Testing, testing...testing?

How do you test your products?  Is it easy?  Is it accurate?

No matter what you make, no matter how you make it, we all worry about the quality of our products.

Quality in today's world can not be overstated...losing the trust of a customer from a bad good or article is incredibly expensive.  Either through the loss of business or the struggle of retaining that business (discounts, offering money-back, replacement units, expedited shipping, rental equipment, ect).

So what's the best way to to check for quality before it ever leaves your facility, or better yet before it ever leaves the operator?  You can try things like checklists and operator inspection or next-operator-in-line inspection.  These things work to a degree, but you are still counting on an operator(s) to catch the mistake.  We're all humans...not only do we make mistakes, we also miss them upon inspection.

Maybe you need a test station?...let's look into it

 

What is a test stand?

A test stand or test station is a device used to measure, validate, and confirm the intent of an article.  By article I mean anything you produce, be that a machine/fabricated part, an assembled unit, a sub-assembly, or a control system.  If you want to make sure something meets specifications, whatever those requirements may be, then a test stand should be made in such a way that it gives you an answer quickly that the article is in spec.  Usually test stands have some data collection capability as well as operator feedback (discussed more below).

 

Why would I use a test stand?

Lot of reasons!  Is there a specific point in your process that you would like to test the part/article?  If so, can you quantify the result you are wanting?  If so, how hard is it to test for those results?

If it's easy or achievable, why not make a device or stand to test this.  It can only make your process and quality better!  We all know the compounding effects of missing a mistake early on and catching it at the end, or worse yet, missing a mistake that affects an entire batch of parts.  Consider testing early in the process and continuously as the article changes form and new elements need to be tested.

 

What to keep in mind?

When designing or building a test stand, keep these elements in mind.  You may want some of these or all of these.

- Operator feedback

Operators need a way to interact with the stand, how will they know if the article being tested passed inspection?  Can this be simple feedback (green light, red light) or does the operator have to read gauges and make calculated decisions to determine if an article has passed?  Keep in mind, the simpler the feedback, the less human error comes into play that can affect the results.

- Data collection

Do you need to track the test?  Is data important to warranty claims or trend spotting down the road.  In most cases, yes, data is very important to access and understand when testing.  Consider data collection when designing a test stand and picking out components.  There is some upfront cost but the feedback and ability to better integrate processes down the road is exponentially valuable.

- Fully automated/Partially automated/No automation

How important is the process of testing and what variables play into the results?  If no automation is included or partial automation does this effect the outcome of the test?  If no, then a non-automated solutions may work just fine.  Our experience tells us that when companies choose not to automate a feature of a test stand, then that forever becomes a variable that can effect the results of the test.

- Visual inspection

Do you want to visually inspect an article or measure a certain feature(s).  One option is to have the operator visually inspect the unit, looking for all the bits and pieces.  But again, anytime an operator is involved human error comes into play.  Consider an Integrated vision system, these have come a long way and are very affordable for most inspection type applications.  Literally a vision system can look at your part and determine if components are where they are supposed to be, as well as measure tolerance areas and look for key features.

So...

No matter your approach, keep quality in mind and make sure you are testing for the right thing...not just testing to test.  In those cases not only do the results not help, then can actually complicate things down the road.

If we can help you with any elements to your testing or if you'd like to discuss a custom stand made just for your process...we would love to chat!  Contact us here

OTHER ARTICLES: