I have a couple of very busy weeks coming up and I wont be able to make much progress on this project. So here is an update:
-
HX711 and HX717 support has been implemented. I looked at a log of different implementations but found that none of us had gotten it exactly right. But of course Prusa had to get it right and their implementation is a great resource. Looking into this also led me to notice that Prusa is using the
Bchannel on the ADC to measure hot end temp in the MK4 (at least that’s what the code seems to say). Long discussion in this thread. Basically if we want to get an MK4 toolhead and graft it onto a Voron as a test platform (which seems like the fastest way to solve our
vs 
dilemma) we might have a sticky problem to solve. -
I set up a project backlog so that you can see what I’m working on and whats left to go.
-
I spent 3 days looking into Grafana and it was a waste of time. For the kind of high frequency data coming from the load cell, it’s display engine seems to be a bad fit. Basically any data arriving faster than 1Hz makes the graphs flash and become unusable. So I’m going to move to ObservableHQ notebooks. Still javascript but much more useful for doing data science than Grafana. Initial testing of the websocket connection works so another day or so of hacking to get pretty graphs again. (task link)
