DK does a lot of projects. The first, and most important, step to any project is to get a global perspective of the objectives
. Because it's from the global
perspective that any compromises must be negotiated.
Yesterday, I almost posted in someone's thread about some very challenging problems they are about to face, but then thought better of it - that would have removed a valuable learning process - problem solving. They will be better off, down the road, from the experience of problem solving than from being told the problems and solutions, first.
The Aquavac project is a great example of the problem solving process.
First, my global objectives are to develop a tool with certain capabilities that will fit into the overall scheme of optimized automation in shrimp keeping.
So to this end, I needed a tool which would:
- clean an incoming water stream, (debulk the mulm from a tank)
- vacuum up small particles and isolate them, (remove substrate for recycling or clean algae patches or plant fragments)
- be usable on any of my tanks
- be robust and dependable
- be cleanable/disinfectable easily
Now, in any building project I tend to use some global parameters:
- make the construction heavy duty
- design in a modular fashion
- make any aspect of the design easily alterable or replaceable to the extent possible
- use commonly available, inexpensive materials
- keep it as simple as possible, given the above
- good, but cheap - good comes first
Aquavac v.1.0 used a bucket with gamma seal I already had, that was supposed to be robust and air tight. The capacity was within the range I wanted. I had the bulkheads and plumbing parts, and bought what I thought was a suitable pump to drive the AV with enough power and flow to accomplish the technical goals.
Problem was, the gamma seal would not seal, mostly under pressure but also under vacuum. Both conditions occur - pressure as one primes the chamber, then later vacuum as the pump is circulating. The whole thing wasn't at all elegant, but would have been just fine, had the gamma seal worked. I tried a number of fixes, and couldn't get it to seal in the manner I needed.
Aquavac v.2.0 utilised a different chamber, a locknlock container that I knew would create an airtight seal under vacuum. It was moderately robust, but I worried about that. I did, however, discover that it would pop the seal and leak under pressure, so it went bust. Bummer, I wasted a locknlock, however I had bought it clearance and its cost was maybe $3.
AV v.2.0 solved two other problems from v.1.0.
First, it had a see-through chamber, so when the filter sock gets full one can see that and change it out, easily.
Second, in v.1.0, the tubings were coming off the apparatus straight up and were kinking over time when they flopped over. So v.2.0 mounted them horizontally rather than vertically.
What 2.0 lacked, I soon discovered, was a bleed valve that would become necessary to make pump priming easy. I could have installed one, but the big picture told me to move on from the locknlock container - every hole in a container is a point of potential failure. That FOUNTAIN GRADE pump driving the system is not to be disrespected - a failure will result in a LOT of water displacement in a short time.
It also had an awkward footprint and a layout that was vulnerable at a number of stress points - not robust from this perspective. DK hates babysitting. DK likes set-it-and-forget-it, and not having to worry about some weakness.
Time to get REALLY serious about a robust chamber.
The hilarious thing is, when DK went to her bins and barrels, thinking "pressure vessel, AIR TIGHT seals under PRESSURE and VACUUM, transparent, robust, about a gallon volume" - she realized she had bought the very part ORIGINALLY for this project, to use, but then had relegated it to parts, due to capacity
. It had all the other characteristics needed, easily. (Such a part IS available in a larger capacity, but isn't exactly cheap, so she had moved on, not ready to compromise YET.)
That was the moment, recently, of forehead smacking and laughing hilariously.
OK, this post is long enough, for now. Plus, I need to shoot some pics later for illustration.