Introduction

Stage 0: General Setup and Project Prep

Hardware & Equipment You Need
pulling in hardware alternatives from ‘iterating and improving’, make this page overall more easy to read
Storing Baseline Data
Setting Up Your Raspberry Pi or other devices
Setting Up openaps and dependencies

Stage 1: Logging, Cleaning, Analyzing Data

Setting up visualization and monitoring
Include NS related alerts and openaps system alerts for looping/not *DIYPS style net basal insulin reports should get added to NS reports
Setup script
github.com/openaps/oref0/blog/master/bin/ns-uploader-setup.sh with some tweaks)
Analyze your existing data
basals & ISF, use NS report or Dex Studio/Clarity, whatever people prefer

Stage 2: Creating an Open Loop

Configuring and learning to use openaps tools
much of this covered by script, so mostly pulling suggest and enact from old version
Running an open loop with oref0
most of the content previously from creating loop and retry logic
Building preflight and other checks

Stage 3: Understanding Your Open Loop

Determining basal recommendations
https://github.com/openaps/docs/blob/dev/docs/Build-manual-system/Understand-determine-basal.md
Run for enough days to decide what max basal should be
How often did you disagree or counteract what the loop was recommending?

Stage 4: Starting to Close the Loop

Cron instructions (or daemon)
this is obviously new and needs a lot of thought/contributions
Observing the closed loop
Starts with LGS mode only
Don’t move forward until observed for 3 days
If any consistent net negative low insulin amounts (check NS reports), then you may be overbolusing for meals and/or basals are too high 1.5-2 hours before the lows are occurring.
Troubleshooting the closed loop
Info from validating & testing

Stage 5: Tuning the Closed Loop

Going beyond low glucose suspend mode
(moving beyond max iob = 0)
Tuning your targets
After 3 days, at least one night with zero low alarms should occur before you consider dropping the max target from 160.
Only drop 10 points at a time, again observing the outcomes over a few days.
If going low three+ days, the “min” target may need to be raised

Stage 6: Iterating on the Closed Loop

So you think you’re looping
3 nights consecutively with no major system problems and at least 1 night without low alarms – fill out form and get number
Tests for when you are running for 24 hours
Run 24 hours for 3 days (doesn’t have to be consecutive) to get a feel for how the system behaves when you do meal boluses with the system on
Observe again net basal insulin delivery after 1 week of looping; see times of day when the system gives net negative; again, may be overbolusing and/or basals too high
Enabling meal-assist and other advanced features
First, regular carb entry required (so don’t accidentally trigger wtf-assist after meal-assist & co are enabled
Confirm 1 week successful daytime looping
Enable meal-assist & co to enable more aggressive high temps (setup script)
Understanding the –assist features & how it works
Close observation for a week

Resources

Provide feedback

Saved searches