The Definable Distinction Between “Cheap” and “Sub-Standard”
Chapter 1 – “That was Scary”
I have a wrist watch that I describe this way, “it’s like me – old but mostly functional.” What I don’t mention is that when it’s not working properly, the fault is not with the watch, but with the wearer. The watch is “self-winding” which really means it must be wound with motion. And, since I am not the most focused of people, usually it’s running more than a bit slow. A friend sensed the opportunity for a superlative gift and purchased me an Automatic Watch Winder Box.
The box is beautiful. It’s dark brown burlwood with a high gloss finish, a patterned velvet-pile interior and a winding motor that features selectable programs to properly maintain a self-winding watch. It also came with a wall-plug adapter which gets us to the Cheap vs. Sub-standard part.
The box lives in my bathroom on the counter near the sink. One morning, after about a year of service, freshly showered and grooming near the sink, I needed to move the adapter from one NEMA 5-15 outlet to another. I gripped the adapter as one should—by the PC case—and gave a firm pull remove it from the outlet. I was startled when the ultrasonic weld that held the clamshell case together just gave up the ghost and the top half of the case ripped off in my hand exposing the PCA—primary side and all—still plugged in and powered. I found a rigid plastic ruler and pried the exposed adapter converter from the AC wall plug. No people or animals were harmed during the entire episode.
The knee-jerk reaction by a typical individual, and by myself at the time, was to exclaim, “What a piece of junk!” But, as a power-products person and having been part of teams that write power-converter specifications, project development plans and validation and verification protocols for newly developed power products, I needed to take a deeper dive.
When the primary side of the power supply was exposed, I knew what not do, but my friend who purchased the winding box would not have known. My home is equipped with a GFI breaker in the circuit, so if anyone touched the PCA and shorted the converter to ground, in theory, the breaker would have tripped with no one feeling “overly energized”. What I would like Commodity and Program Managers to deliberate is not the root cause of the failure but rather, loosely, what was the probability that the selected power converter, a critical component to the watch winding box system, was going to fail and was the risk of this failure proportionally higher than other blocks of the system (the wooden box, the velvet interior, the DC motor, etc.). As part of a project post-mortem I would break it down into the questions below:
- Was the power converter, viewed as a system block, sub-standard?
- Assuming the power converter, as a sub-system, had a low-cost requirement, i.e. “it needed to be cheap”, could the overall risk to the system have been mitigated?
- Were the final requirements of the power converter sub-system, especially the low unit cost, when viewed as part of the entire system, worth the risk?
Taking a closer look at the wall-plug adapter, the first thing you would note is the clamshell case is fitted with a USB A socket so the DC cord set, terminating in a standard barrel plug, is removable. My opinion—for what it is worth—is using this type of configuration was brilliant. Assuming the required DC output of the adapter is 5V and the load requirements of the system can be met within the confines of the USB power standard, the power supply sub-system has been commoditized to the highest possible level. First, scores of vendors, big and small, have standard product that meets these requirements and, this is the key, within the that vendor group, many can be labeled as Tier I Suppliers. They have the engineering depth, operational process expertise and financial stability to consistently supply product that meets specification.
Secondly, driven by the mobile phone market and other mostly portable consumer products, the quantity of devices that have standardized on USB power provides incredible volume that can be leveraged by the device maker, in this case the maker of the watch winder box. These two items combined result in a very high probability of success for finding an incredibly low-cost adapter from a Tier 1 Supplier, likely a genuine standard part from two or three Tier 1 suppliers could be on the BOM. The system power supply requirements, especially utilizing the USB A socket really answer question 3 above. The risk as viewed by the Program manager (engineering) and the Commodity (operations) and Quality groups should be well below any threshold that would be considered dangerous or ill advised. In fact, given that the risk was so low, the requirements of the power converter, including a low unit cost and added profit margin it provides, was worth it, even if when using a pareto analysis, the cost of the power supply sub-system was only marginally significant.
Without a detailed technical analysis and ignoring for now the quality of the converter design, just by hearing the circumstances of the failure, we can safely assume that the power adapter manufacturer had a process control problem. Regardless of whether the PC adapter case was molded improperly or the ultrasonic weld was done improperly the result is the same. If the plastic was overly brittle due to a processing issue at the molder that should have been found during incoming inspection. If the welder was not set up or being operated correctly, that should have been discovered during a reasonable set-up check or early in the production run during a weld check set at regular intervals.
Another fact to note, a quick inspection of the power supply PCA showed the solder workmanship was well below the IPC-A-610 Class 1 acceptability standard. The poor PCA and final mechanical assembly of the power adapter make it highly likely the vendor used for my watch winding box system was not a Tier 1 supplier. This allows us answer questions I and II above.
Yes, the power converter as viewed as a system block was sub-standard (a serious piece of junk) and yes, any overall risk to the system driven by the requirements, including a low-cost converter, was avoidable. By being more discerning in vendor selection and using a Tier 1 supplier with systemically sound process controls a dirt-cheap power supply could have been found that had a very low risk of failing in the field. In this case my watch winding box maker could really have had their cake and eaten it too.
