Ethernet & Ferrite Chokes

If you search you'll find plenty of advice.

If you have a problem with radiated RFI then you should put ferrites on any cables attached to your audio kit.

I have ferrite on the ends of all Ethernet cables

NDX-Switch,   Switch-NAS  &   Switch-Hub (router/switch)

These are all Cat7 (screened) - not sure what/how/if ferrite works with screened cables,  but I cannot detect anything detrimental.

I also have ferrite on all the cables feed or output from the various  SMPS's, & the phone leads to router & phone.

Yes it might be overkill,  but I have yet to hear or see anything detrimental.  

Mike, chokes work well twisted leads and shields, ie where there is a common mode RF current. 

Simon

I use them on all Ethernet cables. I move them about from time to time and none of it seems to make any difference in my system. So I leave 'em on, just in case. They're so cheap in any case. I see no harm in it.

IME, the positive effects of ferrite chokes are subtle in most use cases, but when it came to my turntable, they turned out to be essential.

My turntable sits atop my "brawn" rack, and my Superline phono stage used to sit immediately below it, connected by a .5m RCA interconnect made by Auditorium 23. I moved the Superline to an open shelf on my "brains" rack, and connected using a 1m version of the exact same cable. All was fine for a couple of days, and then to my surprise, an annoying hum started coming through my speakers. I powered everything off and back on, and all was well. A couple of days later, the hum was back again. I stupidly allowed this cycle to repeat one more time before remembering that I had forgotten to move the ferrite chokes off my old RCA cable and on to the new one. Did so, and now six weeks have passed with no more RF "breakthroughs". Problem solved.

ATB.

Hook

Hook, are you trying to make me bald?  I'm going to be scratching my head for weeks trying to work out possible ways how that can happen!

Hi Huge -

Going bald? Ha -- feel free to join the club pal!    Here's a few more details I left out for brevity's sake...

I had issues with RF on the Superline when I first bought it five years ago, I found that it sounded best with no capacitance plug installed, but I would occasionally hear Radio Moscow through the speakers  1nF was too much -- it really dumbed down the sound -- so I ordered a custom 470pF cap plug, and used it for a long time. Had no further problems with RF, and at some point, just out of curiosity, I removed the cap plug. Interestingly enough, I had no further breakthroughs.  Never figured out why, but now I believe that by the time I removed the cap plug, I had already added ferrite chokes to the RCA interconnect.

When I started hearing this new hum through my speakers, my first thought was a ground loop. But that didn't make any sense, since ground loops just don't appear out of nowhere -- they are either there when we power on a system, or they are not. My second thought was a cable or Superline fault -- I had recently changed carts from a DV to an EMT, and had my SL converted to a low-gain "E" version. So, had the chokes not eliminated the noise, my next steps were to try different interconnects, and then send my SL back to AVOptions for a checkup.  Also, just to make sure, I disconnected my NDS after the second or third breakthrough occurred, leaving only a very simple, single source vinyl setup with no possibility of any grounding issues. But the hum still returned after a couple of days!

So yes, this was a strange one. Each time it happened, the noise sounded a bit different, both in frequency and in volume. Besides being the only explanation I can think of, and because this interference went away after the chokes went on, I am now pretty much convinced that RF was the attributable cause.  But I would welcome any other theories or explanations you can come up with!

ATB.

Hook

Hi Hook,

Well, to misquote a great Dane: "There are more things in heaven and earth Hook, than are dreamt in my philosophy."

I have a germ of an idea, but I'm not sure if it could actually occur!

I need to think - damn! more head scratching.  I think the risk now includes splinters as well as baldness.

Hook interesting - specifically radio break through is often not the same as RFI - and usually have different causes. Radio breakthrough is interesting as for AM to decode into audio there would be a diode function of some sort in the path of the RF current otherwise it would appear  as a  solid tone/frequency. It is this rectified RF signal one hears as music/speech AM radio breakthrough. FM   radio breakthrough is back to sums and differences again. (and with speech / music / Taxis the sums and differences will be modulated and hence potentially audible)

Regular RFI from clocks and oscillators usually affects feedback and causes intermodulation distortion (sums and differences between frequencies) and can create a soup of spurious HF signals (hash). Again computer / signal modulation might be audible as a soft whine or buzz if one of more of the created intemodulated frequencies is in the audio band. But often the created frequencies are outside the audio band but are still affecting the audio electronics such as feedback systems and this is when we hear things like 'hardening' or 'loss of detail' with RFI.

Radio pickup in phono cartridge and their leads is not that uncommon as effectively the cartridge and lead appears as a loaded antenna to the phono amp. Personally I think its inexcusable to design a phono stage to allow it to act as RF front end in this way - but that's just me - as many - but not all - phono amps I have come across do this.

Braiding or effectively shielding the phono signal cables is usually the best cure - with the former being preferable in my experience as a shield itself can act as an antenna is not accurately designed.

Simon

Hi Simon,

What interests me about this isn't so much the breakthrough / AM decode, (although interesting in itself; at low current, the non-linearity of a transistor base-emitter junction can act as an inefficient AM detector).

What particularly interests me here is the LF tone that builds up over a couple of days, that it takes time to re-appear, varies in frequency and volume, and is prevented by the chokes.  It sounds rather like a charge accumulation problem but that's odd in a phono-stage.  Do you know if the Superline uses MOSFETs anywhere?

What really gets me is the link to the RF chokes.

Do they suppress a RF parasitic oscillation that's being rectified to obtain the charge?

Is there a possibility that a HF oscillation is being stimulated by the RFI?

I've met charge accumulation problems before in instrumentation amps, but never in audio.

Exactly and this is usually what is causing the rectification.

To your point on reappearing interference over a few days is probably to my mind caused  by minute changes in the contact connections of plugs and sockets and annoyingly dry joints - and as such at a small signal level (this take me back   ) that junction will appear complex with reactance and resistance and therefore if this varies subtly  - say through oxidation - or mechanical movement - then the tuned circuit parameters of that contact will vary.

As far HF oscillation / instability being triggered by RFI - it is possible but would be a very very poor circuit as it is essentially unstable.

RF chokes add inductance - and inductance reactance is directly proportional to frequency - and so it is this reactance that resists the RF current and turns it into heat.

In some environments ( and not home audio systems thank goodness) RF chokes can actually get warm or even become saturated.

Simon

So we have a tuned circuit arising from the contact characteristics selecting an RFI carrier frequency, intermodulation giving the LF tone, and non-linearity giving the AM detector; all amplified to audible levels by the gain of a phono amp.

Small changes in the tuned circuit will be frequency down shifted giving relatively larger changes in the tones in audio band by the intermodulation and will account for the LF frequency variance.

That all sounds feasible and worth testing.  Hook, have you got a 'scope?  (OK, just kidding ).

As for the potentially unstable amp, sort of like a power amp without a Zobel network?

I believe it's the magnetic complex reluctance that is the magnetic 'lossy' property that causes the heating.  The pure inductive reactance is lossless.

(And hence the frequency dependence of ferrite properties)

Wow...you two are talking some serious science!  Am heading out now for a day of customer meetings, and will read your posts more carefully this evening.

Did I mention that the RCA cable was pseudo-balanced, with the grounded end connected to the Superline? Given I also have a digital source (NDS), I was advised that the SL end could better deal with the noise being collected by the shield. Perhaps not?

Thanks guys -- more later.  And apologies to the OP for the thread jack!

ATB.

Hook

I stand corrected

Indeed - and remember the story of the person who heard radio voices in their head .. in turned out they were near a transmitter - and some tooth decay was actually forming a diode junction and converting an AM carrier into an audio signal that they could hear....

Who needs silicon