iCub HeadEdge – Substitute the AEA3 on the neck mk3 w/ the AEA4

[Nicogene]

Task description

During the board meeting, we decided to tackle this issue:

• iCub HeadEdge – Investigate about the noise in the econder reads of the neck #201

substituting the AEA3 w/ AEA2, because the AEA2 has the CRC check wires shielded and should give a less noisy signal.

Warning

These should be only changed in the physical setup, not in the CAD/WinGST
cc @MeliTaliani @salvi-mattia

The idea is to have the neck mk3 asap mountable on an existing robot, for the ones we are building we will find other solutions (fix AEA3 or think about AEA4)

Definition of done

We mounted the AEA2

cc @Fabrizio69 @maggia80 @fgarini @pattacini

[pattacini]

@fgarini aims to do a quick test on the 5V with AEA3 before the replacement.

[Nicogene]

Yesterday we spoke w/ @fgarini about this activity and before proceeding to change the aea3:

• https://github.com/icub-tech-iit/electronics-boards/issues/527

we need to find a way to measure the noise before and then after the intervention, in order to see if it has been somehow resolutive.

We can either read the raw encoder via the RawValues interface (cc @MSECode) or track the statistic about how many times the warning of the yarprobotinterface is triggered.

[fgarini]

I put the three boards behind the setup, they are in their pink bags

[martinaxgloria]

The new aea3 were mounted on the setup and tested (with bad results) in:

• iCub HeadEdge – Investigate about the noise in the econder reads of the neck #201 (comment)

cc @fgarini

[pattacini]

We decided to go with the upcoming AEA4 instead of downgrading the sensing to the use of AEA2.
I've updated the title accordingly.

cc @Nicogene @maggia80 @fgarini