Noise and DC on mains and new radial

KRM posted:

Huge posted:

KRM posted:

The spike was fitted by an electrician.

The advice to fit a spike came from Naim (via the dealer), not from Russ Andrews, although I did ask him if it would be ok and he said it would be "fine" as long as it was fitted by a qualified electrician.

The advice from Naim was:

"Use a dedicated earth for the sockets. Use the house earth to protect the cable run, but don’t connect it at the sockets. Get the sockets’ earth from a totally separate rod outside the house. The rod should be as close as possible to the sockets, but needs to be somewhere where the ground doesn’t dry out too much."

Keith

 

If you are doing this then you MUST have an RCD in ALL circuits; without this the arrangement just isn't safe.

 

 

(As a very minor point, how are you running the extra earth wires as this will affect the audio & RFI performance?  Also note that they must be green/yellow insulated.)

That's why I called Russ Andrews and they said it was "fine".

The Consumer Unit is sold with a socket labelled for "not protected by RCD - for hifi use only". 

Keith

Keith, did you tell them you were changing over to a TT earthing arrangement?

The necessity for an RCD is shown below (I've assumed a normal 32A MCB is in place, if a higher current limit - 100A switch? is involved the situation is worse)

If an equipment fault connects the 230V line to the equipment case...

With an RCD:  Then at 30mA residual current (e.g. by earth leakage) the RCD will trip.  The RCD will still trip even if the earth spike has 7,600Ω impedance.

With no RCD:  You are relying on the current through the MCB to trip it and disconnect the mains voltage from the case.  This will require more than 32A to flow through MCB and out through the earth spike.  So the MCB will trip only if the earth spike has an impedance of less than 7Ω.

If the ground around the earth spike were to dry out, the impedance could rise above 7Ω.

Any thoughts on SmartMeters transmitting digital gunk?

John.

Hi Huge,

Yes I did. I asked the dealer who asked Naim who emailed a section of a white paper. I then employed an electrician and consulted with Russ Andrews to make sure the consumer unit would be compatible with a spike.

Belt and braces, you would have thought!

Keith

Then I just hope you don't get a line / earth fault when the ground's dry or the cases of your HiFi are likely to be come live.

J.N. posted:

Any thoughts on SmartMeters transmitting digital gunk?

John.

No should be fine.. many smart meters  and home automation now use Zigbee which is an IEEE 802.15.4 specification for low power personal networks. Zigbee operates in Europe on the 2.4 GHz ISM (wifi band) and the 868MHz ISM bands and is extremely low power. Naim are also using Zigbee as the remote control frequency on the new Uniti models. There is also an equivalent low power wide area network in development for urban and semi- urban areas called LoRa... which may be used shortly by utility telemetry (remote metering/smart metering) in the U.K. which won't rely on a separate back haul such as GSM or  broadband.

Huge in the U.K., the maximum ground impedance  for a TT electrode is 200 ohms.. anything higher is deemed unstable. To get comfortably below 200 ohms where the ground has limited conductivity then the electrician should use multiple earth electrodes. 

The consideration difference between TT and PME, is that with TT the max fault disconnection time is 200mS where as with PME I believe it's 400mS, but a RCD is not specifically needed if the earth loop reading is very low (very conductive earth) and in which case you could use a MCB. So it is very much down to installation and environment, but in most regular TT installation an RCD would be used, but doesn't have to be used with a TT system.

Simon, yes indeed, that's why an RCD is needed with a TT system.

To be clear an RCD  it is not required in all circumstances with a TT installation, BS7671 17th Edition  NOTE 1: in 411.5.2

200Ω max, let's say 20Ω ground impedance so we're nowhere near considering the worst case.

Short line to earth:  230V @ 20Ω = 11.5A, so the MCB won't trip and the earth line will be live, as will be everything connected to it.

I personally don't consider that safe, even if the regulations allow it!

Ok. So after absorbing all the comments above (thank you, All) I plan to ask my Sparks to take a radial off the tails of my mains into a MK Amendment 3 CSU. Into latter he'll put a MK 32amp Type C MCB. From there he'll go into a balanced power supply and from there to a double, unswitched MK Logic wall socket. (Should he star earth inside the socket?) He'll use 10mm2 Prysmian cable throughout. I'll ask him to earth as he sees fit; but, is there a choice to run the earth to either the dedicated CSU, the main household CSU or meter board? If there is, which is best?

There should be an earth terminal near your meter, ask your electrician to connect the new consumer unit to that directly, rather than via your main consumer unit, to give greater isolation. (Assuming, that is, that you're not going to persuade him to stick an earth spike in your garden!)

Great, thanks Chris. Considering the above, I'm going to leave spikes to the vampire slayers!

Wooden ones aren't much good as earth spikes anyway!

Mains measure 234v so I'll not be getting a BPS with adjustable taps.

Bt, Airlink offer 1) a standard balanced power supply and 2) a conditioning balanced power supply :

'The Conditioning Balanced Power Supply has the addition of a three-stage EMI suppression filter between the line input and the transformer, designed to suppress either common and differential mode noise or interference.'

This refers to mains high frequency noise, I believe.

I remember from past posts a red flag being waved at 'conditioners' with people saying it had resulted in their music becoming less dynamic. Is anyone able to comment on whether this would be a danger in the instance of the CBPS above?

 Thanks.

Hi Eagle, I cannot comment on the merits of the conditioning circuit. What I can say is that my BPS does not have the conditioning circuit. The Engineering Director of Airlink recommended the one that I went with.

Best regards, FT

Thanks FT. The MD recommended the conditioning unit to me but I think I'm going to stick with the standard version (5000 va) in light of no reassurances about music becoming less dynamic. Less expensive too.

SimoninSuffolk I vaguely remember reading comments you've made in the past about high frequency pollution. Sorry if I'm imagining it but any thoughts on this one? 

I'd just like to say, as OP, that I've presented a distillate of all the above to my fully qualified sparks and anyone thinking along the same lines in terms of improvements to mains power should do the same. Unless they're qualified not to, obviously. Anything that gets done will be done by him and will, therefore, be entirely safe and legal. Having stuck a pair of scissors into a socket aged 7 and been blown across the room, I tremble when replacing a lightbulb.  

So were there any meaningful measurements of DC offsets and noise on the mains? Or are we still in the land of speculation and presumption?

My mains measured 234V and occasionally fluctuated to 233V. We didn't check DC offset etc. - because I forgot all about it. I was focused on the mains measurement and which BPS I'd therefore need. Actually I didn't know you could even measure it. I just assumed it's there after finding that turning off all the breakers in the CSU, save for the hifi rooms wall sockets, resulted in less harsh leading edges. Also, my 250DR hums and the hum can become distractingly loud from time to time. I'm hoping the stable power delivery of the BPS will sort this. I'll report back as soon as install is complete. In the scheme of things, this entire undertaking is not expensive - a few quid more than a single Powerline. I'm hoping it turns out to be a no-brainer.

..I should say 'stable and 230v-accurate power delivery..'

I think the issue was dc and noise... small movements of main AC voltage is irrelevant

Quick question for the RFI officionados - should the cable feeding my dedicated radial socket be situated a min distance from the cable feeding a socket on the house mains? (My 10mm2 is unshielded Prysmian) I was planning on siting the dedicated socket a few inches above the house ring socket which would bring the cables to within a few inches of each other. Will RFI, or anything else, jump from house ring onto dedicated radial if they're too close at any point? Instal is on Monday

Thanks.

Ian

[@mention:26908937791111692] - RFI is Radio Frequency Interference - what do you think.....?  Greater separation would be good, IMO, but earthing and/or screening are the real 'fix' to RFI. If RFI is really a problem you have...?

All Naim power amps and power supplies hum - some more than others - but the hum isnt through the speakers generally. Most live with it.... YMMV clearly.

You say a PowerLine is not expensive....? Wow.

Hum from a transformer isn't due to RFI.