Don't be stupid and try and repair a delicate piece of PPE, just buy a new one!<\/p>\n<\/div>\n
Device overview<\/h3>\n
The Senko SP-SGTP (datasheet<\/a>, user manual<\/a>) is a personal gas monitor that is designed to measure the concentration of a specific chemical in the atmosphere and trigger an alarm if that concentration falls outside of a specified \"safe\" range.<\/p>\n The \"O2\" in the product code denotes that my monitor contains an oxygen sensor. I use it when exploring abandoned mines. In these confined spaces natural bacteria can consume oxygen as they break down organic matter, reducing the concentration of oxygen in the environment and replacing it with poisonous CO2. Not wanting to asphyxiate and die, I therefore use a gas monitor to make sure that the air I'm breathing contains enough oxygen to live (and therefore probably<\/em> not enough CO2 to do me harm). There are other risky gases that can build up in confined spaces, particularly abandoned mines, but they are much less likely to occur in the mines that I'm exploring (or have less severe effects than asphyxiation), and multi-gas monitors are expensive, so I have made a decision not to monitor for them and to control my risk exposure in other ways.<\/p>\n The second \"P\" in the product code denotes that this is a maintainable device rather than a disposable one - i.e. you can (in theory) switch out the battery and sensor if they fail.<\/p>\n Senko<\/a> is a South Korean company with quite a good reputation for making this type of PPE, but this particular device is definitely on the budget end of their offering.<\/p>\n At first glance this device certainly does appear to be pretty maintainable. The back of the case can be removed with 6 identical machine screws, giving easy access to the battery which just pops out for replacement.<\/p>\n [insert image here]<\/p>\n The circuit board is held into the front of the case with 2 shorter self tapping screws. Removing these allows the board to be removed, allowing the filter on top of the sensor to be removed and replaced.<\/p>\n [insert image here]<\/p>\n But what about the sensor itself? On further inspection it appears to be a fairly standard three pin electrochemical sensor, with the pins labelled on the board:<\/p>\n My life has been hijacked by my smartphone. In some ways it's my fault, because I made a series of incremental choices that led me here, but in many ways it's not my fault, because each individual choice wasn't unreasonable. How could my poor primitive primate¹ brain possibly know what it was getting itself in for?<\/p>\n The 2025 Smartphones in Education (SIE) report<\/a> suggests that \"the average student now spends 5.5 hours per day on their phone-equivalent to 25 years of their lifetime\", and perhaps more strikingly, \"68% want to reduce [their] phone usage\" but presumably feel unable to.<\/p>\n My numbers aren't quite that bad, but they're still really bad<\/em>, and I'm 30 and supposed to be in control of my life in a way that adolescents aren't. So what am I doing about it?<\/p>\n What I'm complaining about are my own bad habits, and it's interesting just how unconscious these are. My mind can be entirely elsewhere but I can still unlock my phone and open an app with zero conscious effort - it's baked-in muscle memory. Fingerprint readers and facial recognition unlock have facilitated this; it's easier than ever to unlock your phone without having to engage any mental faculties.<\/p>\n Recognising that these are habits is helpful because it gives us the tools to fight them, and that is to interrupt them.<\/p>\n It's very difficult to make it impossible to access Facebook, or LinkedIn, or BBC News on a smart-phone - but you don't need to make it impossible, you just need to interrupt the habit and give your conscious brain the chance to object. So I've uninstalled my worst offending apps² and blocked their web equivalents using uBlock Origin on my phone's web browser.<\/p>\n Now, when my muscle memory gets me unlocking my phone and opening something that wastes my life away, I am interrupted for just long enough to be slightly disgusted with myself. I am incredibly grateful for this disgust, because it is negative enough to prevent me from just working around the simple block that I've implemented.<\/p>\n This isn't perfect. I want to know what's going on in the world, and blocking my news outlets prevents me from doing so. I want to hear about peoples caving discoveries on UKCaving<\/a> and Buddle Pit<\/a>, and that's harder for me now. There are sacrifices to be made with this, but they're not that substantial. Ultimately, I'm happy to be making them, because I think it's making me happier, and healthier, and more interested, and more interesting<\/em>.<\/p>\n I think \"screen time\" (particularly \"phone screen time\") is quite a nice loss function to be trying to target, because it doesn't have that many negative consequences. Even if I stopped using my phone entirely I don't think my life would look all that different - there would just be some more areas of inconvenience.<\/p>\n So if I could ditch my smart phone entirely, why...<\/p>","image":"\/user\/pages\/05.blog\/screen-time-loss-function\/5868625180358723881.jpg"},{"title":"Unsolicited opinions: caving torches","date_published":"2024-12-14T11:14:00+00:00","id":"https:\/\/cooper-davis.net\/blog\/caving-torches","url":"https:\/\/cooper-davis.net\/blog\/caving-torches","content_html":" Caving (and mine exploring) is an unusual outdoor pursuit in that it's time spent almost entirely in darkness. As such you need a good source of light - usually a torch.<\/p>\n After 10 years of caving (and mine exploring) I now feel that I have earned the right to give my unsolicited opinion on what makes a good caving torch.<\/p>\n TLDR<\/strong>: Caving introduces some niche requirements for torches. Feel free to skip to my recommendations<\/a>.<\/p>\n Before I make some recommendations, let's look at some things that you need to consider when choosing a caving light. I've split this up into three lists; must haves<\/em>, nice to haves<\/em>, and don't wants<\/em>.<\/p>\n Brightness<\/strong>: a caving torch must produce enough light to cave with. This isn't actually all that much light, in fact early cavers and miners used candles as their only source of light. A candle has a luminous\nflux<\/a> (good luck with that rabbit hole!) of about 12.5 lumens. They had a horrible time though, so let's go for a minimum brightness of 10x that: 125 lumens.<\/p>\n However... lumens alone aren't all that useful as a metric, as it doesn't consider the area that this light is spread across. A candle spreads a small number of lumens over a large sphere, whereas a laser pointer crams it all into a tiny spot. Most torches have a tolerable beam-shape, though, so I'll relegate beam shape to the \"nice to haves\" section.<\/p>\n<\/li>\n Robustness<\/strong>: you (really!) need your torch to survive your trip. For me, this means it has to be waterproof<\/strong>. Thankfully we have a nice ingress protection standard<\/a> that allows us to directly compare waterproofness. As you may well be wading, swimming, crawling through puddles, or even going entirely underwater for sumps and ducks, I think IPX7 is a good minimum.<\/p>\n Robustness should also include impact resistance<\/strong>. Now we do have a standard for this too<\/a>, but it's not very human readable, and has such hasn't seen very widespread adoption. Instead you have to judge robustness on manufacturer ~lies~ claims, user reviews, and an instinctive assessment of the material and design of the product itself.<\/p>\n<\/li>\n Head mount<\/strong>: you need your hands when you're caving, so your torch has to live on your helmet. However, I have chosen to use \"head mount\" here, because your backup light(s) shouldn't rely on you still having your helmet - what if you've dropped it down a pitch<\/a> and smashed it to smithereens?<\/p>\n Head mount also encompasses weight<\/strong> and bulk<\/strong> as you have to be able to wear it on your head for hours at a time. I don't know what my limits are for weight really, except having worn a 160g headlight (inc. batteries) once or twice I wouldn't want to exceed that by much.<\/p>\n<\/li>\n Friendly UI<\/strong>: so that you can access all those nice brightness modes. This is a matter of preference, but it's really important that your torch is...<\/p><\/li><\/ul>","image":"\/user\/pages\/05.blog\/caving-torches\/emisar-dw4.jpg"},{"title":"Viseron and the Tapo C310","date_published":"2024-08-26T16:03:00+01:00","id":"https:\/\/cooper-davis.net\/blog\/viseron-and-the-tapo-c310","url":"https:\/\/cooper-davis.net\/blog\/viseron-and-the-tapo-c310","content_html":" A TP-link Tapo C310<\/a> security camera streaming audio\/video to the FOSS Viseron NVR<\/a> software running on an Odroid-C2<\/a> single board computer\/home server.<\/p>\n Because I couldn't resist the temptation of a technical challenge, despite my dislike of surveillance in general (sorry principles!).<\/p>\n The Tapo C310 makes a couple of authenticated RTSP streams available at These streams are a bit quirky; they're fixed at 15 fps, and use the Still, this is enough for us to plug into a Network Video Recorder (NVR) program which can pull the feed, process it, and save it.<\/p>\n Viseron is a fairly straightforward open source NVR program that should be just the job. I like it because it's modular, written in python, and quite efficient. This means that you don't have to play with all the bells and whilstles of AI object detection or graphics cards if you don't want to. It'll manage all the boring stuff, like deleting old video files, making the setup fairl simple.<\/p>\n Even better, it's shipped in Docker containers so it's super easy to spin it up and get it going without worrying about dependencies, daemons, or other faff like that! So we spin up a container<\/p>\n Note that we're binding some directories on our filesystem as volumes in the container. This allows the recordings and config files to be stored on an external HDD plugged into the server, rather than doing all that writing on fragile flash memory (which is a great way to kill your micro-SD card\/eMMC!). We're also port forwarding on 8888 to allow us to view the admin interface on the local network. And finally we're using the What about setting up Viseron itself? Well the documentation for configuring Viseron<\/a> is pretty good, and as we're only using the basic First...<\/p>","image":"\/user\/pages\/05.blog\/viseron-and-the-tapo-c310\/vaporwave.jpg"},{"title":"3D Scanning of Abandoned Mines","date_published":"2024-08-14T13:13:00+01:00","id":"https:\/\/cooper-davis.net\/blog\/3d-scanning-abandoned-mines","url":"https:\/\/cooper-davis.net\/blog\/3d-scanning-abandoned-mines","content_html":" I have recently been working on a project exploring how I can use the laser surveying tool that I'm developing to produce high resolution surveys and 3D scans of abandoned mines and caves.<\/p>\n With the support of an innovation grant from the Deep Digital Cornwall<\/a> project, run by the experts at the Camborne School of Mines and the University of Exeter, I have been able to accelerate the development of my surveying tool and obtain some more advanced hardware for collecting high resolution LiDAR scans at survey stations.<\/p>\n Whilst the surveys of the mines will take a significant amount of time to draw up, the first 3D scans are now processed and ready for viewing. I was very lucky to be invited to view these models using DDC's Visualisation Suite<\/a>, which gives an incredible virtual reality view of these environments, which otherwise require considerable ropework and mine exploration skills to experience.<\/p>\n I will also make these models publicly available so that you can explore them at home.<\/p>\n Stacked Deads<\/p>\n This scale model shows a level in Wheal Pell, an abandoned mine in the St Agnes Consols Sett in West Cornwall. Waste rock (\u201cdeads\u201d) are stacked up against one wall.<\/p>\nWhat's inside?<\/h3>\n
\n
![\"Screenshot%20From%202025-08-26%2017-10-56\"](\"\/user\/pages\/05.blog\/screen-time-loss-function\/Screenshot%20From%202025-08-26%2017-10-56.png\" "\\"Bar")
<\/p>\nMuscle memory<\/h3>\n
Add friction<\/h3>\n
Is it worth the sacrifices?<\/h3>\n
It's personal<\/h3>\n
Considerations<\/h2>\n
Must haves<\/h3>\n
\n
![\"IPRATING_uitleg_MM\"](\"\/user\/pages\/05.blog\/caving-torches\/IPRATING_uitleg_MM.png\" "\\"Ingress")
<\/p>\nWhat?<\/h3>\n
Why?<\/h3>\n
![\"vaporwave\"](\"\/images\/9\/f\/3\/0\/6\/9f3066a89a1036f82cd97eed1ad1a5b1afc0e470-vaporwave.jpg\" "\\"vaporwave\\"")
<\/p>\nHow?<\/h3>\n
Tapo C310<\/h4>\n
rtsp:\/\/camera.local.ip.address:554<\/code>:<\/p>\n\n
\/stream1<\/code>: a high res stream (1920x1080 or 2304x1296 depending on how you set it up)<\/li>\n\/stream2<\/code>: a low res stream (640x360)<\/li>\n<\/ul>\npcm_alaw<\/code> audio and h264<\/code> video codecs. This audio codec choice is a pain, as it means you have to re-encode if you want to put it in .mp4<\/code> containers rather than .mkv<\/code> containers, which aren't as well supported (I'm looking at you, Firefox!<\/a>).<\/p>\nViseron<\/h4>\n
docker run \\\n -v \/mnt\/nvr\/segments:\/segments \\\n -v \/mnt\/nvr\/recordings:\/recordings \\\n -v \/mnt\/nvr:\/config \\\n -v \/etc\/localtime:\/etc\/localtime:ro \\\n -p 8888:8888 \\\n --restart unless-stopped \\\n --detach \\\n --name viseron \\\n roflcoopter\/viseron:latest<\/code><\/pre>\ndetach<\/code> and restart<\/code> flags to keep the process alive in the background.<\/p>\nffmpeg<\/code> camera and background_subtractor<\/code> motion detection components the setup is quite simple:<\/p>\n## Cameras\nffmpeg:\n camera:\n hillview_front:\n\n ### Recording stream\n name: Tapo C310\n host: cameras.local.ip.address\n port: 554\n path: \/stream1\/\n username: username_set_in_tapo_app\n password: password_set_in_tapo_app\n\n audio_codec: aac\n codec: h264\n fps: 15\n width: 2304\n height: 1296\n\n ### Low-res stream\n substream:\n path: \/stream2\/\n port: 554\n\n audio_codec: aac\n codec: h264\n fps: 15\n width: 640\n height: 360\n\n recorder:\n lookback: 20\n extension: mp4\n\n## Motion detectors\nbackground_subtractor:\n motion_detector:\n cameras:\n\n hillview_front:\n width: 320\n height: 180\n\n trigger_recorder: true\n fps: 1\n area: 0.01\n threshold: 15\n alpha: 0.1\n\n## NVR\nnvr:\n hillview_front:<\/code><\/pre>\n\n