• Using hostapd to turn a Rpi3 into a wireless access point on 2.4Ghz band.

    • How can I estimate the broadcast power of this AP?

    • PS adapter is 5V 2.5 A. P=IV = 12.5 watts, but, I think that's just the available source adapter power and not really what would be used.

    • What could I do to measure the broadcast power of this?

  • SoC: Broadcom BCM2837

  • Broadcom BCM43438 chip provides 2.4GHz 802.11n wireless LA
  • Networking: 10/100 Ethernet, 2.4GHz 802.11n wireless

Looking at the specs of the BCM43438 I see:

pi wifi power vbat and current draw

So looking at the table I see a range from 37 to 41 mA * 3.6V = max 0.1476 Watts.

Does this look right? And, is that powerful for an access point if so?

  • \$\begingroup\$ This will almost certainly be a "pulse" load (where the pulses might as much as double the baseline consumption), so you need some sort of time-domain current meter, such as an instrument which logs measurements over serial/USB to a PC, or a current-to-voltage converter and a scope, or at minimum a peak-hold ammeter. \$\endgroup\$ – Chris Stratton Jun 29 '17 at 19:04
  • \$\begingroup\$ Thanks. I'm not calculating this for an statistical purpose so don't really need any meaningful data. I just want to make sure it's not something too powerful so I can tell my IT department anticipated range while it's on for a test. I'm actually just wanting to compare the max power to something on the market and want to see if I've arrived at it correctly. \$\endgroup\$ – rpi3ap Jun 29 '17 at 19:06
  • \$\begingroup\$ @Chris Stratton if you have the time do you think I could do this with a wifi spectrum analyzer? I recall seeing power and peak hold options on it, but, I have no idea how to focus on this test AP and filter out all the other 2.4Ghz traffic we have going on here. \$\endgroup\$ – rpi3ap Jun 29 '17 at 19:11
  • \$\begingroup\$ It's unclear if your question is about power consumption or about RF output. For the RF, this would essentially be duplicating the FCC (or whatever) certification test process. You could just find the applicable report. \$\endgroup\$ – Chris Stratton Jun 29 '17 at 19:49

The specifications to look at are the dBm transmit numbers. dBs are always an expression of a power ratio. In this case it is the logarithmic ratio the AP output power compared to a 1 mW transmitter.

To convert dBm back to power output in watts, divide the dBm figure by 10, then raise 10 to this power and finally multiply by 0.001 watt. For example, to convert the 20 dBm figure to watts, take the 20 dBm divided 10 to get 2. Then raise 10 to this power (102) to get 100. Then multiply 100 times 0.001 watts to find that the output power is 0.1 watts.

Here is the same thing in its full formula:

Watts = 10(dBm/10)*0.001

Of course power output is not the only variable affecting range or effectiveness. The gain of the transmit antenna, the gain of the receive antenna (often shared with the transmit antenna), the sensitivity and selectivity of the receiver, interfering signals, and the distance between APs all play a role.

| improve this answer | |
  • \$\begingroup\$ Thanks manThanks manThanks manThanks manThanks manThanks manThanks manThanks man \$\endgroup\$ – rpi3ap Jun 29 '17 at 19:47
  • 1
    \$\begingroup\$ Raw dBm won't actually be all that meaningful. What you probably want are the regulatory numbers, which are in field strength units after the antenna gain. And then of course there's the receiver performance - it does no good to be able to talk to clients that you can't hear. \$\endgroup\$ – Chris Stratton Jun 29 '17 at 19:48
  • \$\begingroup\$ @chrisstratton The number you are referencing is an EIRP limit of 36 dBm (~4 watts) for a 2.4 GHz transmitter in order to comply with FCC WiFi regulations. There is no gain limit on the receive system. \$\endgroup\$ – Glenn W9IQ Jun 29 '17 at 20:30
  • \$\begingroup\$ I'm not referring to a limit, I'm referring to an actual measurement of the device. A actual field strength measurement of a typical unit made in an accredited lab for certification purposes will be much more useful to determining range than the software-commanded (ie, uncalibrated) power output of the radio, both because it is an actual measurement, and because it includes the antenna. \$\endgroup\$ – Chris Stratton Jun 29 '17 at 23:12
  • 2
    \$\begingroup\$ @GlennW9IQ your comments keep getting more absurd - you propose an extremely expensive piece of test gear and modifying the device to connect it (and still having to make guesses about the antenna) over reading an existing report of such tests that have already been performed, which is available online???. \$\endgroup\$ – Chris Stratton Jun 30 '17 at 3:06

Not the answer you're looking for? Browse other questions tagged or ask your own question.