Fibre and contention

Is fibre contended the same was as ADSL? I remember years ago, there was a contention figure of 50:1 on residential lines. I accept that as more people sign up to a system, the potential for slow down increases. My Sky fibre until recently is always syncing at 39996. The line is still synced the same but during the evening, throughput drops from 32MB (wireless) to down to 14MB. During the day, speeds are as they have always been. This has happened over the last month, I would have expected a gradual reduction.

Will try to get hold of a LAN cable to test the connection wired.

My stats are:

Port Status TxPkts RxPkts Collision Pkts Tx b/s Rx b/s Up Time
WAN WANoE 13759507 18932939 0 2096 1083 555:33:58
LAN Up 12165406 8987943 0 1331 1120 555:34:24
WLAN Up 13497549 8825495 10 0 5738 555:33:53
Broadband Link Downstream Upstream
Connection Speed 40000 kbps 9996 kbps

(sorry about the formatting)

Mark

there is always some form of sharing. I assume this is related to the line from the exchange to the core sky network. What exchange are you on?

In my case I am not getting any drop in speed (that I can detect) unless daughter is on youtube.com and I run a speedtester at the same time.

I have found it must be local interference somewhere as running a wired test has given me 35mb while another wireless was 14mb!! I am only sat about 10 feet away from the router, so signal strength isn't a problem.

Just wondering why my wireless signal would be so bad in the evening? Weird...

it is contended but its dynamic, now days isp's dont stick to a fixed contention ratio as customers needs keep changing. Sky seem to authorise upgrades if visible congestion occurs which is basically a sufficient policy.

Edited by Chrysalis (Fri 18-Jul-14 21:37:02)

I would suspect that in the evenings, your household and your near neighbours, having returned from work, are using WiFi more intensively than during the day; and also are more likely to be using other non-WiFi items that share the same 2.4 GHz "Unlicensed" Radio Band, such as microwave cookers.

Keep in mind that WiFi snooping programs such as inSSIDer only show you the WiFi streams that it can decipher; but do NOT show interference from other 2.4 GHz Band non-WiFi, although that interference does affect the WiFi signals.

Think of yourself having a quiet conversation in English in a quiet pub, with another person, in the afternoon. No problems.

That evening, you and your companion return to the pub; and a large noisy group has arrived. Some are speaking English, whilst others are speaking in other languages.

You can no longer hold a quiet conversation; and it is with difficulty that you occasionally manage to exchange a word or sentence with your companion, hence slowing down the transfer of information.

Then a noisy motorbike starts up outside ...

Best thing to do is to change the wireless channel, trying only 1, 6 and 11.


Open a new browser window and visit http://192.168.0.1
When prompted enter the username: admin and password: sky (in lowercase).
If you have a white router select Wireless Settings from the left hand menu.
If you have a Black or Charcoal router then it's the Setup tab in the black navigation bar you need to select.
This will bring you to the Wireless Settings page where we can adjust the wireless signal.
You'll see the Wireless Channel option with a dropdown menu, simply change this option to either 1, 6 or 11. Don't use 2,3,4,5,7,8,9,10,12 or 13. They overlap the frequencies.
Once you've changed the wireless channel number, scroll down and click the Apply button to save the settings.

I'm sure one of them will deliver better speeds as it's likely the routers just using the same channel as a nearby neighbour whos using the connection extensively e.g. they could have something on same channel such as a doorbell, a cordless phone, a bluetooth keyboard, another wifi router, a baby monitor etc. The list is endless.

In reply to a post by Discus:

I have found it must be local interference somewhere as running a wired test has given me 35mb while another wireless was 14mb!! I am only sat about 10 feet away from the router, so signal strength isn't a problem.

Just wondering why my wireless signal would be so bad in the evening? Weird...

Like eckiedoo has said in their reply, other people near you are using wifi in the evening - it's as simple as that.

Whilst UKHardy07 is correct to suggest that you try each of those three Channels, 1,6 &11, in turn, I would add that if your Router allows you to do so, try it on "AUTO" Channel Selection as well.

On AUTO, the Router will switch the Channel in use, to whichever it finds "quietest", whether the "noise" is from other WiFi sources or from other devices such as microwave cookers.

Your WiFi devices can track such AUTO changes normally - I certainly have no obvious problems working that way.

-----------------

Interestingly with the EE Bright Box 2 also having the other WiFi Band available at 5 GHz, although inSSIDer can not detect other near-by WiFi devices in that band, it tends to AUTO-switch in the range of Channels 36 to 48.

My lady-wife's Laptop is the only PC etc accessible by me, that has the 5 GHz band available; and defaults to using that higher band, although the lower band is also available to it, yet the 5 GHz Band signal is generally 10 dbs lower/weaker than the 2.4 GHz Band.

I can only presume that the Channel Switching over that range of 36 to 48, is caused by non-WiFi "noise" sources, one possibility being harmonics radiated by microwave cookers - BUT no proof!


Keep in mind that all WiFi devices are very limited in the amount of RF power that they can transmit, both physically and legally, so by definition, noise from other sources can readily swamp them, just like the noisy motor-bike.

A microwave cooker is between 500 Watts and 1,000 Watts.

A WiFi device is down around 10 milli-Watts, so between 1/50,000th and 1/100,000 of that level, both to restrict its range and likelihood of interfering with others; and also to minimise power consumption in battery-powered PCs, iPads etc.

A mobile phone typically radiates more power; and being in the 800 MHz range, their third harmonics at 2,400 MHz (2.4 GHz) could readily be a source of interference as well.

In reply to a post by eckiedoo:

A microwave cooker is between 500 Watts and 1,000 Watts.

But most, if not all of that radiation will be contained inside the oven, otherwise people standing near the microwave will be cooked along with the food!

Agreed fully, Oliver.

But leakage at a level below endangering humans, animals etc, could readily interfere with WiFi.

To quote yourself-
"
most, if not all of that radiation will be contained inside the oven,
"

At least with microwave cookers, they are in a containment system to minimise those risks, where-as mobile/cell phones are specifically designed to radiate similar, lower frequency RF; and note the complaints when the user discovers that he/she is out of range.

And how well do folks (know to) clean the RF sealing area of such cookers?

In reply to a post by eckiedoo:

Think of yourself having a quiet conversation in English in a quiet pub, with another person, in the afternoon. No problems.

That evening, you and your companion return to the pub; and a large noisy group has arrived. Some are speaking English, whilst others are speaking in other languages.

You can no longer hold a quiet conversation; and it is with difficulty that you occasionally manage to exchange a word or sentence with your companion, hence slowing down the transfer of information.

Then a noisy motorbike starts up outside ...

I'm liking that analogy

In reply to a post by eckiedoo:

Whilst UKHardy07 is correct to suggest that you try each of those three Channels, 1,6 &11, in turn, I would add that if your Router allows you to do so, try it on "AUTO" Channel Selection as well.

I always leave the router on auto., currently channel 1, it was on 11 yesterday afternoon.

Mark

Afternoon Discus

If you have not already done so, I suggest that you download "inSSIDer 2", which is still available free from some download sites.

This would preferably be on to a laptop or other portable, WiFi equipped device.

Be wary when observing it.

A part of the 2.4 GHz Band may look "quiet" in WiFi terms; but it may be that other AUTO WiFis are avoiding non-WiFi noise in that part of the band.

So regard it as an observational tool, rather than an implicit guide.

---------------------

Other sources of "noise" in that band are BLUETOOTH Devices, a significant part of the modern electronic device world.

And think of the many other devices now offered with the "convenience" of WiFi connections - central heating/air conditioning installations, refrigerators and their remote supermarket purchasing lists, audio systems etc.

Most of the middle to long-distance buses in this area have (hopefully) internal WiFi along with their necessary connections to one or more mobile phone systems to give the onward connections.

I am aware of one person who could pick up the bus WiFi in his flat, about 50 feet, 15 metres from the stationary bus.

Most pubs, many supermarkets, fast-food chains etc.

Mobile phones being used as (temporary) WiFi Hot Spots.


Of course, it may be that you are relatively isolated from many of those public sources; but you may be tightly encompassed by domestic systems.

In reply to a post by eckiedoo:

A part of the 2.4 GHz Band may look "quiet" in WiFi terms; but it may be that other AUTO WiFis are avoiding non-WiFi noise in that part of the band.

AUTO is a great idea - sadly some router manufacturers have software that is seriously buggy in AUTO mode. Only a couple of years ago all the VM Superhubs in my street ended up on the same channel. Thankfully they've auto updated firmware.

5 GHz is a useful improvement as the range is so much less that RF pollution between homes is reduced - but also interference from other unlicensed equipment (microwaves are okay, its baby monitors, TV senders and older doorbells that seem to be worst).

Agreed that poor software could achieve that situation - but it could also be that there were non-WiFi devices causing severe interference on all the other channels; and that the forced, common channel was the "least bad".

Without a Spectrum Analyser etc, it would be very difficult to come to a definite conclusion.

-------------------

Agreed about the greater absorption and attenuation on the 5 GHz Band. As I have noted, it is averaging 10 dbs lower/worse/weaker around my house, compared with the 2.4 GHz from the same BB 2 Router, almost simultaneous measurements at each of 10 locations.

I note that the BB 2 has four WiFi aerials incorporated as etchings directly on to its PCB.

From the relative sizes, it looks as though one is 2.4 GHz, and the other three are 5 GHz.

---------------------

Having noted that on glossywhite's youtube posting, I opened up the now-redundant BB 1.


Much to my surprise, it has two aerials, both metal pressings rather than etchings.

By size, one seems to be 2.4 GHz and the other, distinctly smaller one appears to be 5 GHz.

The 2.4 GHz aerial is clearly visible through the vents on one side; and is held in moulded slots from the top and bottom main faces, the PCB having a clear rebate to allow ithe aerial to stand clear, with short wire connection to the PCB.

The other, apparently 5 GHz, aerial is near the other; but mounted directly on the PCB.

---------------------

I did a Search yesterday; but have not found any mention of its presence in the BB 1; nor any suggestion that the BB 1 may have been originally intended to be Dual Band WiFi.

This is based on their relative sizes, as expected the 5 GHz aerial is about half the size of the 5 GHz one.

That also applies to the BB 2 aerials, as far as can be discerned on that youtube posting, as all four show up vaguely through the PCB, rather than being seen directly.

With the BB 2 ones, both sizes are based on slot arrays; and there is on each, almost a "horn feed" in to a larger slot. also there is what may be a "Quarter-Wavelength" stub.

---------------------------

Thinking of such stubs and the about 12.5 cm wavelength, about 5 inches, for 2.4 GHz, leads on to considering particularly nails in walls etc around houses, so many nails, screws etc are about 1.25 inches or quarter-wavelengths; and the problems those can produce!

Fortunately, 2.5 inch or half-wavelength nails are much less frequently encountered.

Agree completely about a proper analyser - WiSpy looks good but you'd need to be a semi-pro to justify the cost.
http://www.metageek.net/products/wi-spy/

Interesting about the routers with antenna on the PCB. I think my mates TPLink is designed like this, with 2.4GHz on the PCB, and 5GHz on external antenna.

My ASUS RT-AC68U has 3 external antenna, and I suspect the 2.4GHz are internal, as the design has the box sitting upright. Of course AC only works at 5GHz but I get coverage in my flat brilliantly on 5GHz to my iPhone and laptop, and I control which one I use with two different SSIDs. The 5GHz gives me full FTTC speeds (50Mbps sync here) whereas with 2.4GHz in evenings when the kids are home on the estate, I get about 10Mbps at best :-/

Interesting idea about nails, I hope they're not affecting! Its been a long time from my CB radio days, LOL!

glossywhite's interior views of the Bright Box 2-

https://www.youtube.com/watch?v=Ly6kV-UUuZg

From 13 minutes and later.

Particularly at 13:04, one of the aerials shows up relatively clearly through the laminate.

Note the transmission line coming in from the bottom, opening out in to almost a cross-section of a satellite LNB!

The Gadget Show did a test last year on WiFi routers with 3 microwaves and baby monitors in the same room.

Very little difference was found with them off or on. The main difference was when they moved the laptop to another room then outside the building.

Seems like different wall construction was the big issue.

Regarding Attenuation/Power Level, I would accept that fully, from observing how it is in the house where a grand-daughter lives.

It is dates from about the 1880s, massive brick and stone walls throughout. Former Mine Manager's house, so nothing spared!

Not sure of the attenuation figures; but in speed terms on WiFi beside the router, about 11 Mbps.

Straight along one corridor, then at right angles in to a room - wavering 1 to 2 Mbps. Distance about 40 feet, say 13 metres.

Edited by deleted (Mon 21-Jul-14 17:16:36)

I live in a Victorian terraced house hailing from the 1890's and apart from Victorian bricks having a high iron content whoever did the renovations lined just about every wall with foil-backed plasterboard.

The router in the front room covers the two bedrooms at the front of the house and I need another access point in the main bedroom at the back to cover the ground floor at the back.

There's virtually no signal at the back of the house from the router which is a distance of around 25 feet.

Foil backed pb is the worst for allowing a signal to penetrate.

Most new houses have stud partition walls but whether they are using foil backed pb - who knows?