WiFi Question

[NY]

The sort of houses that have 3' stone walls often have another wifi killer: a huge metal cooking range surrounding the open fire. A large amount of cast iron has interesting effects on wifi. Old stone cottages tend to have another problem: they may be in remote parts of the country which have poor phone lines (eg 8 km from the exchange) which means that you get slow or intermittent ADSL. Try downloading a bloated HP printer driver (with all its unwanted extras!) or an anti-virus package over a link that is about 250 kbps at best but which keeps dropping out (ping times vary between a very respectable 40 msec and upwards of 2500 msec over the course of a few minutes). (*)

The worst house for getting internet in various rooms was a cottage that had previously been several different cottages. It had the lot: thick stone walls, two hot water cylinders in different parts of the house, multiple mains circuits fed from different meters, walls that definitely shouldn't be drilled through or be defaced by cable trunking,

I struggled for a long time with the router in various different places to get best wifi coverage from it and then using several wifi repeaters to transport the signal to the periphery of the building. I tried Powerline until I realised that the signal wouldn't reach some parts of the house because they were on a different meter (and maybe even a different mains phase).

(*) At least in the case that I was working on the other day, the customer said "it gets worse whenever it rains" which sounds like good grounds for getting BT Openreach (via the ISP) to check for water in underground cabling over the 8 km journey to the exchange. Thank goodness the house had been wired so all the internal phone wiring was the customer's own wiring which could easily be disconnected to prove that the problem still occurred. It's a different situation when (like in our house) there are two extensions permanently wired to an old GPO lozenge box (which you Must Not Touch) so you can't prove whether or not ADSL problems are caused by extension wiring. But that's all a side issue to the matter of wifi and Powerline.

[NY]

I agree with you. I'd never use Powerline if there was an easy way of installing Ethernet. My parents actually drilled a hole through the internal wall of their house (which was brick, not plasterboard) to get a cable from the router in one room to the computer in the next room (they have "his and hers" offices in adjacent bedrooms of the house).

Where we live, there are very few wifi networks, and certainly none on one of the three "magic channels" 1, 6 or 11 which are guaranteed not to overlap with each other, so I can use one of these (I forget which) and know that I have no interference from neighbours' networks. But wifi reception can still be very variable, even in the same location. I suspect some other device that uses 2.4 GHz (or produces harmonics in that range) which is not visible as a wifi network and therefore doesn't show up on InSSIDer. Whatever it is causes a lot of mush at the low end of the VHF waveband as well in one bedroom. (And yes, I've unplugged all the wall wart power supplies round about, as well as the Powerline devices, in case it's one of those which is radiating crap).

Do wireless intercoms and baby alarms use 2.4 GHz?

[Harry Bloomfield]

NY has brought this to us :

Much longer distances have been managed than that.

I really don't follow how a copper tank full of water, can attenuate a signal more than the same tank, but empty.

[newshound]

I can see the one from next door, and usually at least three from across the road. They are well down on signal of course. I have no reason to suppose they cause significant interference, I get much the same speed cabled to the main router as on WiFi through the extender.

[Harry Bloomfield]

It happens that newshound formulated :

Using a Wifi scanner application, I see around 14 AP's using my laptop on the ground floor. Two of them I believe from next door, which are logged as stronger signals than my own, even one just 8 feet away in the same room.

Doing the same scan on the top floor, I can get a few dozen AP's logged.

What puzzles me, is that the signal strength graph show the usual slight variations, but occasionally shows a massive spike in the signal strength of some of the signals. It is perhaps an anomaly of the scanning software.

[Roger Hayter]

If something acts as an absorber of electromagnetic waves and is lossless it also acts as a radiator. If the energy is taken out of the system by a matched electrical connection (as in an aerial) then that energy will not be re-radiated. I guess maybe the water could act as a lossy absorber of electrical energy and have the same effect. Like you, I am not very convinced that a water tank is going to act as a highly effective 2.4GHz aerial in the first place, though.

[DerbyBorn]

Harry Bloomfield wrote in news:o9bi34$1gu$ [e-mail]:

If it was my house, I would.

[dennis]

Maybe there are some MIMO APs there and you get one of the beams formed in your direction.

[Harry Bloomfield]

Roger Hayter submitted this idea :

Quite, it will though act as a screen, much reducing the signal strength behind it.

Try microwaving water in a metal pot - the water will remain cool, because it doesn't absorb any of the radiation, it is screened by the metal.

[tony sayer]

Indeed !

What a lot of people miss is that there isn't much room in the 2.4 ISM band and congestion is a big problem which does make for poor performance more often than not..

[Harry Bloomfield]

I had to look the MIMO part up - three streams of data, occupying three channels. No, I am only seeing the spikes on the single channel they are occupying.

[Lobster]

Most respondent seem to be assuming that if you use a mains-borne ethernet signal, you're stuck with hard-wired internet rather than wifi. It's defin itely not an either/or: eg I cured a wifi blackspot in my house (which has lots of thick stone walls) using one of these:

(or

share the same frequency band as the 'hub thing' they halve the speed.

I currently use my auxillary extender thingy on a completely different netw ork to the main one, so my devices swap over to the best signal when necess ary. I must admit I had a lot of grief getting it set up, and once it worke d I just left well alone... I've never quite understood the whole speed-hal ving malarkey; am I avoiding it by what I'm doing? Or can I simply change the names of the two routers to match; or do they clash then? Ideally I'd much rather have just the one wifi network, for sure.

David

[Dave Liquorice]

Is that another wired AP? If not how does it get it's connection to the LAN?

On a given channel only one thing can transmit at a time. With a repeater (aka extender) retransmitting on the same channel as the AP each packet is transmitted twice, once from the AP then again from the repeater. Thus halving the number of time slots available.

Certainly if the two APs are on different channels. If on the same channel and they can't hear each other there is potential for a "mush" zone between them where a device can hear both and if the APs transmit at the same time stomp on each other. Also the device may start switching between them.

The SSIDs can be the same and even on the same channel provided the traffic levels aren't high on both at the same time.

[Dave Liquorice]

Some do, and video senders and blue tooth and ...

[Dave Liquorice]

They certainly did, I'm not sure if they are still on 2.4 Ghz as the kit has changed a couple of times since the orginal installion 15 odd years ago. That used Cisco Aeronet 350 series bridges, 15" dish one way and "long" yagi the other.

[Dave Liquorice]

Or the metal is a better absorber than the water and soaks up all the energy... You need to repeat the experiment with the water in a plastic container outside of the metal pot.

[tony sayer]

Reflector rather than absorber Dave...

[tony sayer]

5.8 Band C is the preferred way nowadays works much better for longer range links. We have got one in daily reliable use over some 18.7 miles...

[The Natural Philosopher]

2.4 is not used much for point to point. It is used for 'village wifi' and can get a km or two with the right antennae, BUT it does tend to stop working in rain 2.4Ghz is probably good for rain radar tho.

Point to point is at a lot of other frequencies.

Rain and microwaves is a very interesting subject. generally te higher the frequency the worse the effect BUT there are as has been pointed out, deep and narrow notches in longer wavelengths like 2.4Ghz.

Which is why 2.4Ghz is left for 'amateurs' use. In practice that means microwaves, wifi, model radio control, some local telemetry and a few other short distance uses.

Tony Sayer will be along with chapter and verse on what is used professionally.

[The Natural Philosopher]

Probably not. 27MHz (CB, ancient model radio control) is available for low bandwith localised shit