{"id":273,"date":"2021-12-07T18:17:01","date_gmt":"2021-12-07T22:17:01","guid":{"rendered":"http:\/\/www.functionalprototype.com\/news\/?p=273"},"modified":"2022-09-06T17:57:58","modified_gmt":"2022-09-06T21:57:58","slug":"self-configuring-sensors-self-configure-and-thats-a-problem","status":"publish","type":"post","link":"https:\/\/www.functionalprototype.com\/news\/2021\/12\/07\/self-configuring-sensors-self-configure-and-thats-a-problem\/","title":{"rendered":"Self-configuring sensors self-configure, and that’s a problem"},"content":{"rendered":"

Little backstory on the sensorBox project.\u00a0 Things were going well until early in the Summer when the sensors threw out what I’d call garbage data.\u00a0 CO2 levels so high I should have headaches on a day whet the windows were all open is when I thought, “maybe I’m getting exactly what I’m paying for: sensors where the low price reflects that are disposable\/replaceable”.<\/p>\n

I worked on some other projects, did a little more research, and learned something about the CCS811, not well documented, if at all:<\/p>\n

It continually<\/em> self-configures.<\/p>\n

The docs are pretty spare, but it’s clear that you calibrate by reading the baseline in a stable, normal CO2 environment; store it somewhere permanent; then use that value to set the baseline when you launch your code.\u00a0 That works pretty good when you’re launching the code at least once a day or several times a week — you’re repeatedly resetting the baseline.<\/p>\n

What if you let it run for weeks without resetting?\u00a0 It can reset the baseline in an high CO2 environment and now all the readings are off.<\/p>\n

Here’s how I started figuring things out — I started doing a get_baseline() every hour and logging it in the events file.\u00a0 Here’s the event log for one night when I baked a potato in a gas oven using my sensorBox in the kitchen.\u00a0 Entries starting with a comment (“#”) are automatic, the rest I did using the log-entry.py tool.\u00a0 The baseline is clearly wandering in a kitchen where I’m not doing anything, then gets worse when I start using my gas oven in convection mode.<\/p>\n

$ more kitchen-events-2021-11-12.log
\n#boot time 2021-10-31 14:25:09
\n#description of event
\n1636693203,#baseline at day rollover 48713
\n1636693203,#baseline 60 min read 48713
\n1636696812,#baseline 60 min read 48713
\n1636700421,#baseline 60 min read 48713
\n1636704031,#baseline 60 min read 48713
\n1636707640,#baseline 60 min read 48713
\n1636711250,#baseline 60 min read 48713
\n1636714859,#baseline 60 min read 48713
\n1636718408,#baseline 60 min read 48691
\n1636722017,#baseline 60 min read 48670
\n1636725627,#baseline 60 min read 48670
\n1636729236,#baseline 60 min read 48670
\n1636732845,#baseline 60 min read 48655
\n1636736455,#baseline 60 min read 48655
\n1636740004,#baseline 60 min read 48655
\n1636743613,#baseline 60 min read 48630
\n1636747223,#baseline 60 min read 48572
\n1636750831,#baseline 60 min read 48572
\n1636754438,#baseline 60 min read 48572
\n1636758046,#baseline 60 min read 48572
\n1636761654,#baseline 60 min read 48572
\n1636763990,set oven to 350F
\n1636764253,turned on kitchen exhaust fan
\n1636764904,oven to 350 convection
\n1636765162,added potato
\n1636765201,#baseline 60 min read 48531
\n1636768811,#baseline 60 min read 48416
\n1636769479,remove potato from oven, turn off oven
\n1636772420,#baseline 60 min read 48416
\n1636772792,CO2 and tVOC aren’t dropping, forcing a restart
\n1636772797,sensor-box.py started at 2021-11-12 22:06:37.084918
\n#boot time 2021-10-31 14:25:09
\n#description of event
\n1636772798,#baseline at day rollover 48713
\n1636772985,turned off exhaust fan
\n1636776046,#baseline 60 min read 48717<\/p><\/blockquote>\n

Here’s a simple graph, you can see at the end where things go wonky as the self-configured baseline self-configures in a high CO2 environment:<\/p>\n

\"kitchen<\/a><\/p>\n

It took some iterations and checking the baseline and it turns out it’s best to reset it every ten<\/em> minutes.<\/p>\n

Here’s me baking another potato but resetting it every ten minutes:<\/p>\n

\"kitchen<\/a><\/p>\n

Here’s a clip from the event log with debug logging and how often the baseline changes while I’m using the oven.\u00c2\u00a0 It changes once while the oven is preheating, then nine more times while the potato is baking.\u00c2\u00a0 Each of those times is a self-calibration *changing* the baseline.\u00c2\u00a0 Since I can only read the sensors once a minute, resetting every ten minutes seems like a decent adjustment.<\/p>\n

1638830503,set oven to 350F convection
\n1638831035,#baseline 10 min read 48635
\n1638831035,#baseline 48635 does not match baseValue 48713, resetting
\n1638831132,put potato in oven
\n1638831637,#baseline 10 min read 48654
\n1638831637,#baseline 48654 does not match baseValue 48713, resetting
\n1638832238,#baseline 10 min read 48452
\n1638832238,#baseline 48452 does not match baseValue 48713, resetting
\n1638832840,#baseline 10 min read 48324
\n1638832840,#baseline 48324 does not match baseValue 48713, resetting
\n1638833441,#baseline 10 min read 48226
\n1638833441,#baseline 48226 does not match baseValue 48713, resetting
\n1638834042,#baseline 10 min read 48168
\n1638834042,#baseline 48168 does not match baseValue 48713, resetting
\n1638834644,#baseline 10 min read 48093
\n1638834644,#baseline 48093 does not match baseValue 48713, resetting
\n1638835246,#baseline 10 min read 48072
\n1638835246,#baseline 48072 does not match baseValue 48713, resetting
\n1638835623,turned on exhaust fan in kitchen
\n1638835686,opened oven to check potato temp
\n1638835847,#baseline 10 min read 48062
\n1638835847,#baseline 48062 does not match baseValue 48713, resetting
\n1638836142,opened oven to check potato temp
\n1638836449,#baseline 10 min read 48098
\n1638836449,#baseline 48098 does not match baseValue 48713, resetting
\n1638836509,opened oven to check potato temp
\n1638836556,0ven off
\n1638836565,took potato out of oven
\n1638837050,#baseline 10 min read 48260
\n1638837050,#baseline 48260 does not match baseValue 48713, resetting
\n1638837651,#baseline 10 min read 48713
\n1638838253,#baseline 10 min read 48713<\/p><\/blockquote>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

Little backstory on the sensorBox project.\u00a0 Things were going well until early in the Summer when the sensors threw out what I’d call garbage data.\u00a0 CO2 levels so high I should have headaches on a day whet the windows were all open is when I thought, “maybe I’m getting exactly what I’m paying for: sensors […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"footnotes":""},"categories":[21,34],"tags":[42,43,45,44],"class_list":["post-273","post","type-post","status-publish","format-standard","hentry","category-design","category-environmental","tag-ccs811","tag-co2","tag-sensors","tag-tvoc"],"_links":{"self":[{"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/posts\/273"}],"collection":[{"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/comments?post=273"}],"version-history":[{"count":0,"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/posts\/273\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/media?parent=273"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/categories?post=273"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.functionalprototype.com\/news\/wp-json\/wp\/v2\/tags?post=273"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}