my investigations

[lumapu]

First of all: Thank you for creating such a good place to collect information.

Without knowing about this repo I started to measure on my printer (A1 with AMS lite) to determine if there is any data transferred on the free AMS port. I'll add some pictures which can be part of a documentation later on (let me know if I should open a PR for adding documentation).
The captures only consist of data which is transferred while the printer is in idle mode and AMS lite connected.

Using Picoscope I measured both lines.

20250804-both.psdata.zip

export as CSV (first column shows byte in packet, there are 64 packets in total, second column is the data)
20250804-both.csv

I had to configured the serial decoder like this

The signal looks like this

[lumapu]

I further investigated the data I captured.

I can't confirm, that long packets always have a value lower than 0x80 in the second byte. It feels that there are two different types of packages, where the byte defines different package headers. For now I only found C5 and 05.

250804_BambuBusResearch.ods

Is there more documentation available? How was the protocol analyzed which is referenced in the tables of the readme.md?

[leonllrmc]

Hey, I may have more data in stock somewhere, will send it when I'm back from vacations.

Part of the protocol was figured out by me and especially 4016N research, we were also given a parser script.

by the way, at least on the A1 mini, both AMS port uses the same serial to RS485 converter and so have the same signal.

P.S: yes a PR to add the documentation would indeed be good, just remember to include a pdf version of your documents =)