Pioreactor dev log #16 - Cables

As we start to take a more critical eye at our project, we are revisiting parts of the Pioreactor that started as temporary hacks, but now need to be updated. Two really positive changes that we've decided on are our cable assemblies. Let me explain. 

The most basic use of the Pioreactor involves one LED, and two photodiodes. These components need to be connected from the glass vial to the Pioreactor HAT. For R&D, we are using two-wire jumper cables: 

This worked, but it had a lot of drawbacks: it looked sloppy, and required us to build custom assemblies with an LED on one end, and our polar connector on the other. 

Similarly, our heater PCB was connected to the Raspberry PI's GPIO pins and a PWM output, again using jumper cables. You can barely see this in the photo above (it's the white/brown/red/yellow/black cable). Again, this worked fine for R&D, but we are adults now. We can do better. 

Overall, we can a lot of colourful jumper cables, and everything just felt ready to fall apart on a light touch! 

Settling on cables

Through a lot of back and forth with my co-founder, we decided on nice solutions. 

First, the LED and photodiode assemblies. Our female connectors on the HAT are the lovely 2-position TE Connectivity AMP connectors, seen below:

There exist pre-assembled, flex-cable male connectors, like this: 

These cables feel really good, look really good, and are still pretty low cost (relative to buying the components ourselves and paying for assembly / manufacturing ourselves). They are also accessible to users to buy, if they wish to change LEDs or photodiodes. 

The heater PCB now has a flex cable, instead of the jumper wires. This cable connects to a corresponding connector on the HAT, so no more using GPIO pins and getting connections wrong. 

Overall, these cable changes make product feel more secure and reliable. And signal integrity is improved, too, since the cables are higher quality and shorter.