[time-nuts] timing properties of "spectru spreading", clocks

[Alan Melia]

Hi David, thanks for that succinct explanation. The SSC is a "fudge" to 
defeat the measurement technique, which has nothing to do with the 
psychological effect of the interference, or the BER effect on the 
legitimate user. In fact it is often easier for the legitimate user of 
the frequency to avoid a fixed clock harmonic than deal with one 
spluging all of the band. It legitimises poor design!

Alan Melia G3NYK
RSGB EMC Committee mmember

On 21/05/2019 18:00, Dave B via time-nuts wrote:
> Date: Mon, 20 May 2019 20:41:51 -0500
> From: "Bill Byrom" <time at radio.sent.com>
> To: time-nuts at lists.febo.com
> Subject: Re: [time-nuts] timing properties of "spectru spreading"
> 	clocks
> Message-ID: <5e9c8b54-c30e-4e03-aba2-d4d96e5d8345 at www.fastmail.com>
> Content-Type: text/plain;charset=utf-8
> 
> SSC (spread spectrum clocking) is widely used in both commercial (servers etc.) and consumer products. The big advantage is at harmonics of the spread clock, where the higher frequency energy can more easily escape the product case, shielding, and cables. CISPR (EU) standards in general specify a receiver bandwidth of 120 kHz for signals below 1 GHz and a receiver bandwidth of 1 MHz for signals above 1 GHz.
> 
> <Snipped>
> 
> ------------------------------------------------------------------------
> 
> Hi.
> 
> "Spread Spectrum" clock's were designed as a quick and dirty fix for out
> of specification emissions from all sorts of digital based products (and
> some non digital!)
> 
> It does *not* reduce (or increase) the level of leakage, it just spreads
> it over a wider spectrum.   (Never mind the quality, feel the width!)
> 
> It was specifically designed to "Fool" the Quasi Peak (QP) detectors
> used in the EMC Emissions test receivers, so that a lower "Under the
> limit" reading is produced.
> 
> Such emissions tests are usually done in two stages.  A fast pre-scan
> with a Peak detector.  If the product passes that, job done.   If any
> signals extend over the Peak threshold limit, then those frequencies are
> again tested with the QP detector, that is quite slow to respond.   If
> the signal is still over the limit, it's a fail.   If it's now under the
> QP limit (as it often is for SS signals) it's a pass.
> 
> The limits are there to protect broadcast signals, not to limit
> interference to other systems, or small signal communications links.
> 
> In reality, it's a "bean counter" induced fix to poor technical design.
> "Short term cost saving".
> 
> Just look up the history of the QP detector, it's utterly inappropriate
> for what it's now being used for.  It's original purpose was to get some
> sort of quantitative and repeatable measure, for pulse interference to
> AM broadcast signals, such as that experienced from poorly suppressed
> spark ignition systems in passing vehicles.   That of course go past and
> away after a short spell.
> 
> Declaration:
> 
> I work in the EMC "Industry".
> 
> I'm also a government licensed Radio Amateur, who's utterly P'd off with
> the RF pollution from poorly designed and/or implemented digital systems
> and switched mode power converters.
> 
> Such levels of wide band interference is already starting to bite the
> hand that feeds it.  Some LED lighting systems can and do wipe out
> terrestrial DAB reception!   Of course, it's not the LED's themselves,
> but the SMPS based "drivers".
> 
> The EMC rules are lax, and they are not monitored, let alone enforced in
> the marketplace.   A lot of the industry is fed up with it all too.
> 
> The soap box is creaking.
> 
> 73.
> 
> Dave B G0WBX.
> 
> 
> 
> 

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus