I'm curious as to when the responsibility of a manufacturer ends when dealing with emissions and RF testing, part of FCC and CE tests.

I repair TVs as a hobby and as business. I got a TV in from someone which would not power up most of the time. I found the soft-start circuit was stopping the TV starting, so I inhibited that. Now it works fine. But the PFC transformer makes a very audible whining sound, and the PFC bus voltage now measures 328V - when it really should be nearer to 390V. This tells me the PFC is not working (although the TV is fine with the reduced bus voltage) and so the TV probably wouldn't meet its emissions requirements. In this case, would the manufacturer be responsible for this?

Another example, an old Compaq power supply had a very odd "feature". When the 5V output was shorted, it wouldn't switch off. It would melt whatever was attached (including its own cable harness) and in the process make enough EM noise to cause an AM radio within half a metre to lose a station. Given that the 5V shorting is not a default configuration of the PSU, but it could happen, why is it covered in FCC/CE logos?

See, I am designing a power supply which might have an expected lifetime of 50,000-100,000 hours. But it won't last forever, that's pretty difficult to do. At the beginning of its life it would meet the requirements. But as it gets older, electrolytics degrade, ripple incrases, noise increases. Eventually, it won't meet the original requirements.


3 Answers 3


I live in Europe, and I can't comment on FCC regulations.

About CE certification there seems to be much misunderstanding. The CE mark on a product claims that the product complies to all relevant CE regulations, but it's the manufacturer who decides to use the marking, not some certifying organisation. If I decide to manufacture some product and place the CE marking on it, I'm free to do so. Did my product undergo required testing? No. Am I liable in case of technical problems due to non-compliance (for instance too much EMI)? Yes. A third party (usually a customer) can challenge my CE-compliance claim and then I have to prove the product is in fact CE-compliant.

(I wasn't sure about this myself, but I was told so by the director of the quality department of our company, who had a seat in several of those CE committees.)

  • \$\begingroup\$ Yes, although if the product has relevant CE regulations then you must apply the CE mark to the product or it's documentation. As you say, you don't have to have tested the product - you just take responsibility for any consequences. But whether there is some limit on the length of time a product can reasonably be expected to remain compliant is an interesting question. \$\endgroup\$
    – MikeJ-UK
    Jul 15, 2011 at 8:37
  • 1
    \$\begingroup\$ I would also bet that by having had a 3rd party verify your product, you would have a better defense if down the road someones device wasn't conforming. \$\endgroup\$
    – Kellenjb
    Jul 15, 2011 at 15:53
  • 1
    \$\begingroup\$ @Kellenjb - Yes, it pays to do the testing anyway, better safe than sorry. The third party could be a competitor who's done some competition research (when I worked with Philips Audio we had two engineers who did only that!), and wants your product to be taken out of the market. \$\endgroup\$
    – stevenvh
    Jul 15, 2011 at 16:15

If you replace blown components, the equipment should meet the original specifications so could assume compliance holds, but if you remove sections of circuitry, that assumption is clearly invalid. Therefore you should retest to current standards before placing the repaired/modified equipment back on the market (selling it). The original manufacturer is not responsible for the non compliance of a modified item, unless the repairer is working under orders from the manufacturer.

I found this section on CE marking: "Reconditioned and Reconfigured Apparatus

The directive does not apply or reapply to a repaired or reconfigured apparatus that has been modified with restoration as the intent. Only if equipment is considered "as-new apparatus" will the directive apply or reapply to a reconfigured apparatus that has been modified with the addition (upgrading) or the removal (downgrading) of one or more parts. Note, however, that though the modification may not result in an as-new apparatus, the modifier must, in any case, document any changes, EMC analysis, tests carried out, and the modifier's final conclusions. In certain cases, the modified configuration already may have been envisaged and documented by the original manufacturer as an EMC-confirmed variant. In such instances, the EMC Directive does not need to be reapplied."

It doesn't seem reasonable to expect the power supply to be compliant under fault conditions. It would be tested only under different operating conditions (no load, fully loaded, different combinations of loads on the outputs). Expecting it to remain compliant while not necessarily operational under fault conditions would require the manufacturer to think out all the possible misuses. Whilst I have heard that in the US, manufacturers can be responsible for injuries sustained even during product misuse, this doesn't appear to be the case in Europe yet.

As far as staying EMC compliant over the life of the product, it should be designed for that, but it is very difficult to test for. In the same way, products are often only EMC tested once at the end of the design/start of manufacturing cycle, but should be regularly sampled to ensure that they are still being made to the same standards (components and materials get substituted over time). If you expect a large change in the EMC performance with aged components, then a heat/soak test for some hours prior to EMC testing would be advisable. But most products wouldn't be expected to change that much it might be considered unreasonable to simulate years of use, depending on how critical the application is.


I can't answer all your questions, and am not a lawyer anyway. However, I'm pretty sure in the first case the manufacturer is off the hook since you modified the TV. In fact, I'm guessing you're now responsible should the TV interfere and someone were to file a complaint.

Disabling the soft start circuit was a very bad idea. It sounds like either you disabled more than the soft start, or that circuit had other functions you haven't bothered to understand. Unless the manufacturer has published a service bulletin advising of your "fix", I expect you now own all the legal liability should this TV go on and do something bad, including causing radio interefence or zapping someone. It won't matter whether you think your modification had anything to do with it.

  • \$\begingroup\$ N.B. The TV would power up before hand, but only occasionally. The soft start is only for the secondary side control - not the primary side - to shut down in case of the rails going out of spec. Very common problem with the 17MB22 series. Of course, what you said still applies in terms of legality. \$\endgroup\$
    – Thomas O
    Jul 15, 2011 at 16:40
  • \$\begingroup\$ For the purposes of this question, I'd like to make it clear that the TV powered up occasionally, and did the same thing (PFC bus out of spec, whining noise.) I suspect it developed the fault over a long period of time (before it started getting stuck in standby); is the manufacturer required to guarantee their product will continue meeting spec over time? \$\endgroup\$
    – Thomas O
    Jul 15, 2011 at 16:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.