[Here] is the chip I'm using. Below is the symbol I just made:

Now to make the footprint and 3D model...

Hmm? I wonder if I could've used this to create the BMI160 footprint...

The 3D pads are understandably wrong, but it looks like there's promise with the footprint.

It's actually just as I manually modelled. Amazing.

It's just a shame that I have to click each pad line individually when projecting it. For some reason I couldn't just click+drag to select all. To save time, I just projected half and circular patterened the other half.

---

Anyway, back to what I was supposed to be doing... connect these pins... join these lines... done.

Talk about Team Close (TM). Thankfully, I can populate half the 595 with the pins that just power down or reset other components, meaning that they're very unlikely to need to be changed often. There is then 1 free pin I can use for the latch. I should be fine with nSleep and DAC_SD, but I'll have to check to make sure that the 1.47" LCD and ADC don't expect to be communicating at the time of the reset pin.

For the ADC, it seems that as long as it has an actual clock source (and I don't try and share the ADC clock with the SPI clock like I proposed doing in the past) I should be fine here. Actually, looking at the diagram just above it in the datasheet, it looks like the SPI clock isn't even on all the time:

I wouldn't be able to share the 2 sources anyway, since the MIP display expects maximum 2Mhz SPI clock and the ADC uses 8.192Mhz for it's CLKIN.

On the LCD side, it actually looks like I could even put D/C on the 595 if I needed to, since it's just a toggle between telling the display if you are sending either data or a command. I can't imagine you'd need to rapid toggle for that. Still though, the less pins are on the 595, the faster it is to update all of them.

---

Me about a week ago was worried that "How am I going to get Tetent done by the end of summer at this rate?", but wow now I've almost finished the primordal schematic! I've now got to 

• do the power circuitry of the ESP32 (if I need more than just a bypass capacitor), 
• create pads for the load cell and magnetometer,
• create the potentiometer for position selection,
• expose the USB pins (and UART if I need those for programming),
• add test pad connections,
• remove the grounds in the Tetrinsic_PCB connector, 
  • It's probably better to go with either another 3V3 or another pin
• and look into the TB chip connections. 
  • It's likely that it too will have some non-I2C pin that needs to be connected to the MCU.