Enlarge /. The top floor of our test house is relatively straightforward – although like many houses it suffers from a terrible router placement that is not near its center.
Here at Ars, we've spent a lot of time researching how Wi-Fi works, which kits perform best, and how future standards will affect you. Today we're going to do something more basic: we'll teach you how to find out how many Wi-Fi access points (APs) you need and where to place them.
These rules apply regardless of whether it is a single WLAN router, a mesh kit such as Eero, Plume or Orbi or a series of wired access points such as the UAP-AC line from Ubiquiti or the EAPs from TP-Link. Unfortunately, these "rules" are necessarily closer to "guidelines" because there are many variables that a chair a few thousand miles away cannot fully account for. However, if you become familiar with these rules, you should at least develop a better practical understanding of what you can expect from your Wi-Fi equipment – and what not – and how you can make the most of it.
Before we start
Let's go through some of the RF theory (radio frequency) before we start with our ten rules. Some of them make a lot more sense if you understand how to measure RF signal strength and how it attenuates over distances and obstacles.
Enlarge /. Note: Some RF engineers recommend -65 dBM as the lowest signal level for maximum power.
The graph above shows some simple loss curves for free space for Wi-Fi frequencies. The most important thing to understand here is what the units actually mean: dBM convert directly to milliwatts, but on a logarithmic base ten scale. For every 10 dBM drop, the actual signal strength in milliwatts drops by a factor of ten. -10 dBM is 0.1 mW, -20 dBM is 0.01 mW and so on.
The logarithmic scale enables the signal loss to be measured additively and not multiplied. Each doubling of the distance reduces the signal by 6 dBM, as we can clearly see if we look at the bold red 2.4 GHz curve: At a distance of 1 m the signal is -40 dBM; at 2m it is -46dBM and at 4m it is up to -52dBM.
Walls and other obstacles – including, but not limited to, human bodies, cabinets, furniture, and devices – further attenuate the signal. A good rule of thumb is -3 dBM for any additional wall or other important obstacle that we'll talk more about later. You can see additional curves shown above in finer lines for the same distances, including one or two additional walls (or other obstacles).
While ideally you should have signal levels of no less than -67 dBM, don't be upset about trying to keep them much higher – there is usually no real difference in performance between a hot -40 dBM and a much cooler -65 dBm. on a chart as far apart as it seems. Wi-Fi is much more than just raw signal strength. As long as you exceed this minimum, it doesn't matter how much you exceed it.
Indeed, a signal that is too hot can be as much a problem as a signal that is too cold – many forum users have complained about low-speed test results until a wise head finally asks, "Did you put your device right next to the access?" Move it a meter or two away and try again. "Sure enough, the" problem "resolves itself.
Rule 1: No more than two rooms and two walls
Our first rule for placing access points is, if possible, no more than two rooms and two interior walls between access points and devices. This is a fairly fudge-y rule, as different rooms are shaped and dimensioned differently and different houses have different wall structures – but it is a good place to start, and it will do you good service in houses and apartments of typical size and reasonable standard with modern sheet metal wall construction ,
"Typically large," at least in most United States, means bedrooms about three or four meters per side and larger living areas up to five or six meters per side. If we take nine meters as the average linear distance that covers "two rooms" in a straight line and add two interior walls at -3 dBM each, our RF loss curve shows that 2.4 GHz signals are at -65 dBM are fantastic. 5 GHz, not so much – if we need a full nine meters and two full walls, we have dropped to -72.2 dB at 5 GHz. This is certainly enough to connect, but it's not great. In practice, a device at -72 dBM at 5 GHz will likely achieve roughly the same raw throughput as a device at -65 dBM at 2.4 GHz. However, the technically slower 2.4 GHz connection tends to be more reliable and has consistently lower latency.
Of course, all of this assumes that distance and damping are the only problems we face. Rural users – and suburban users with large shipyards – are likely to notice this difference and the rule of thumb "2.4 GHz is great, but man, 5 GHz is shit" is internalized. Urban users – or suburban residents in housing estates with postage stamp yards – usually have a completely different experience, which we will cover in rule 2.
Listing picture by Jim Salter