RouterOS Tx Power Issue with SENAO/EnGenius EMP-3601 radio

RouterOS Wireless Networking
1

I installed in a wifi AP/hotspot (802.11b/g) a SENAO/EnGenius EMP-3601 (EU2) LFP (chipset is Atheros AR5213)

Card’s datasheet declares the following TX powers (ETSI version):

802.11b

  • 1..11 Mbit: 20 dBm

802.11g

  • 6..48 Mbit: 20 dBm
  • 54 Mbit: 19 dBm

When enabling the card, RouterOS reports the following “Current TX Power”:

\

  • 1..11 Mbit: 16 dBm

  • 6..24 Mbit: 15 dBm

  • 36 Mbit: 14 dBm

  • 48 Mbit: 13 dBm

  • 54 Mbit: 12 dBm

(note: same values for “Tx Power”, “Real TX Power”, “Total TX Power”)

As you can see, the reported values are quite lower (up to 7dBm of difference) than the declared ones.

Card’s relevant settings are:

  frequency-mode=manual-txpower
  tx-power-mode=default

I tested two different radio cards in two different boards (RB433AH) and two different RouterOS versions (4.17 & 5.14) with exactly the same results.

I’m not able to measure the actual rf-power sourced by the card right now, so I would like to know if there is any known RouterOS issue with this model (i.e. the effective TX power not being reported correctly) and if I can safely ignore the values reported by RouterOS and assume that with the above settings the card is working at its max power.

Or, alternatively, can I safely set “tx-power-mode=manual-table” and manually assign the values indicated by the datasheet without the risk of burning the card?

Thank you.

2

Ok guys,
may be my question was too stupid to deserve an answer or the matter so well known to everyone to require a comment: well, it was not for me, so I did my own measurement.

I really don’t believe that nobody is aware of this but, just in case, here are my results so to help anyone else should face my same problem.

There is definitely a “power offset” in SENAO/EnGenius EMP-3601 ETSI minipci card which is not correctly taken in account by RouterOS. The difference between the effective power sourced by the card and the value reported by RouterOS is almost exactly 4dB for “b” mode and around 6 dB (+/- 1.3 dB) for “g” mode. The card appears to give its declared max power with RouterOS “default” setting, so there is NO NEED to manually override the power table, and it might by unsafe for the power amplifier stage.

About this last point: there are two versions of EMP-3601, the “european” version (ETSI regulation) and the “US/international” one (FCC regulations), the latter declaring an higher power (as ETSI regulations limit the power to 20dBm). May be that the rf hardware is exactly the same, the ony difference being the default power settings and the values programmed into the card and returned to the driver (a quite common scenario), so I tried to override the power table of my (ETSI) card just for the few seconds required to take the measurements: the effective power, so as the current drained, increased as expected and the rf spectrum did not show anomalies (up to 10GHz). May be my hypotesis is correct. Anyway I think it’s better to NOT overdrive the ETSI card as (apart of breaking the regulations) it would probably require a proper heatsink. Just for your reference, these are the power values I used in my test:

+-----------+----+----+----+----+----+----+----+----+----+----+----+----+
| CARD RATE |  1 |   2| 5.5|  11|   6|   9|  12|  18|  24|  36|  48|  54|
+-----------+----+----+----+----+----+----+----+----+----+----+----+----+
| dBm       |  18|  18|  18|  18|  17|  17|  17|  17|  17|  15|  13|  12|
+-----------+----+----+----+----+----+----+----+----+----+----+----+----+
| GAIN (dB) |   2|   2|   2|   2|   2|   2|   2|   2|   2|   1|   0|   0|
+-----------+----+----+----+----+----+----+----+----+----+----+----+----+

The card is now working without problems at its default power setting.

Regards,
rock