Icicle Kit Case

[hsuenaga]

Hi, I share STL files of ICICLE Kit case. If you looking for case, this may help you. You can print STL files with 3D printers.

The case has access port for: 3 micro USBs, 2 ethernets, SD card, DC input. other ports/sockets are disabled.
You need 10 M2.5x8mm countersunk head screws to tighten icicle kit and 4 M2.5/M2.6x6mm countersunk head screws to tighten lid.
Probably, different length (< 10mm) screws are OK. I recommend to glue M2.5/M2.6 nuts to lid of the case.

The case is designed using FreeCAD. I share a FreeCAD file too.

Thank you
ICICLE_BOX.zip
.

[AntonKrug]

Great job, probably you want to share it on sites as thingiverse as well.

Wondering, would you be able to generate STL files without the RISC-V text?

As it currently will cause excesive use of support matterial while without text the whole face could be flat againts the print bed.

[hsuenaga]

I measured temperature of SoC using external probe. This is not a junction temperature, just a reference. Probably, we can probe more accurate temperature using TVS, but I don't know how to access it from Linux environment at this time.

I used stress(1) as simple software load.

[TommyMurphyTM1234]

So the case seems to be adding c. 10°C to the temperature?

[AntonKrug]

It might be even more, as the 'lid open' control was already with closed bottom, might not have as much impact as closed top, but might have a impact.

[hsuenaga]

At least 10[C] is added(note that the temperature is not saturated within 14:00-15:00). PolarFire's recommended operating temperature is Tj=100[C] and absolute maximum is Tj=135[C]. Though I didn't measure Tj directly, the temperature is not so high.

Hot spot appeared in FLIR wasn't PolarFire, it was Vitesse VSC8662 GbE phy regardless of link status... but I didn't probe it. I will probe it later if I have time to do.

[AntonKrug]

Shiny metal surfaces can't be measured directly with FLIR, they are to IR something like mirrors are to visible light.

https://www.youtube.com/watch?v=lfi5MOhgRbM