Smaller WIFI range with Annex32 1.5x

[PeterN]

Tnx!

It is simple ... but no way to understand it at the moment:

My first Foto shows situation of a freshly startet Atom with the Toolkit-serial-monitor connected.
The ATOM can not connect the WIFi and so it opens his own AP.
You see the weak yellow signal of the AP despite the ATOM antenna being next to the notebook antenna.

IMG_8204.jpeg

Now I disconnect the toolkit-serial-monitor and connect the Module with the webbased Installer console.
The ATOM restarts ... starts his own AP - VOILA!! with the expected HIGH LEVEL !!

IMG_8205.jpeg

Now I can close that monitor connection an restart the ATOM. And it will come up with the WEAK signal only again!!
As soon as I start connected with the web console the signal is OK .... even if the ATOM is in WIFI client Mode !


So I know now:
The naked Annex32 1.51.6 ATOM starts in any case with low TX_Power. No matter if as WIFI-client or in AP-mode.
The connected web console will lift the TX-power up.

What does the connected web console do that increases the TX-power??? While the TX-power is always low in any other case!!!



And one more detail in the fotos : The Atom running an AP with Annex32 1.48 is the strongest graph at the left side up to chan7, while this module is located in a distance of 1m . With this Annex version 1.48 the TX-level is always high.




As for Full English Breakfast: potential Health consequences did not matter in those days! And it was necessary - cause we did not find a good restaurant and were forced to have fish&chips for dinner )

[cicciocb]

Hi All,
I'm not able to give any feedback on that subject as nothing should change if the serial is connected or not, in particular for the ESP32 as the serial link continue to be active, even if not physically connected.

[PeterN]

Good morning Francesco

The healing effect of the web-based console is a nice feature.
The real problem is that the TX power seems to be set to a low level. And unfortunately that cannot be circumvented with WIFI.POWER as the command does not seem to do anything.

[cicciocb]

If the problem is the power level, I can check in the code.
I remember that I did something for the ESP32-C3 for fixing a H/W design failure of some modules; maybe that fix have some side effects?

I'll check

[PeterN]

Thanks
Yes, it's my ATOM with Annex 1.5x listening to a loudly screaming WIFI AP and only responding with a whisper.
When resetting using the web console, the transmission power seems to be initialized differently.

[PeterN]

To give some more hints:
To locate the version with the problem, I took a fresh erased ATOM and flashed 1.48.22 and 1.49.2 which then show a strong TX power in AP mode and under all power and USB-serial circumstances.
The reduced TX signal appears for the first time in Annex 1.50.5 . However this is a bit hidden, because as long as the flasher is still connected all seems OK after flashing and the automatic restart. But if I close the webbased flasher and reset , then the AP restarts always with weak signal.
And the reduced transmission power in AP mode guarantees that the transmission power in WiFi client mode is also reduced


Edit: Testing with AP-mode activated has the advantage that it is very easy to observe the power of its constantly transmitted signal with my inSSIDer wifi sniffer.

[cicciocb]

I've spent some time trying to understand something but I cannot reproduce your behavior, the WiFi.power command works and I have the same wifi signal with the PC connection or using an external power pack.

I modified the firmware to see if inside the power is correctly set and it seems the case; however I noted that, with different power sources (i.e. with a USB power supply) the strength is different. I also noticed that, in this case, the ESP32 reports a max power slightly below the value set.

So, for what I understand, it is probably a question of power source.

edit :
I tested with ESP32 and ESP32-S3 (the esp32 has a max power of 19.5dbm and the ESP32-S3 has a max of 21 dbm)

[PeterN]

Thanks for testing

Did you observe the TX strenght in AP-mode with an external sniffer ?
"Wifi.power 19.5 " for max power ? This or other values did not change anything at my output

Edit: It is not only a question of some few dB
I will test with other then the ESPpico, perhaps this has influence.

[cicciocb]

try wifi.power 2

[PeterN]

Had no effect with wifi.power 2 . .... and also not in a slow loop with other values