open-source-silicon.dev

I see "A cell based buffer is used to drive the external 100 ohm differential output." mentioned in a project readme, but don't see references to this elsewhere

<@U016H5X1K62> :wave: hope you don't mind the ping, the satcom stuff is awesome (and the VCO GDS is art!). Am I interpreting your usage of the harness correct? (this message and above)

<@U01PK07UZEJ> yeah, the signals go through the mux which is obviously a signal integrity nightmare.
I don't need to get any RF power out of the chip for this submission. So my hope is that by having enough drive strength and using a LNA and low bandwidth I can get just enough power through that I can still measure the output signal to some degree.
There was no other reasonable choice for this submission so I'm just gonna try and see how it pans out

Makes sense. I wonder how feasible it is to build an antenna out of the top metal layers.

Could be an alternative way to test things like VCOs and transceivers

(I see a bunch of papers that demonstrate this on other processes)

Yeah, people have definitely done it although I imagine those are at 10's of GHz.
At 2.4 GHz a half-wavelength patch is 30 mm which is absolutely huge by IC standards.

The other slight issue is that due to the chip scale packaging used the 'top' of the chip actually becomes the bottom and get's bond pads added using a redistribution layer.
So you'd want to put the antenna on the lower layers which is still possible of course just a bit more challenging

But having said all that it would still be very cool to see someone try it and see how it turned out

In any case, I expect better RF pin options going forward on the future shuttles

Ah good point about the chip-scale packaging.

I have seen a couple papers about 2.4ghz antennas in about 0.77mm by 0.66mm, which should fit in the user project area...

Incentives get weird when you're not paying by the mm^2 :)