You are here

Ubiquiti Power (not)

3 posts / 0 new
Last post
KE2N
KE2N's picture
Ubiquiti Power (not)

I had a look inside the M2 using SSH.  If I ask for iw list it shows me the expected frequencies, but the power limit for all frequencies is 18 dBm (not the 28 expected).

Since (by coincidence) 18 dBm is the power of a LinkSys router, I wonder if somebody made a mistake? 

Or is there a more nuanced explanation?

Thanks

Ken

KG6JEI
Ubiquiti devices have (from
Ubiquiti devices have (from all observations and analysis and power merger testing ) an "off chip" amplifier. These range from as little as 1dbm to I belive IIRC 12dbm is the highest we have seen. Not 100% sure why Ubiquiti choose to put an off SoC amp on the devices and integrate it with the SoC but they appear to have done so. The speculation is they felt it was more reliable or more linear than the chip itself. This is one of the items we adjust for in the GUI to compensate for user experience, the kernel thinks it's 18 but with the 10dbm off SoC amplification it's actually 28.
KE2N
KE2N's picture
thanks

By coincidence, UPS dropped off an eBay find of mine today: an HP8900C peak indicating power meter and 84811A sensor.  Checking it against a 2.4 GHz signal generator that I have, the power levels agree to within 1 dB.

Using to make a few checks, I found my M2 puts out 30 dBm (times two) and my LinkSys puts out 20 dBm (times one).  These are peak outputs observed by watching the meter while the unit runs a speed test at max power setting. The Ubiquiti output seems to be not linear versus the GUI, with the power ramping up quickly and then flattening out from the 22 through 28 dBm settings. I suppose the radio chip really is quite linear but the off SoC amplifier is saturating. That is why they want you to back off the power at the higher MCS operation, where linearity is needed. My test unit was within 1 dB of rated output at a setting of "22".

Ken

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer