A cranky biologist who means well. My hobbies include long walks off short piers and anything science related.

  • 0 Posts
  • 81 Comments
Joined 1 year ago
cake
Cake day: June 25th, 2023

help-circle












  • Well the energy recovered from plant lights is actually really pathetic when using standard panels. Had to switch to amorphous panels which capture just enough energy to keep the battery sustainably charged day after day under typical loads.

    Overall, the system reclaims just enough energy to run the air pumps for up to about 15 minutes total daily use. That’s plenty to achieve the goal of giving the plants a short soak in the nutrient medium which just drains away when the pump stops. It also facilitates filtered fresh air exchange and aeration of the liquid medium which is vital.

    The next stage will be to make the system more thrifty with power (MOSFETs instead of relays and other tweaks to reduce current leakage when the microcontroller is asleep).

    I want to add the ability to inject CO2 and monitor levels before I consider the product full-featured. Recent studies are showing that can result in huge gains in growth rates. Based on my estimates, a 16g CO2 cartridge (think paintball or whipped cream) can keep a typical 3-liter culture vessel at 4000ppm (10x over atmosphere) for a year, even if you completely flush with fresh air twice daily.

    Thank you for your question and interest in my latest obsession.


  • My hobbies mostly sit at the intersection of plants and electronics/programming esp. microcontrollers and managing fleets of them.

    Im obsessed with making things grow and relatively simple types of automation can make a huge difference to a plant. A trickle of water applied at the right frequency can turn an unforgiving sun-scorched balcony into a garden.

    Im currently working on prototypes of a device destined for mass production. It’s a power unit for a temporary immersion bioreactor used in plant tissue culture. The benefit of my approach is that the power unit can work with almost any growing container and the unit doesn’t need any power hookups.

    The unit is powered by the plant grow lights and my Mark I prototype proved it can harvest enough energy to perform any published temporary immersion protocol I have seen.

    I think this qualifies as ‘weird’ because it usually requires explanation to justify the ‘why’ of this project. Plant tissue culture is not a common interest.

    But if you want to plant a trillion trees and ‘save the planet’, we will need to develop some new propagation methods. This is my little attempt to address some of that need.







  • meyotch@slrpnk.nettoLinux@lemmy.mlSystemd timer unit
    link
    fedilink
    English
    arrow-up
    2
    ·
    7 months ago

    Your systemd file looks ok, but I think it’s doing exactly what you are telling it.

    The solution may lie in the backup.service. Is that code you can modify? The OnCalendar=weekly doesn’t specify when in the week the service should run so that config may be vague.

    If I understand the desired function here, you will need the service up all the time. It will just wait politely and occasionally run the specific backup script. It’s up to the backup script to determine when the last backup was made and either exit early because it hasn’t been a week or run the backup and reset a flag file.

    At least that’s the approach I would take. Systemd is a very vigilant, but very stupid, service manager. It just watches and triggers services based on just a few criteria. Any logic more complex needs to go in the service itself.