• Insufficient sample size – Not enough parts were tested to meet sample requirements. In extreme and particularly nefarious situations only one device may undergo the required tests, although the implication is that the proper sample was tested!
• Inferior level testing – This is a sort of bait and switch. For example, you might get min/max results that are actually the result of curve trace testing instead of a large data set from the DC testing that you paid for.
• Incomplete testing – Results are listed for a category type (e.g., surface testing) instead of broken down into each individual method (e.g., blacktop/acetone, blacktop/1-Methyl-2-pyrrolidinone, blacktop/DynaSolve and scrape test). In this case, only simple non-heated testing may have been done.
• Inadequate test duration – It may take a longer period of time for some issues to reveal themselves. For example, a solderability test requires an 8-hour steam bath as part of the procedure; thus a 24-hour turnaround is not possible without shorting the duration of the test.

Inadequate testing is simply going through the motions and thus is doesn’t truly mitigate the risk of using or reselling substandard parts. With that said, it’s something you should try to avoid.  These warning signs can help you do so.

ADVANCE WARNING SIGNS

• Lab suggests house-level testing – In-house, tier level testing is likely minimal testing that actually provides little to no value; although it may be profitable for the test house.
• Very short turnaround time – Although it is generally a plus when a facility can perform testing within a shorter timeframe, unusually short turnarounds could be a sign that the lab is shorting testing in some way.
• Surprisingly low estimate – If the quote is very low, it could be a sign that the lab is not planning to perform all the testing you really need. Please note that not all low-quality testing is cheap; we’ve also seen evidence of subpar testing that was actually on the pricey side!
• Surprise clauses – Vendors might include (and fail to explicitly mention) clauses that reference an inferior level of testing than you expect.
• Lack of specificity – Quote contains only general references with respect to tests to be performed. For example, “authenticity testing” instead of “AS6171 moderate risk level, model 2.”
• Lead time subject-to-change clause omission – No lab can guarantee they can do authenticity testing in a short timeframe because there is no way to be sure what they will find and how much extra work that will entail: research, communication with manufacturer and/or customer, additional testing and interpretation. And that all takes extra time.

TEST RESULT RED FLAGS

These red flags on test reports may be an indication you were a victim of subpar testing.

• Incomplete results – They are very brief and do not provide the data sets behind them.
• Results don’t make sense – If the results don’t match the data sheet there may be a problem. For example, differences in dimensions, pinout configuration and lead finish.
• Wrong data sheet – The data sheet provided with the report is not a direct match to the device tested. Two potential disparities are the date code and the manufacturer.
• Fine print – Footnotes or other fine print might actually reveal that an inferior level of testing than expected was performed.

BEST PRACTICES

There’s no fail-safe way to guard against less competent or even unscrupulous test houses BUT there are some things you can do to help ensure that the testing you receive provides the insight and assurance you need for the results to be reliable, actionable and, basically, valid.

• When at all possible, request testing to a specific military or industry standard (such as MIL-STD-883 or AS6171) because if you don’t, the results you get will not be all that meaningful. Cherry picking test methods may be cheaper for you and easier for the lab but it’s inherently insufficient.
• Avoid testing to any facility’s house standards, which are basically cherrypicked test methods that primarily give the appearance of valid testing.
• Always insist, up front if possible, on read-and-record data with results.
• Review past test reports and check for red flags. If you see a disconcerting pattern then you may want to consider trying a new lab.

Also concerned about subpar electrical testing? Read our 2016 story:  Avoid Paying for Subpar Testing.