aircrack-ng / rtl8812au

RTL8812AU/21AU and RTL8814AU driver with monitor mode and frame injection

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Even with `txpower` set to 3000mW, current draw is too small.

rklasen opened this issue · comments

Hi,

I measured the current draw of an Alfa AWUS036ACH with a FNRSI FNB58 Power meter. To repeatably generate a current draw,
I'm using

aireplay --test wlan0

while the card is in monitor mode (device name remains wlan0, but iw dev shows monitor mode). The txpower is set to 20dBm. When I set the txpower to 3000mW, iw dev reports 30dBm dutyfully, and the max current draw is 700mA (so about 3500mW drawn, which would be fine).

aircrack-3000mW

For some reason, the driver accepts higher txvalues up to 4000, but the drawn current decreases:

aircrack-4000mW

So clearly that can't be true. The adapter has integrated power amplifiers as you're probably aware (see here for details), but I'm not sure how they're handled by the driver.

I can hear something hiss in the adapter during transmission, and the volume seems to correlate with the txpower (if at most 3000mW).


I'm comparing this driver with https://github.com/morrownr/8812au-20210629. The other driver only accepts a txpower value of up to 3000mW, confirmed with iw dev, but the burst current draw is much higher, about 1200mA.

morrownr

I think that is plausible, since the power draw should be much higher than the transmit power, but clearly the two drivers behave differently. The hissing is much louder this time too.


So, which one is correct? I don't have an RF power meter, so I cannot measure the actual adapter output directly.

I'd be glad to run more tests.

The system is Arch, so both driver versions are checked out from git fresh.