No, under the pcb is a copper stud which seems to have a threaded metal insert that is pressed into the chassis (2 mm thick metal). It's an M3x8.
The exposed parts of the tracks and pads are plated, but no solder is on them. Any empty hole was masked before going over the wave solder machine (the pcb definitely doesn't look hand-soldered).
Thanx. Visually checking for faults is second nature to me because it's part of my job. I estimate about 90-95% if the faults I encounter in the testing stages of production can be found just by looking over a pcb (e.g. loose pins, wrong polarity, missing components, solder shorts), or perhaps comparing it with a "golden board" (e.g. wrong value/type components).
CE mark does not mean that the appliance has passed the tests, just that the manufacturer claims that the tests have been passed. This is the wonder of self-certification. CE means nothing, as compliance is not enforced. In some parts of the world regulations are only followed when it is known that someone is checking.
Using filter capacitors with no ground seems increasingly common. I discovered this when unplugging my FM antenna from my tuner. The antenna is connected to the TV via the downlead diplexer, and the TV, DVD and VCR each inject a few 100uA of mains into the 'ground'. This was properly grounded via my homebrew FM tuner, until I unplugged it. Not dangerous, but an unpleasant tingle.
Using filter capacitors with no ground seems increasingly common. I discovered this when unplugging my FM antenna from my tuner. The antenna is connected to the TV via the downlead diplexer, and the TV, DVD and VCR each inject a few 100uA of mains into the 'ground'. This was properly grounded via my homebrew FM tuner, until I unplugged it. Not dangerous, but an unpleasant tingle.
Often no tests are required for the CE mark. The regulations are that no appliance may be brought into traffic in the EU without the CE mark, when certain laws or regulations apply to the product. So the manufacturers are more or less forced to mark those products.
By CE marking a product the manufacturer guarantees that the product complies with all applicable norms, laws and regulations in the EU. The so marked products usually come with a declaration of conformity. As long as nothing happens, that is all.
Should something happen, e. g. should harm come to a person or her belongings, the person can sue the manufacturer and then he has to prove the conformity. That may involve tests. In some cases a theoretical proof is acceptable, in others following common practice in the product design is considered sufficient.
Some products must not be marked at all. And the responsibility for the marking lies with the person who puts a product into circulation which is not always the manufacturer. For practical reasons manufacturers assume that responsibility, because e. g. a shop owner may not be in a position to assess the necessity for a CE mark or whether a product complies.
By CE marking a product the manufacturer guarantees that the product complies with all applicable norms, laws and regulations in the EU. The so marked products usually come with a declaration of conformity. As long as nothing happens, that is all.
Should something happen, e. g. should harm come to a person or her belongings, the person can sue the manufacturer and then he has to prove the conformity. That may involve tests. In some cases a theoretical proof is acceptable, in others following common practice in the product design is considered sufficient.
Some products must not be marked at all. And the responsibility for the marking lies with the person who puts a product into circulation which is not always the manufacturer. For practical reasons manufacturers assume that responsibility, because e. g. a shop owner may not be in a position to assess the necessity for a CE mark or whether a product complies.
Dito.
But here my interpretation was different. But I guess, if you locked two nuts together...
Whether it is achieved via testing or design, the intention of CE marking is that it shows that the product satisfies the regulations. The relevant ones here are probably the Low Voltage Directive (it won't electrocute you or burn your house down) and the EMC Directive (it won't interfere too much or react badly to reasonable incoming interference). CE marks often appear on goods which clearly don't meet the spirit, and probably the letter too, of the Directives. In some cases it is likely that the design would meet the requirements, if only the manufacturer did not deliberately omit certain components. The fact that this goes on shows that some factories are simply putting two fingers up to CE marking.
What is more interesting about this particular case is that the filter is injecting noise on the mains into the interconnect shield, especially if another box in the system is earthed. This is most likely these days if there is a desktop PC connected.
I wonder if it would be worth bonding all of the system cases together with braided cable.
I wonder if it would be worth bonding all of the system cases together with braided cable.
Cheap PC power supplies (inevitably made in China for incredibly little cost) omit and downgrade anti-EMI filter components as a matter of policy. In all the 'inexpensive' unit's I've serviced (and upgraded out of a sense of duty if nothing else) the design was probably good but, to save a few cents per unit, the manufacturers omitted 80% of the vital filtering stage. Quite often they use one massively under-specified generic toroid inductor with a few windings of hook-up cable around them instead of two proper split devices. Pretty horrible stuff when it comes to dumping massive RF frequencies into your home's power circuit. Always buy reputable brands when it comes to switch mode supplies!
Perhaps this article (http://audiosystemsgroup.com/SurgeXPowerGround.pdf) is of interest to you. The author calls it "local bonding" which can be done for equipment in close proximity sharing the same PE conductor of the outlet.
Even that is not sufficient. I bought a Philips DVD player, which radiated a lot of RF hash even when switched "off"; touching the metal case made it worse. It was, of course, made in China. Philips "customer support" were no help so I took it back to the shop and got a refund. I bought a Sony instead and that is fine. It injects mains into ground connections (which at least shows it has some filter capacitors present) but it does not radiate much.
Try to find either a brand who put the fear of God into their Chinese suppliers so they build things right, or a unit with a genuine on/off switch so off really does mean off.
20+ years ago I found that IBM was stripping EMC components from PCs for the UK market compared with PCs for the USA. This was back when PCs cost the price of a small car. I had to use Compaqs to find a machine near FCC Class B limits before we even put the card to be tested in.
Now a PC psu costs from UKP10, so quality has to have slumped since then
I read that the original AT PSUs were capable of supplying over 200-300W of power - whatever that figure was but they were capable of supplying 1) whatever their labels claimed and 2) much more power than whatever computers drew then.
Then came the era of cheap "400W" PSUs that exploded when the user tried to install a "power-hungry" CPU + graphics card, total power draw - <200W.
Last edited:
- Status
- Not open for further replies.