On 3/7/2016 11:56 PM, Jim Brown wrote: > Actually, switch-mode power supplies run in the range of 10-50 kHz, > NOT at HF. It is the unfiltered high order harmonics that we hear at HF. Yep... I was wrong, you are right. Not fundamental, but harmonics. I should stop writing in the middle of all-nighters.
> But the IF and mixing products are NOT at HF. Trash from WiFi systems > is produced by the baseband digital electronics and the switch mode > power supplies, NOT from any part of the RF system. > I agree that the much more likely candidate is the Ethernet switching. But the standard Ethernet signaling is differential... PoE could definitely be a culprit, since one of the methods involves tapping the Ethernet pulse transformer for a voltage source. But, whatever the SOURCE of the trash, it is generated inside the device and inadvertently being radiated over the UTP. Bottom Line: There is no shortage of potential places inside a typical Wireless Switch/Router where RF issues can be created... especially when their designs are targeted to meet a price point rather than comply with PART 15. I'm a bigger fan of eliminating the source than I am of mitigating the effects... at least these days I am. There was a time when I enjoyed "making stuff work". I confess, I don't design wireless data radios, so I have no idea what the IF/oscillator freqs are. I was simply stating that wireless data radio is a radio. >> Perhaps I should have been more precise... logical ground and a chassis >> shield. Although, there ARE in fact many electronic devices where the >> device itself has an isolated logical "ground" and the entire device is >> located WITHIN a totally separate shield from end to end essentially >> floating the devices on each end and the signals inside their own >> "space". > > As Ott has observed, there are holes in that design logic. Don't shoot the messenger. Tell the DoD, NASA, NSA, and a host of private companies to include Cisco, etc. I'm just telling you what's out there. Ott wrote a lot... are there specific references you'd care to share? I'd be interested to read some of his theories with respect to this subject. Online refs if possible; I've exceeded my book budget for the month... heck for the year. ;-) > I don't get "drain to ground." I apologize... perhaps I stated it wrong. Many shielded cable types (not typically STP cable) come with a "drain" wire to make it easier to terminate the shield... not used a lot in data cable... but I sometimes use the term to refer to a conductor that connects to a cable connector housing or to the cable shield whose purpose is to assist in grounding the cable shield or creating continuity between the cable shield and the shielded termination. In the case of the "plastic crap" I described above without additional shielding added and internal connection to the cable connector somehow, the alternative solution of trying to make the STP cable a less efficient radiator involves using "drain" conductors to try and shunt the RF to ground before it can radiate. > It IS possible to make unshielded equipment RF-quiet with proper > circuit layout and construction. Yes, of course. But MOST manufacturers have "discovered" that they can design to a price-point and ignore Part 15, et al. because of the unlikelihood that the FCC will pursue them for their massive violations (think of the number of non-compliant consumer data devices in the filed and sold each year). Sad really, because it doesn't take all that much more time to design and manufacture a proper board. One would think that a couple hundred thousand HAMs making complaints against a massive number of obviously non-compliant crap consumer electronics might generate a response... Needs to be coordinated. <PCB design criteria was here> We agree on that entirely. > Not quite. What I've observed is that MOST of the noise we hear on HF > is common mode current on cables connected to the device, NOT due to > the device being poorly shielded. Yes, SOME noise is radiated directly > from poorly shielded and/or equipment with the PCB layout issues I > noted, but that part of the noise is rarely dominant because the > antennas inside the box are too small to be efficient at HF, whereas > the antennas carrying the common mode current (the interconnecting > cables are usually much longer (including path to other equipment, > power supply, facilities ground, etc.) Gotcha... Having never needed to chase RFI to HF in my own home/office, I'll defer to your obviously superior level of direct experience. I think it might be interesting to actually have the harmful RFI tracked to specific causes in a large range of device models to see if the sources cluster around a particular design "flaw"/choice(s). > I would NEVER waste my time shielding a poorly shielded box that was > bad enough for me to hear. I would dump it and buy something without > that problem. Absolutely... Today, I would agree with you, but there was a time when I would not have had the economic luxury of "dumping" a piece of equipment. It's nice to know that you CAN do something because you've done it before. Have a good day, Jim! Clay, KG5LKV ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html Message delivered to [email protected]
