gehenna:

 

If you do decide you need two, I've got a Cambium for sale with a PoE injector ;)

 

Is it the e410?

It's the e400 cnPilot

gehenna:

 

It's the e400 cnPilot

 

Dumb question maybe, but does the e400 and e410 work together?

That I do not know, but I imagine given they're the same product line (400 series) they should.

Looks like one is Wave 1 and one is Wave 2, so they may not be compatible.  But if you find out otherwise and you end up wanting it just pm me!

gehenna:

 

Looks like one is Wave 1 and one is Wave 2, so they may not be compatible.  But if you find out otherwise and you end up wanting it just pm me!

 

Thanks, I'll defo want Wave 2 though.

This is probably stating the obvious, but given it's a new build, I would very much be running cables all over the show. Cable is cheap and easy to install neatly pre-gib. Post-gib... not so much. Even if you leave it in the wall/ceiling unterminated, you've at least got it if you need it. And at super minimum, at least a huge chunky conduit from the patch panel up into the roof space so you can run cable later.

At some stage, we'll probably end up with 802.11ad/WiGig, operating at much higher frequencies for Moar Internetz. But physics says higher frequencies won't go through walls (damn you James Clerk Maxwell!). So at some stage, a WiGig access point may be required in every room.

I have spun "It's an investment in future proofing" into all sorts of new tools and other fun things to play with.

mdf:

 

This is probably stating the obvious, but given it's a new build, I would very much be running cables all over the show. Cable is cheap and easy to install neatly pre-gib. Post-gib... not so much. Even if you leave it in the wall/ceiling unterminated, you've at least got it if you need it. And at super minimum, at least a huge chunky conduit from the patch panel up into the roof space so you can run cable later.

 

At some stage, we'll probably end up with 802.11ad/WiGig, operating at much higher frequencies for Moar Internetz. But physics says higher frequencies won't go through walls (damn you James Clerk Maxwell!). So at some stage, a WiGig access point may be required in every room.

 

I have spun "It's an investment in future proofing" into all sorts of new tools and other fun things to play with.

 

Thanks.

Yes, I was planning on having CAT6 cable running to several areas in the ceiling, to future proof.

One question: Should I buy CAT7 cable instead?

danepak:

 

mdf:

 

This is probably stating the obvious, but given it's a new build, I would very much be running cables all over the show. Cable is cheap and easy to install neatly pre-gib. Post-gib... not so much. Even if you leave it in the wall/ceiling unterminated, you've at least got it if you need it. And at super minimum, at least a huge chunky conduit from the patch panel up into the roof space so you can run cable later.

 

At some stage, we'll probably end up with 802.11ad/WiGig, operating at much higher frequencies for Moar Internetz. But physics says higher frequencies won't go through walls (damn you James Clerk Maxwell!). So at some stage, a WiGig access point may be required in every room.

 

I have spun "It's an investment in future proofing" into all sorts of new tools and other fun things to play with.

 

 

Thanks.

 

Yes, I was planning on having CAT6 cable running to several areas in the ceiling, to future proof.

 

One question: Should I buy CAT7 cable instead?

 

I am using CAT6 for everything. It's cheap, and all all the connectors and stuff are easily available.

IIRC, TCF was recommending 6A for a while; I am not an expert but that seemed to be a widely-derided recommendation. Latest version of their guidance is CAT6.

mdf:

 

danepak:

 

mdf:

 

This is probably stating the obvious, but given it's a new build, I would very much be running cables all over the show. Cable is cheap and easy to install neatly pre-gib. Post-gib... not so much. Even if you leave it in the wall/ceiling unterminated, you've at least got it if you need it. And at super minimum, at least a huge chunky conduit from the patch panel up into the roof space so you can run cable later.

 

At some stage, we'll probably end up with 802.11ad/WiGig, operating at much higher frequencies for Moar Internetz. But physics says higher frequencies won't go through walls (damn you James Clerk Maxwell!). So at some stage, a WiGig access point may be required in every room.

 

I have spun "It's an investment in future proofing" into all sorts of new tools and other fun things to play with.

 

 

Thanks.

 

Yes, I was planning on having CAT6 cable running to several areas in the ceiling, to future proof.

 

One question: Should I buy CAT7 cable instead?

 

 

I am using CAT6 for everything. It's cheap, and all all the connectors and stuff are easily available.

 

IIRC, TCF was recommending 6A for a while; I am not an expert but that seemed to be a widely-derided recommendation. Latest version of their guidance is CAT6.

 

Cool, just trying to be as future proof as possible. Happy to pay a bit extra.

Would prefer not to have cables of an old standard in (let's say) 15 years times.

hio77:

 

danepak:

OK, so a wireless access point can spread the signal in an effective way behind it as well?

 

can yes, the design of the antenna though means that's Quite difficult to propagate and will likely have signal loss, re-transmits etc.

 

@Crowdie is the man to really explain why this is.

 

Access points commonly deployed in residential environments have internal omni-directional antennas.  These are designed to be mounted to ceilings around 2.7 to 3 metres high and primarily propagate signal across that floor.  The image to the left shows typical omni-directional signal propagation from a “looking down” point of view.  You will notice that the coverage is not perfectly circular and that is because we cannot make a perfectly isotropic antenna yet.  The image to the right shows a “side on” view and really shows how “imperfect” signal propagation is.  You can see that some signal is propagated up and the weaker area of signal propagation is directly below the access point – this is typical for omni-directional antennas.  Therefore, it is not unreasonable to expect some signal from an access point on the ground floor of a residential site to be detected on the upper floor.

Antennas have a passive gain, measured in dBi, and, as a rule of thumb with omni-directional antennas, all things being equal a higher dBi gain antenna will propagate less signal vertically and more horizontally.  Therefore a 2.x dBi gain omni-directional antenna will propagate more signal vertically and less horizontally than a 6.x dBi gain omni-directional antenna.

When you are purchasing any 802.11 product it is really, really important to look at the antenna specifications.

Crowdie:

hio77:

danepak:

OK, so a wireless access point can spread the signal in an effective way behind it as well?

can yes, the design of the antenna though means that's Quite difficult to propagate and will likely have signal loss, re-transmits etc.

@Crowdie is the man to really explain why this is.

Access points commonly deployed in residential environments have internal omni-directional antennas.  These are designed to be mounted to ceilings around 2.7 to 3 metres high and primarily propagate signal across that floor.  The image to the left shows typical omni-directional signal propagation from a “looking down” point of view.  You will notice that the coverage is not perfectly circular and that is because we cannot make a perfectly isotropic antenna yet.  The image to the right shows a “side on” view and really shows how “imperfect” signal propagation is.  You can see that some signal is propagated up and the weaker area of signal propagation is directly below the access point – this is typical for omni-directional antennas.  Therefore, it is not unreasonable to expect some signal from an access point on the ground floor of a residential site to be detected on the upper floor.

 

 

Antennas have a passive gain, measured in dBi, and, as a rule of thumb with omni-directional antennas, all things being equal a higher dBi gain antenna will propagate less signal vertically and more horizontally.  Therefore a 2.x dBi gain omni-directional antenna will propagate more signal vertically and less horizontally than a 6.x dBi gain omni-directional antenna.

When you are purchasing any 802.11 product it is really, really important to look at the antenna specifications.

danepak: Thanks.

 

The reason why I was hoping for one AP, is that if we have one located centrally downstairs, it's in the dining room.

 

I guess I could place it in the garage, where the patch panel will be located (on wall close to the kitchen). Not central, but would still increase the coverage downstairs.

 

When you are deciding on access point mounting locations ALWAYS think of it from the point of view of the worst performing wireless client – commonly Chromecasts, Smart TVs, low end smartphones, etc.  The radios and antennas in these devices will always performance poorly in comparison to the access points so we get throughput from the access point to the wireless client than vice versa.  The consequence of this is that the CRC error and retry rates from the wireless client to the access point is commonly higher than vice versa.  Therefore, when you are looking to determine a mounting location use a cheap smartphone/tablet running a video streaming application.  When you start losing frames or the image starts getting "blotchy" you are past the "real" coverage area of the radio.

The most frightening thing about the replies you have received is not one of them has mentioned roaming - moving from the radio in one access point to the same spectrum radio in another.  To get a nice, quick handoff from one radio to another is a real skill.  From a technical point of view this is a reassociation (both access points working together to ensure all your information is "transferred" from one access point to the other) not a disassociation (drop) followed by an association (initial join).

The first thing you need to determine is where the roam is going to occur.  In your building it is likely to be the stairwell.  Things to take into consideration:

• Mounting location - No matter what you do you can't fix a poorly placed access point.  On the ground floor the "Entry" area may be the best location to mount the access point - I am assuming you don't need HD video in the garage.  The toilet area should help reduce the signal strength as you head up the adjacent stairs.  You shouldn't place access points directly above each other so I would look at mounting the upper floor access point in the hallway outside Bedroom 2.
• Transmit power - When you are designing wireless networks you have three types of signal strength - want (typically -65 dBm and greater), don't want (-66 dBm to -82 dBm) and don't care (-83 dBm and less).  Modern wireless deployments typically involve time sensitive applications and these typically need a signal strength of -65 dBm or greater so this is the signal level we "want".  The 802.11 standard states any any signal detected at -82 dBm or greater must be processed so, in this scenario, any signal strength lower than our "want" down to -82 dBm will adversed effect performance so we "don't want" it.  Any signal detected lower than -83 dBm, according to the 802.11 standard, doesn't need to be processed so we "don't care" about it.  In an ideal world we could just use a Star Trek transporter to beam signal into the "want" area and not into the "don't want" area but in the real world when we transmit into the "want" area we also transmit into the "don't want" and "don't care" areas.  The key concept here, and most people don't know this, is that when you increase a radio's transmit power you make the smallest change to the "want" area (good), a larger change to the "don't want" area (bad) and the largest change to the "don't care" area.  Therefore, transmit power is a double edged sword and should be used appropriately.  
• Data rates - One of your key weapons in helping roaming are the data rates.  Commonly residential wireless routers don't allow you to change the data rates and should be used as a router only with access points deployed.  Data rates, contrary to popular belief, are not a speed but a measurement of a potential payload.  This is because a transmitted frame may get corrupted in transmit or the receiver may not detect it so the transmitter will have to send it again.  Data rates have three "variants".  "Basic" or "mandatory" data rates are used for transmitting beacons (the "advertisements" for the SSIDs), broadcast and multicast traffic.  The beacons generally transmitted on the lowest basic data rate so only having a single basic data rate and making it a higher data rate, say 12 Mbps, will make wireless clients roam quicker.  This is because higher data rates, all things being equal, don't propagate as far as lower data rates.  So raising the single basic data rate will make the radio appear to have a smaller coverage area.  On your radios you may want to disable all data rates below 12 Mbps, set 12 Mbps to basic/mandatory and set 18 Mbps and higher to optional/supported.  If your wireless clients are disassociating before they can roam lower the basic/mandatory data rate to 9 Mbps (never use 1, 2, 5.5 or 11 Mbps unless you have a wireless client that specifically needs one or more of those data rates) and test again.  "Supported" or "optional" data rates are used for transmitting unicast traffic while "Not supported" or "disabled" data rates are never used. "Not supported" data rates are also important because as a wireless client moves away from a radio the error rates rise until the wireless client and access point renegotiate a lower data rate.  When a wireless client wants to renegotiate a lower data rate but all the lower data rates are "Not supported" then the wireless client will initiate a roam to a radio on anther access point.  So using data rates carefully you can make your wireless clients roam where you want them to. 
• Interference sources - The microwave oven in the kitchen will, when activated, create a large "interference field" that will easily cover the dining area.  Mounting an access point here may be problematic.

You may need to use some "trial and error" testing to get the locations, transmit powers and basic/mandatory data rate correct.

mdf:

 

At some stage, we'll probably end up with 802.11ad/WiGig, operating at much higher frequencies for Moar Internetz. But physics says higher frequencies won't go through walls (damn you James Clerk Maxwell!). So at some stage, a WiGig access point may be required in every room.

 

The WiGig working group have not given building wide coverage as a typical use case for WiGig.  The most commonly commonly use case is multimedia transmission within a single room.  Wireless clients will be tri-band - 2.4, 5 and 60 GHz.