Before diving into the details, please let me express that I find it amazing what the ESP8266 offers in relation to it's price - but unfortunately, I'm having an issue with the output power:
The ESP8266 data sheet claims that the maximum output power of the chip is +20.5 dBm (at 1 Mbps) and the ESP-07 data sheet says that the typical output power of the module is +18.5 dBm (min +17.5, max +19.5). +18 dBm would be fine for my application (adding a 2 dBi antenna). Now the problem is that when measuring the output power (using a $20K digital spectrometer), then it is only +12 dBm!
For the measurement, the ESP-07 was connected to the spectrometer via a u.FL pigtail. The resistor to the on-board antenna was removed (as required when using an external antenna through the u.FL connector). Increasing the supply voltage to 3.7 Volt didn't made any difference (using a strong laboratory power supply and 10µF cap added close to the ESP-07 module, so the power supply isn't the problem). TXPOWER was set to 82 (+20.5 dBm) and the ESP-07 was programmed in SoftAP mode so that the beacon was sent 10 times per second. The beacon was sent at 1 Mbps (BPSK/DSSS), checked with a USB-WiFi dongle and wireshark on a PC. Two other modules have been measured: a Carambola 2 with maximum specified +21 dBm was measured at +20 dBm (with the same u.FL pigtail as above, so the pigtail probably isn't the issue) and a Zigbee module with a power amplifier specified at +21.3 dBm was measured at +21 dBm. So The measurement appears to be ok.
So at the moment the output power of the ESP-07 module is at least 6 dBm below my expectations.
Any feedback is highly appreciated!
Bernhard