In conclusion I will always say that the CE mark represent quality in the way that you can expect the product to adhere to a minimum standard of manufacturing and testing.
The CE marking is not a quality control stamp, it's a declaration of the company that makes the good that they follow EEA rules.
You are both right. CE Mark represents quality but not quality control. This is why it’s not a panacea and also why EU parliament members complain when products are imported into Europe with fraudulent marks, i.e. China Export when there isn’t an actual CE mark.
CE Mark is self-certification. I could inaccurately declare my product meets the safety standards. It would be fraud if done intentionally, but companies could lie or make mistakes. Lying has economic consequences such as not being able to bring other goods, and fines.
I certify a product and then that’s it. It’s not a quality control stamp because a) an independent third party hasn’t double checked my initial claim b) I don’t
need to do ongoing testing off production lots
For electrical goods, requirements can be things like no lead solder used indiscriminately (RoHS), that the electrical insulation is appropriate that it’s not going to kill you, etc.
The CE mark REQUIRES but does not VERIFY in advance, the presence of a
- A bill of materials (BOM), product descriptions, and/or design notes.
- Any testing performed.
- Certificates from approved test laboratories (notified bodies) if required.
So I am self certifying such a set of documents exist, suitable for inspection if asked, but not required to be proven.
In practice, there is a lot of liability if there is a problem with the quality and you are a Sony, Masimo/Sound United, etc. so companies like that have ISO certification which ensures ongoing testing/lot clearance, etc. Everything is risk managed. There is an acceptable level of SINAD risk which is set by the company. We accept that x% of products fall below our design spec, so we advertise a level of performance that ensures that less than x% will fail. If we have more, we trigger an investigation. X will be pretty high since the hazard analysis will be low. No one will die from a product shipping with higher noise or THD, but it does hurt reputation and we have to replace the component that fails, etc. We accept that less than y% will electrocute and kill someone. This number has to be so low that it is effectively as close to zero as feasible. The get this level of confidence, we need to have independent third party testing, etc.
You have companies like Benchmark which get UL listing for products like the AHB2.
Key is that amps and voltage can be dangerous and there are reports of injuries from things like 12V car batteries and things under 50V (though 50V is a threshold from OSHA).
Bad actors will always do bad things. In the pharma world, “Bottle of Lies” is a great read about one company that was caught basically fabricating everything…