[Nottingham] Apple Macs (and WiFi interference)

James Moore jmthelostpacket at googlemail.com
Fri Nov 16 17:53:50 UTC 2012


On 16/11/2012 17:33, Martin wrote:
> In brief-ish!
>
> On 16/11/12 17:12, James Moore wrote:
>> On 14/11/2012 20:50, Martin wrote:
> [---]
>>> Is this where the RaspberryPis overwhelm all users to save the Internet
>>> and computing as it should be?! ;-)
>> I've yet to even see one of these gadgets, let alone get my stickies on
>> one. Are they actually any good for anything beyond SDTV/HTPC/Jukebox apps?
> Take a look on:
>
> http://www.raspberrypi.org/
>
>
>>>> Now...why the hell is my WiFi so laggy...
>>>>
>>>> *This one was bought for the insane discount.
>>> That's just conflict with your neighbour's WiFi and interference from:
>>>
>>> EU standardises hamtagonistic powerline network tech
>>> http://www.theregister.co.uk/2012/11/14/plt_standard/
>>>
>>> "The second vote on EN50561-1, the EU standard for running Ethernet over
>>> mains circuits, has passed..."
>>>
>>> "... long opposed the standard, arguing that existing requirements that
>>> no device generate undue radio interference should be applied to PLT
>>> devices, which would make the vast majority of those in use today
>>> illegal. ..."
>>>
>>>
>>> Well... Those darn things are wide-spectrum broadcast transmitters after
>>> all! Possibly even powerful enough to locally obliterate broadcast radio
>>> and ADSL! The comments to that article are rather apt... I guess the PLT
>>> manufacturers care little for interfering with other users, especially
>>> so when denuding the profits of other suppliers...
>> I can certainly vouch for this. Someone in my neighbourhood uses PLE and
>> I can tell you now, nobody for several blocks around has adequate
>> grounding. The interference is enough to weaken my satellite reception
>> by over 85%.
> Expect a lot worse to come, and all way beyond the capability of most
> mere mortal users to trace where the problem is other than to be
> aggrieved with: "sometimes it doesn't work"...
>
>
>
>>> On an unrelated note, I've had a surreal argument about the earthing for
> [---]
>> *By "adequate", this consists, at a minimum, of a solid copper rod of
>> minimum cross section of 2.5mm^2 buried in the ground to a minimum core
>> depth of 1m (BS7671). Check the document itself for guidelines on this
>> simple fix, protective multiple earths, bleed earths (lightning
>> conductors and antenna groundplanes) and metal grid earthing.
> I suspect they are being cheapskate and have not bothered with the
> "protected multiple earth" bits...
>
> They've got the lightning conductor running to a separate pit with
> earthing spike. The metal building frame itself is left electrically
> floating on a concrete base. There might be pilings under the main
> supports but can't be sure, and would the (concrete) pilings offer good
> electrical earth grounding in any case?
>
> I'd best check with a multimeter at the next visit to see how many volts
> of mains (pickup) might be floating around on the network and the cable
> cages...
>
>
> The main questions are:
>
> Is a local mains earth spike always needed for 3-phase into a building?
> Or do you earth back to the substation via neutral (or is that a no-no)?
>
> What's normally done?
>
> Cheers,
> Martin
>

Last time I dealt with commercial mains I used corner-to-corner earth 
spikes - one in each corner of the building, with a ribbon conduit 
running down the flue stack for the lightning conductor (it was a 
5-storey hotel building with the flue stack running down the back of the 
structure). It's not a good idea, in my opinion, to mix passive earthing 
(line circuits and water mains) with bleed earthing (groundplanes and 
lightning conductors), since at the moment of leakage the entire bus is 
live and with no passive leakage circuit breakers to isolate equipment 
that equipment bears the entire force. In case of a lightning strike 
this has the potential to destroy computer equipment. Concrete 
foundations and pilings are *not* adequate earthing materials - which is 
why offshore static rigs have Faraday shells grounded directly to the 
sea bed on their own conductors rather than through the pilings. 
Earthing via neutral is not only dangerous, it's illegal, since any 
closed part of a circuit renders the neutral line live by virtue of the 
potential difference between it and the live line, hence any 
ground-cased equipment will have the same voltage as the neutral line. 
Neutral grounding occurs at the substation where it is done in such a 
way as to not interfere with local earth potential. This ensures that 
ground cased equipment, and local earth, carries no potential in normal 
use.  The only time grounding is not a requirement is when you're 
dealing with double insulated consumer equipment (cellphone chargers 
etc), which are not case earthed anyway.

Bottom line: if in any doubt whatsoever, drive a spike and connect it to 
the earth bus.

Rgds,

JM



More information about the Nottingham mailing list