@Arno thats good to know ;D i imagined one could wire the ATOM and SERVO2s together anyway to pipe the signals. But having a CORE seems fine too so here is the new setup:

its great that the FIRE does have a Battery-Bottom already and i assume this is a very basic setup now. maybe it would be more straight forward to make a single stack but making two interconnected stacks like in the image is surely possible.

Maybe removing the Battery-Modules and get the power right instead. (Though i really like the magnetic charging-connector ;)

okay lets try this setup. from the descriptions it seems to be perfect for a robotic-arm. so just need 6 of these. interconnecting them in a row using the Port.B and Port.C plugs on the Motion-board and Grove cables.
in software the servo-controls get piped from the usb-connection down the Ports through all the ATOMs.

Maybe this sounds reasonable. Sadly the ATOM Motion Kit is currently Out of Stock

edit: in case the servo-controls can not be passed easily/quickly though the row-connection (e.g. the data needs to be received and sent by each ATOM with major delay), it may make sense to connect them using a PaHUB2 and an additional ATOM for parallel distribution?

Hey folks, just want to share my recent ambient-light: 1D and 2D Cellular-Automatas on the M5Atom :D
https://github.com/Schlabbermampf/M5Atom_CellularAutomata/

Have fun!

edit: appears there is already a 2D CA with Conway's Game-of-Life in the Advanced Display examples from M5Stack^^ thats nice, so the sliding 1D CA with random rule seems to be the new thing here :D

On one hand things are going great and i'm learning a lot on the way.

On the other hand i realized that reading the Position of a servo, like it is shown in the Arduino example-code of the Motion Kit, is not sufficient as sensory-data (it just represents the applied position/velocity). Due to physical interaction i want to track the real, measured angle to compare with the desired.

So in the long run i'll use servos with Position-Feedback to read the angle from the additional wire. As noob its hard to compare any specs, but the "KST MR320 PF" may be appropriate as replacement (still confused about the min and max angle it can take, it must not turn infinite, but a range of at least 270° would be great).

For the current prototyping, i will extend each servo with an Angle-Module (potentiometer), combined with a PbHub and connected to the Motion-Kit. This may be a bit overkill (in cables and weight), but cheaper and sufficient to have a nervous-system as processing-inputs.

okay lets try this setup. from the descriptions it seems to be perfect for a robotic-arm. so just need 6 of these. interconnecting them in a row using the Port.B and Port.C plugs on the Motion-board and Grove cables.
in software the servo-controls get piped from the usb-connection down the Ports through all the ATOMs.

Maybe this sounds reasonable. Sadly the ATOM Motion Kit is currently Out of Stock

edit: in case the servo-controls can not be passed easily/quickly though the row-connection (e.g. the data needs to be received and sent by each ATOM with major delay), it may make sense to connect them using a PaHUB2 and an additional ATOM for parallel distribution?

@felmue thank you for clearing this up! the schemantics are surely to trust!
taking a look at the "PLEN5Stack DIY Small Bipedal Robot" it uses a custom "Control-Board" to connect the servos, and has also a custom battery plugged into (1500 mAh)... 8 servos are though an optimistic number

@ajb2k3 okay makes sense somehow, so it is dangerous to change the commands to the servos with a very high frequency. That makes it of course harder to just run a single servo at a time xP
But in general it sounds reasonable to control 12 servos but power only 1 at a time with a battery-module?

@felmue Thank you for your kind answer! Since the shop description of the SERVO2 states: "When the battery base is used for power supply, the maximum output power is 5V / 2A." So it sounds like its possible in general but realistically to drive 1 or 2 servos at a time? or do i miss the point? i understand that the Battery-Module is way to weak to power all the servos at once.

@Arno yes, i can remove the battery-bottom from the FIRE module, put the SERVO2 module instead and attach the battery at the bottom (to use the charger). just thought: to have each SERVO2 has its own power supply (maybe using more SERVO2+Battery modules with each less servos attached?)

@ajb2k3 yes makes sense, thats what the SERVO2 module shop-description says, i just get it to understand slowly ;D (just discovered the Documentation of all the M5 parts!) since i like to use the 360° servos with speed and direction control via PWM it may be possible to ensure (in software) a limit for the applied power in total.

Looks like bridging the 30 pin MBus over some short distance with a cable is kinda overkill so a single stack it shall be.

Power management strongly depends on what the servos have to lift, so overall weight-balance/efficiency of all parts is what its all about then in the long term (even when plugged into a full power source).

So i want to get my hands on these pieces soon and start digging. happy to share some progress here :D
Thanks for your kind and fast replies :D

edit: here is the stack, quite simple. (hope these servos do not fry to soon):

@Arno thats good to know ;D i imagined one could wire the ATOM and SERVO2s together anyway to pipe the signals. But having a CORE seems fine too so here is the new setup:

its great that the FIRE does have a Battery-Bottom already and i assume this is a very basic setup now. maybe it would be more straight forward to make a single stack but making two interconnected stacks like in the image is surely possible.

Maybe removing the Battery-Modules and get the power right instead. (Though i really like the magnetic charging-connector ;)

@paulhdietz thank you that's surely something to consider! i may also think about reducing the total amount of servos as well.