove the electronic part from its packaging, you the customer are liable to pay for it regardless of the outcome. Further the electronic part cannot be returned.

To diagnose or not?

We attend your home and quickly advise the hinges need replacing and one of the elements is down to earth. We tally it up and quote $300 (not a really price). We change to parts and start the testing process, while we do the paper work. This would be the average outcome of 80% of our work. But the other 20% when we get the oven going, we find the thermostat is unable to control the temperature. In hind sight everyone would agree the thermostat should have been checked before replacing the element. And, we won’t argue! In hindsight that’s correct. Setting that aside you decide to proceed as it’s still economical to replace the thermostat. Great, oven is running while we complete the paper work. Ohhh Nooo, the fan starts to make a horrible noise after it has been hot for a while. In hindsight it would have been smart to test the fan before doing the element, hinges and thermostat. Again, we can’t disagree! The only caveat being is the term “Hindsight”. It’s an expensive oven and changing the fan motor is still the most economical choice. We do the paper work all over again, the oven is now ticking away beautifully. We take payment and get on our way to our now late next job! On the way to our next job, you call and tell us the unit has tripped the safety switch as you have been running the oven since we left to give it a good test. We make a time and return, only to find the oven working fine. Grumpy, both of us, we agree it seems to be working fine and you pay for a wasted call. Sure, enough to add injury to insult you call about an hour later and advise the oven has tripped the breaker again.  We make a time and come out again. This time though we start the diagnostic process.

What a horror story! Each step in the above process is a progression toward this almost absurd and rare as rocking horse pooh event. And for sure somewhere along that progression it would have been smarter to start with diagnosis.  However, there are factors that have bought us all to this horror point, the customer new that the fan was noisy when it gets hot, but they didn’t think to mention that while we were dealing with the more obvious first issues. The customer also new that the oven tripped on the odd occasion, but never thought this was relevant to our approach; “diagnose or not”. Equally, we don’t experience this horror event very often. Hence, we let our guard down and do the mundane, routine repairs most days, weeks, months sometimes years before another horror event comes around to shock us back into reality. Things can go wrong and it’s no one fault. Why, you ask didn’t you start with diagnosis! Well, another very good reason is “hot”! Yes, ouch that’s hot! Many difficult to find faults only present under hot conditions for some considerable time. Sometime hours before they present. And no one wants to pay for tests that don’t contribute to a determination. While we carry Adelaide leading test capability such as Thermal imaging, Micro Ohm contact testing, live mA recording, ramp RCD testing, tachometer readings etc. these are not cheap processes and may not each be necessary. At the beginning of the above horror show, we might have started with a temperature test to prove the thermostat works correctly, which means the oven would have been “hot” eliminating our ability to replace the thermostat until the oven cooled. This process would have probably identified the noisy fan as well. But would not have uncovered the earth fault. That requires alternate test methods. In the first part of the process, we have to determine if the oven is at fault or if the circuit is compromised. Often a safety switch protects a variety of circuit. If you have a fridge, toaster, dishwasher and kettle all on the same circuit, each of this might contribute to a cumulative effect causing a trip. Add to this the fact that almost no RCD trips at 30mA as designated. Most trip around 25mA. Then if we divide equally the leakage for each contributing circuit it may only be allowing 5mA of leakage for the oven. Today it is regulation that every oven must be under its own RCD allowing a full 30mA. If this were the scenario then possibly the leakage of the oven might never be an issue. Once we have determined it is not a compromised circuit, we then need to conduct an insulation test. This would identify say 80% of earth faults. But, in our horror story, not the case. To find this fault we need to conduct a live mA leakage test on each component of the oven. Again, the operative word being “hot”! We already know now that the hot condition is a prerequisite of this fault presenting itself. To check each device while the oven is in its hot state is no minor feat. In point of fact, by this stage in the proceedings we have probably well past the repair cost of the overall job. Hence the reticence for proceeding down the diagnostic process first.

This isn’t intended as a means for you the customer or even us to determine which step and when to apply it. The factors that influence these choices are many and varied. If you have a Bunnings $500 oven then most of that is out of the ball park at element and hinges.

Bottom line, we do our best and hope the horror story remains rare! Bearing in mind our best compared to most is competitors is significant in its own right. Many wouldn’t know how to apply half the test mentioned here, much less have the threshold data for determining what the results mean.

0/5 (0 Reviews)