Sigh...That's what seems to be the consensus at this moment. I Agree, the efficiency is a nice benchmark, but when your listening pleasure requires dBuV not dBmV - you have to draw a line in the sand somewhere.
The "ring" as some call it, others call it Hash and to me it's simple noise, I have had to switch back (SIC) to Linear because of the support, once the SPS has left the factory, is a lot like finding Elvis...
Non existent...
At least to specific types of control - or any feedback from the user trying to interweave their shack with one - only to have to remove said device because of the noises it buries in the QRM when you're QRP and need to be.
No one ever spoke up about the use of the various Diode constructions, the older PN, versus Schottky, or the materials used, Germanium, Silicon, or the Carbide swaps.
I just wanted to point out something that, if forgotten, we'll never be able to get it back...
That is the inherited "leaky" effects of the older generation of Diode and the intrinsic layering used in it, seems to offer a sensible way to cushion the effects of the switching "clickety clack" - the older LED's even had this effect, junction capacitance, caused by the layers and from the thickness of the layers. Seems we don't have this big of a problem when the older generational parts are used versus the Schottky and N-less possibilities caused by them.
Ok, this may seem more like a word pun, but I need to express the facts of remembering how we got here, not just the Why..
Some of my methods I tried included a "common mode choke" you could buy at a local Car Radio shop - a simple coil cap combo used to help prevent ground loops - using a common mode choke to help with this, but I've found that by placing an low-values- high wattage inductive (Read:Wirewound) resistors ACROSS the output terminals - to be one of the better methods of reducing the hash but it comes at the price of heat and constant loading on the supply itself just to force the ripple and load "curves" to match and intersect to provide some less noise, it's not perfect, but to me it has worked - as a last resort.
All I was doing was changing the location of the hash to a different range of spectrum.
So yes, in the PDF they even admit they shed noise, and the many test tables that they placed it on, got turned on them because of the technology, being so efficient - may have left left others in the dust, but the EMI is still there.
So before I go back into the bushes and hide out again, I just wanted to make some observations from the Street corner of Nowhere, and Here, that to get somewhere, we need to remember whom may be using the next vehicle everyone wants - may wind up drowning out the ability to hear the lesser of many evils and one being the Radio hobby itself, getting left trampled in the dust pile of ores - no one else wants..
Fans may have gotten noisy, the Hash - was it always there? Or only when the OEM supply parts the board was made up with - got used up a new effort had to be made. Forcing them to find another supply route?
Ok, Did they simply use new cross-referenced parts and so the chain can continue - the keyword here is "chain" - the parts themselves can be the weakest link in all of this.
Why the concern?
- ..- no one ever (or so it seems) bothered to research the parts within the layout, if their internal construction - caused the extra noise - could such design been reworked to reduce the problem and maintain integrity?
- - their integral parts of the device that makes the devices work, has it caused more harm to the design that, in it's replacement part, does the unit as a whole become less effective at being desired as efficient? (Reading the feedback here - this seems to be it's current direction)
- - the effort to try the various parts never made it to the Datasheets as a warning - it seems to have been overlooked.
- The Bane of all this is the traits of SPS versus Linear, having to use one over another seems to be a throwback - backwards as a necessity - to lower the noise floor at the cost of efficiency.
It just seems like poor research.
