Close

reservoir musings

A project log for automated gardening - reservoir infrastructure

Building Reservoirs and Weighing Water for level gauging and flow metering.

helgehelge 05/03/2018 at 11:410 Comments

It's been put off long enough - let's see what can be done in terms of reservoir containers. The whole reservoir, while being an unpleasant necessity, is supposed to be inconspicuous and no-fuss to handle - funnel equals failure. "Just add water".

Units sold as "scientific water bath" or small-scale laboratory recirculating chillers have a lid on top that you can just lift off or open to refill and I like that idea. Somewhere between the lid and the 3/4" port there are the wide mouth lids:

The chiller in the lab has one of those and although you can refill with a bucket it's always a challenge not to spill anything.

Once you're willing to add a permanent shroud (e.g. a matching 3D printed part), new options are on the table. Maybe a canister with larger ID hose nipples would do the trick? One for venting and extrating tubes access while the other would be connected to the funnel?

As fas as sloshing and splashing is concerned I like the idea of having a reservoir canister... so I got a 10L canister, only to change my mind. Ironically, it fits the 30x40cm² euro box very snugly:

And once we're there it becomes rather obvious what this story will digress into:

Maybe the simplest and most useful option is to just use a plain euro box- It's also hard to beat the sheer volume of that thing, plus it keeps out the light and thus suppress algae growth.

Add a laser cut divider to control sloshing. Maybe even a ready-made one will suffice?

A bonus of the laser-cut divider would be the ability to make the pockets unevenly shaped so they don't share a common fundamental sloshing resonance or add holes to create turbulent damping. Maybe that's me taking it a bit too far.

Handling:

That should do it?

But then there is the grip opening which just screams "extendable funnel tray" at me, with the "open the darn lid" method as a fallback.

Discussions