UK Mains

Remove the router, sky box, turntable and external PSU, NAS, Apple TV box etc from your Hifi-spur unless they are feb by a linear PSU.

Most will probably be fed by an SMPS (switch mode power supply) which about negates the advantages you have from having a separate spur in the first place.

This move will free up enough wall sockets to feed the linear PSU's you have.

+1

This is the best solution.

I can confirm from my experience that the ND5 XS is significantly affected by RFI (some of the fine detail in the musical presentation seems to be masked in the presence of RFI), and so is likely to be being affected by the switched mode power supplies for the other items you have.

Musicworks Megablock here.

+2 definitely. SMPSs should be kept separate from the dedicated supply. I even have my Radikal on the consumer circuit and it's very easy to demonstrate the effect it has when plugged into the dedicated spur. Whilst playing a CD, when I had a CDP, unplugging the Radikal made a significant difference in SQ. I was skeptical until it was demonstrated. With all the computer kit you have the cumulative effect is likely to be even worse. 

Good luck.

Steve

I found this very good. I bought mine from Grahams before going on to the Hydra and then individual MK sockets for the Powerlines.

+3 to all the above - however given the dilemma of having to share a common supply with the various SMPS for the foreseeable future,  it might be worth considering a Tacima CS929 6-way power strip.  It has a D&C choke & cap filter & is reputed to reduce the effects of mains noise & RFI.

I would also try it for powering the SMPS's,  it might help block their switching noise from getting onto the power line at source.   

Ok the Tacima product for around £30 sounds an interesting option for the SMPS items.

As suggested, I'll use the dedicated spur just for the ND5XS and ATC amps and speakers etc...

Presumably all the SMPS stuff can share a mains distribution block - NAS, Apple TV, Sky HD box, Apple Time Capsule (router), Draytek Vigor modem (for router wifi) and turntable PSU. 

The question now I suppose is whether the distribution block will be okay plugged into the same separate mains spur or whether I'll need to run a lengthy cable from another room because all the sockets in the lounge are on the dedicated spur.

I'm just about to add a Chord Hugo to the ND5XS and wonder if this too is destined for the distribution block (due to it running off internal batteries on constant charge) or whether it will be okay to share the dedicated spur (along with the Naim and ATC equipment)

I look forward to hearing what others suggest because clearly I hadn't realised the impact on Linear power supplies being effected by SMPS kit. Just goes to show the benefit of input from others more 'aware' and knowledgeable in these areas - I had mistakenly thought that as long as all the hifi kit was on a separate spur, all would be good - clearly not 

The problem is you would be putting components on your dedicated spur that you are trying to avoid on the domestic supply. When fitting a dedicated supply for the HiFi it's important to also have sockets from the domestic circuit for just these situations.

Good luck.

Mr Frog

the thing to do is try various permutations of plug configurations and see if you can hear a difference. If you can't then no need to worry. I have to say ive tried disconnecting all other kit apart from my hifi and could not tell any difference. However i don't have a dedicated spur so maybe this is why.

Graeme

Maybe get an electrician to put a socket in the lounge from the domestic supply and plug the distribution block into that - powering all the SMPS equipment. Whilst leaving all the linear power supply equipment on the dedicated separate hifi spur?

That would be your best bet to keep the 'HiFi' and SMPS/computers on separate supplies.

I think so Graeme. A dedicated supply is a very cost effective upgrade IMO.

Steve

Mr Frog,

Is your 'spur' in fact a dedicated ring main for the lounge only, rather than actually an unfused spur?  If it's really an unfused spur, then you'll be limited to one double socket.

It's easy (and relatively cheap) to put an unfused spur through the wall from a socket in another room on the normal domestic ring main.  This unfused spur can have support one double socket.  The ideal is to connect everything using a SMPS to this unfused spur using extension leads with ferrite beads on them (note, this includes devices like DVD/BluRay players, computers, phone chargers and televisions, not just wall warts).

Graeme is also right, test whether it's worth doing:  Some SMPS are actually properly designed and don't emit too much RFI.  One easy way to asses the benefit is to temporarily run an extension lead (or two as needed!) from the adjacent room and plug all the switched mode stuff into that - you'll then see how much benefit you get before you pay to alter the house wiring!

It's probably best to get a professional sparky to do the work, but I believe that if you really know what you're doing an unfused spur is still legal as a DIY job (altering a ring main isn't)*

*  N.B.  This is an oversimplification of the current wiring regulations; for a DIY job you must know them exactly.

How does this work to reduce rubbish on the mains. All rings, spurs come back to the same pair of cables feeding the property. Surely the rubbish will get onto all rings, no matter which ring is generating it.

Graeme

If I may answer this:

The connection from the mains supply to the consumer unit has extremely low resistance.

The ring main from the consumer unit to the sockets is made from 2.5mm copper twin & earth, this has a much higher resistance.

A switched mode power supply plugged into the ring main 'sees' the resistance of the ring main in front of the bus bars of the consumer unit.

The Consumer unit is therefore much more closely locked to the incoming mains supply than it is being driven by the RFI from SMPS on a different ring main circuit.  This ratio of resistances effectively isolates each ring main from the others.

P.S. For the pedants: I'm well aware that impedance is the correct term but more people understand the term resistance and the principle still applies.

OK i get what your saying but if the hash is on the ring and current flows around the ring from the dis board, then would the hash not take the path of least resistance and drain down the neutral.

Oh maybe that's what your saying. It travels around the ring and then once it gets back to the dis board it drains down the neutral of the incoming feed down to earth rather than propagate onto another ring.

graeme

As it flows round the ring, it's impeded by the resistance of the cable from which the ring is made.  When it meets the CU, the CU is being held at the incoming mains voltage by the low resistance of the mains supply cables.

Consider it this way, using a mechanical analogy...

Looking at the consumer unit, the mains input is low resistance, that's analogous to a rigid mechanical connection (an RSJ for instance) joining it to the substation.  The ring main is a higher resistance, analogous to a much more flexible mechanical connection (say, like a piece of copper cable).  The SMPS is on the other end of the copper cable pulling and pushing trying to bend the RSJ using it's connection through the copper cable.

Result, nothing, the RSJ stays put; the copper cable isn't rigid enough to bend it.

The other ring mains are connected to the RSJ, so none of the vibration from the SMPS gets through the RSJ and onto the other connections.

Does that analogy  help?

Think i prefer mine actually

joking

That makes perfect sense, thanks Huge for taking the time to put it in easy terms.

Graeme

You are most welcome.

Electricity isn't always easy to understand intuitively - you can't see it, smell it or hear it (directly at least), and for mains voltages you definitely don't want to feel it!

Huge - I like your illustration, but does it not fall down because we actually are talking impedance and not resistance? RF hash will travel on the edge of the cable and will not significantly influenced by the diameter of the copper. The  impedance it sees will be mostly determined by the dielectric of the insulation and capacitive coupling to earth as well as the tuned circuit that is your domestic ring mains is.

Therefore I don't think you can assume a low imepedance path at 50Hz (ie the ring main to consumer unit - behaves the same way as low impedance path at say >3MHz. Its 'path of least resistance' will be different.

Therefore RFI will quite happily permeate along a separate spur from the consumer unit - in fact RFI will permeate potentially as far as your substation slowly being filtered - along with everyone else's RFI. This is why PME supplies are inherently more noisy.

Therefore if you have RFI on your mains then a separate spur or supply from the consumer will not necessarily provide a cure - unless the rings/spurs are separately filtered at the consumer unit.

Just my thoughts - and based loosely on my experience

Simon

Simon, I am well aware that the analogy is weak and greatly simplified, but I didn't think that going into skin effects, impedance matching, Kirchof's laws etc. would have been particularly useful to the intended recipient of the message.  I think the best answer for >3Mhz is probably still ferrites or some of the other options discussed such as CTE isolation transformers.

At HF with multiple devices plugged in, a lot of complex stuff can happen between a socket and the CU as well as between the CU and the substation!

The analogy was just to give the (highly simplified) general idea in an easily consumed form.

Huge I sympathize with the challenge of simplified models  - however the reason I raise it is because having a separate spur from the consumer unit will not necessarily mitigate RFI/hash on a separate ring main.

So someone investing in a separate spur for their audio could be disappointed with the end result if  they have a separate noisy ring main in their house that has remained unchecked - or even with Ethernet over mains from a close neighbour's ring main.

I have definitely been here.

RFI literally gets everywhere if we are not careful... as I am sure you and I know. And you are right - it is complex which can make it appear largely 'unpredictable'

Simon

What does this mean in physical terms. Do you get special filtered consumer units?

Graeme - I have no idea in practical terms - but that is what you would need to do to stop it in my humble opinion. I have just googled and I read there are quite specialised filters for cables in and out of consumer boards - and consumer board to meter.

The best remedy I suggest is keep well clear of noisy; devices / lamps / solar inverters /devices that RF modulate the mains etc

If you are still  plagued then go for a dedicated mains treatment system for your audio devices,

Google also tells me In consumer space - this pheromone is called DE or Dirty Electricity so I understand - and DE covers noise frequencies from RF down to low frequencies such as 60kHz,