The car door fixes RSRP, apparently

During a 20-minute stationary check with Nemo Outdoor at a 700 MHz site, every time I closed the driver’s door the RSRP jumped about 4 dB and the 512-byte ping loss went to zero; door open, we were back to jitter and drops… I ran it three times, fixed antenna height and cable routing, and logged timestamps, so my protocol says it’s real; anyone else have a car act like a tunable RF absorber and make network reliability look magical?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠‌‌⁠⁠‌⁠‌​‌‍⁠⁠‌⁠​​‌‍‍‌‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠‌​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌⁠​​​⁠‌‌‌‌‌‍‌​‍‌‌‍⁠‌​‍⁠‌‌‌‍‍‌⁠‌⁠‌​​⁠‌‌‌‍​⁠‍‌​⁠‌‍‌‍‌​‌⁠‌‌​⁠​‌​‍​‍‌⁠⁠‌​

But had this at 700 too — when my mag-mount was near the door seam, the gap behaved like a ‘slot’ and I’d see about 4 dB jumps; moving the antenna to the roof center made RSRP and 512B pings steady, though if you’ve got metalized tint it helps less — where was your antenna sitting?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌⁠​⁠‍​​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠‌‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌‍​​‍⁠‌‌‍‍​‌​‍⁠‌‍‍​‌‍⁠⁠‌​⁠​‌⁠​⁠‌‍‍⁠‌‌‍‌‌‍​‍​⁠‌‌‌‌​⁠‌‍‌‌‌​​‌‌‌​​​‍​‍‌⁠⁠‌​

Lock AE before glare-reduction under overhead LEDs; stops pumping — ‘Edge enhancement’ past mid gives halos.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌⁠​⁠‍​​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‍​⁠​​​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠‍‍‌‍‌​‌​‌​‌⁠​​‌⁠‌‌​⁠​‌‌‌‍‍‌​​⁠‌​‍‍‌‌​​‌​‍‌‌‌​‍‌‌‍‌‌⁠‌​‌‍‍‌‌‍‌‍​‍​‍‌⁠⁠‌​