CPSIA: Unit vs Component Testing

Previous entries are here, here and here. These links will be repeated at close.

I don’t know how you’re feeling about this but I feel very discouraged much of the time. No, I don’t make children’s clothes but a great many of my friends do. Even if you don’t either, I’m hoping you’ll follow this closely and lend your support just as if it affected you directly. They don’t have any friends other than us.

At this point I’ll try to clear up a few questions that have been posed and provide other resources. I’m using four main sources for these entries.

  1. The AAFA -which I joined (@$3,000 fee!)- they’re having a Product Safety Council Meeting on December 3rd in Washington D.C. which I will be attending for up to the minute information and strategy. I hope you guys appreciate this; other than the costs I’m not fond of travel. I don’t have anything to wear either. I’ll be so outclassed it’s not even funny. Most of their members earn more in annual salaries than my gross annual sales so you could rightly surmise a bit of trepidation on my part.
  2. The CPSC website. There’s a listserv you can be added to upon request which will send you the rulings as they are released. If you need to catch up, here are all of the General Counsel advisory opinions released so far (scroll down).
  3. The videos of the CPSC Public Meeting on lead. There are three of them: staff and panel presentations and a Q & A. Of these, I’ve found the panel presentations from PLU to be the most interesting. As I write this entry, I’ll be referring to specific speakers in the panel discussion. It’s worth a listen with the exception of one guy I’ve dubbed Captain America for posturing and adding little to the content. This being a public blog I can’t say who that is but anyone who was there or watches it will know exactly who I’m talking about.
  4. The videos of Rick Woldenberg’s portion of the panel presentation. His presentation was so good, somebody took it upon themselves to post it on Youtube in three parts. He’s not a professional speaker and it takes him a few minutes to hit his stride but he is nothing short of compelling and convincing. While a large concern (a manufacturer and distributor of toys), he cogently makes the case of why these laws would adversely affect small producers too. If you only follow one link from this entry, this is the one to watch.

The rest of this entry focuses on unit vs component testing, what it means and why it matters.

Unit vs Component Testing
Component testing means you can test your inputs (fabric, buttons etc) going into the product individually. Unit testing means a completed product is tested rather than individual inputs. In the panel discussion, Stephen Lamar (AAFA) talks about this, some 21 minutes into the video. He also mentions this law is overwhelming the testing community because we don’t have enough labs even assuming everyone had the time, money and logistical infrastructure to do it.

Component testing is preferable for two main reasons.

  1. This saves money and time if you’re repeating fabrications and inputs over several styles -as you should be for continuity purposes and meeting minimums for purchasing.
  2. It achieves the safety intent of the law best.

Unit testing is not preferable for two reasons:

  1. It’s wasteful. If you’re using the same exact inputs in your products across styles, it’s dumb to retest the same zipper that is identical in each unit.
  2. It is less safe, circumventing the intent of the law. Let’s say you melt down a widget or whatever to reduce it into constituencies for testing. The total weight of the product is calculated in comparison to the weight of lead or phthalates so maybe the product is deemed within guidelines -but that doesn’t mean it’s safe. For example, let’s assume the major attributes of the product -like a flat screen tv- are inaccessible to a child and won’t be touched by them but the on and off button is and it’s completely constructed of lead. Should this product pass? Under current guidelines of unit testing it would but it shouldn’t. It would not pass with component testing. In other words, it is more important to test inputs as those can be weighted with respect to accessibility by the child.

Perhaps you’re asking yourself why lead needs to be in anything at all. That was addressed best by Christopher Cleet from the Information Technology Industry Council in the panel discussion (he was a good speaker). He said (paraphrased) “it’s not that we have truck loads of lead and are looking for a place to put it, it’s that it’s needed and useful”. Take your typical TV; that screen glass is full of lead (so don’t let your kids lick it). It has to be, otherwise viewers would be exposed to radiation. Mr. Fashion-Incubator concurs saying lead solder is best. In the EU, they’ve gravitated to tin solder (RoHS). The problem with tin solder is that it’s a friendly and sociable metal and likes to hook up with tin solder nearby forming tin whiskers. This will short out the unit causing failures in everything from pacemakers to video games. He adds that this will happen after your warranty is expired leaving you stuck with a lousy product.

At the risk of being pilloried in the community, it’s only fair to mention two reasons why unit testing is preferable. I didn’t hear any of the speakers mention these and it’s important to address so we can develop appropriate resolutions.

Why unit testing is preferable:

  1. The veracity of component testing can be questionable under certain circumstances. By this I mean that it is very very common in outsourcing situations for your contractor to make unauthorized substitutions for components at the last minute. This is the bane of small manufacturing. While you may have tested the proposed component when in the sourcing stage, unless you’re a large concern who operates your own plants overseas, you often don’t know what you’re getting until it lands at the dock. At this point you don’t have many choices. Maybe you’re “lucky” in that the only last minute change was shortening the sleeves but you’re stuck. You have to get with buyers to see if they’ll take the goods, hopefully without a discount or other losses.
  2. The other problem with component testing is that just as we make prototypes which we then present to buyers for orders, our suppliers do too. Just as you sometimes have to make changes to a style that was ordered and have to go back to your customer and okay the change, we’re working off of prototype components sometimes. You only have assurances on stock items. This is why suppliers will often call you to glean the liklihood of you buying a particular item down the road because they want to know if they’ll have enough orders themselves to justify producing it. As with us, there may be changes between the prototype and production component.

This is probably enough for one entry. I’ll continue in part two.

Related Entries:
New product safety regulations that affect all manufacturers
CPSIA Requirements
National Bankruptcy Day

Related in the forum:
The War Room: CPSIA & Consumer Safety. This is a very active section with nearly 60 different threads and over 1,000 postings. Open to the public.

Get New Posts by Email


  1. Vesta says:

    Lordy. Thank you, Kathleen. I so appreciate your time, effort, and expense. Once we get through “death valley” (the period from Thanksgiving through New Year’s), I’ll help defray the expenses you’re running up. I suggest everyone else who is appreciating this information do the same. (“Make a donation” in the righthand column.)

  2. Maura says:

    I want to add, as an artist and craftsperson who makes one of a kind items (they may be similar in design or use the same pattern, but no two use identical materials). If I were to be able to comply with the requirements, I would have to make two identical items, let’s say dolls, which individually take 5-6 hours to produce, then submit one for destruction in the process of testing. I would then have to work the testing fee into the price of the doll.

    That would cause the dolls to cost well over (many times over) what I would charge for them, and would place them outside the reach of most people for a child’s cloth doll.

    I think that artisans have simply been forgotten in the process of writing the law.

  3. Cinthia says:

    Unit testing is crazy – if you wait until the factory has purchased all the raw materials, and then sew the product and THEN test it – what happens if it fails? When doing a ‘group’ of same products – I’d feel better testing the components first then following up with testing one finished product – maybe an FRX scan just to confirm. Last quote I got for testing one backpack was over $1,000 – even the zipper has 5 components to it. Oh, and it failed, and now the factory says “so sorry – production is finished!”

  4. Marc says:

    I respectfully disagree about RoHS. It has been quite successful in reducing hazardous substances in electronics on a nearly worldwide basis since 2006.

    Today it is difficult to purchase a consumer electronics product that isn’t compliant. Apple, Dell, Sony, Motorola, HP, Cannon, Epson, Sharp and numerous others have been making millions of RoHS products for years now without reliability issues.

    Tin whiskers is a well known, researched issue that has largely been mitigated by new solder alloys like tin-silver-copper. To date there are no significant known documented failures of RoHS compliant products due to tin whiskers.

    Here’s a great link:


    Luckily CRTs are obsolete with the advent of LCD technology, so I don’t see this as an issue long term.

    As for plumbing solder, it has been lead-free for years.

  5. Eric H says:

    I can’t remember the last time I heard about a short in my plumbing, so maybe that is slightly irrelevant? Nevertheless, it’s lead-free because the lead is replaced with silver, a slightly more expensive proposition.

    Oddly, NASA and the aerospace industry ban pure tin because of tin whiskers, and RoHS has exemptions carved out for high-reliability components.

    Remember the NASA investigation into the Toyota unintended acceleration problems? The NHTSA reported, “In one case, NHTSA obtained the pedal and tested it, finding a resistive short in the pedal position sensor. NASA subsequently examined the pedal, confirmed the existence of the short, and found a “tin whisker” to be the cause of the resistive short.” So RoHS hasn’t been as event-free as advertised. Tin whiskers take a while to grow, so banning them one year and declaring victory the following because nobody has report a problem with their throw-away consumer electronics is not without its problems.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.