Watering the background

One of the tricks I have been using to keep things growing everywhere in the paludarium is by regularly watering the background. In this post I will describe how I made a spray bar and how I mounted it inside.

The idea

In order to get water onto the background and have it nicely sift through I have been using a 12mm PVC tube with a series of 1.5mm holes drilled into it. Really simple, and works well even with the smallest of pumps:

Example from an older setup on how to use a spray bar

Mounting the PVC pipe

Now mounting the pipe has always been an issue: I always used aquarium suction cups for this, and you guessed it: They come loose over time. How to overcome this issue? Glue the pipe in place. But that has a drawback that when the pipe get clogged, you’re in for a project of its own cleaning it (or ripping it out and replacing it). To find a more proper solution, I turned to Fusion360 and the 3D printer to come up with this:

3D printed wall mount with a separate 12mm PVC tube clamp that you can slide into the wall mount.

I printed a two-piece tool for this: A wall mount and a PVC tube clamp. I printed 4 sets of these (for a 120cm distance to cover inside). I glued the four wall mounts in place using silicon glue. Now I can slide the PVC clamps in there, and replace them if needed. Next, I can insert the 12mm pipe:

The 12mm tube being held in place right on top of the paludarium using the 3D printed wall mounts and insert clips. The hole on top is used to feed tubing in and out, one of the tubes will go into this PVC pipe using a tube pillar (ordered but not here yet 😉 )

Finally I can start glueing in the background. Here I am used compressed fern root plates which give a nice nutrition for plants who grab hold to it, and it neatly holds water I pump on top:

Compressed fern root plates are glued under the pipe. These plates are excellent for growing plants on, plus they take on water really well and let it slowly sift through to the the bottom of the setup.

Sucking up water… From where?

Final piece of the puzzle is where to get the water from. In my new setup there is a specific sump-like section where all water that drains through the land portion. It is kept there and transported into the sewer if the level gets too high. Also, the aquatic part flows over into this section.

I will use this water to pump up to the PVC spray bar. Usually I’d put a small aquatic pump into the sump and have that push water up, but as I know now, pumps get clogged and fail over time, and replacing such a pump would be a project of it own.

For this reason I will be sucking water up using a small diaphragm pump like this one:

Small 12V diaphragm pump that will be used all over the paludarium; this one will suck up the water for the spray bar.

One final problem? The water the pump will take in will be “dirty” water; there might be particles in there which I may not want inside the pump, and certainly not inside the spray bar. So back to Fusion360 and the 3D printer once again:

A 3D printed inlet filter will make sure the pump and spray bar don’t get clogged by particles in the (dirty) water I use to spray onto the background.

On to the next tiny sub-project inside this huge paludarium project 🙂

The Aeryn module: Palu Air Conditioning

The Aeryn module was designed to allow air to be taken into the paludarium, heated and/or moisturized (mist) before it enters the setup. This is all part of the idea to have “no tech inside” but rather outside for easier access, maintenance and replace with updated versions.

What is Aeryn

Remember the sci-fi series “Farscape”? I originally designed this module named the Aether module. In honor of Farscape I renamed this module to Aeryn 🙂

The Aeryn module is a box-shaped module approximately 60x15x13cm in size, and is inserted into the hood above the paludarium. It lines up with two 12cm holes in the ceiling, where two fans draw in air from the outside. On the end of the module a 40mm pipe leads to the meshed strip on top of the paludarium where the conditioned air gets inserted.

Aeryn section 1: the air intake

The first section of the Aeryn module contains two 120mm fans, RPM regulated (and RPM measured back). These fans are put on top of the module, blowing down into the module. So first the airflow needs to be guided to flow sideways, which is accomplished using a 3D printed guide as shown below:

First section: After the air is pushed in from the top using two 120mm fans, this 3D printed guide directs the air to the left, on to the next sections of the Aeryn module.

Aeryn section 2 : Air heater

The second section of Aeryn is the air heater. First there is a (quick vase mode) 3D printed nozzle to force the air through a large (12x20x7cm) metal heatsink. Under the heatsink there is a 70watt heater element that automatically stays at 70 degrees when powered on:

The second section of Aeryn can heat the air using a large heatsink and a heater pad.

Aeryn section 3: Misting section

The third and last section of the Aeryn module can add mist (moist) to the air forced into the paludarium. There are three 24V mistmakers in a 3D printed holder:

3D-printed holder for three 24V mist maker units. This holder keeps the mist makers in place while allowing water to flow underneath the mist makers as they take in water from the bottom.

These mist makers get placed in a container where I’ll cut out holes for air inlet and exhausts (this is a work in progress). I might split up the mist maker control where I can activate one and/or the other two to more or less regulate the amount of mist.

Second I’ll have an external diaphragm pump able to pump (reverse osmosis) water into this container. There is also an overflow pipe that makes sure the container can never overflow. The overflow ensures the water level inside the container so the mist makers work optimally, and excess water gets inserted into the paludariums aquatic part. I might add a flow sensor to this lead to make the computer detect when there is enough water inside the container (and stop pumping), but I might also “loosely time” this as the overflow will secure the water level inside the container anyway.

This small container will have a lid and holes for air inlet and exhaust, as well as a 6mm hose to fill the container with reverse osmosis water and a 6mm overflow tube to ensure a proper water level for the mist makers.

The way the air leaves Aeryn

The exhaust is still under development; so far I have 3D printed a 40mm exhaust pipe which leaves the module at a 45 degree down angle. This will ultimately lead into a paludarium-wide transparent 50x50mm square tube which in turn will force the conditioned air into the paludarium through the top mesh strip which is 50mm wide:

The paludarium top mesh: a 50mm wide strip through which the conditioned air will be forced inside while keeping animals in 🙂

This concludes the description of the Aeryn module. It should allow me to regulate airflow through the paludarium, heat the air to hopefully direct it at the front window (so it won’t fog up). Finally I can add mist to the air I insert into the paludarium to quickly raise humidity as the sun sets. And combinations of all above; how to exactly regulate this to get to the proper airflow, air temperate and humidity remains to be seen. However, the paludariums “brain” will be capable to control all bits and pieces individually, so I guess I’ll find out soon enough!

Artemis v1.22: The new automation module

As you may have noticed, I love calling everything a “module” more and more. And with good reason: If you build a project of this size, splitting up in modules makes sense. It allows you to focus on PART of the problem, and replace the module if you find something smarter / better.

Required specs for the new automation hardware

As I was building out something new, I figured it should be big and bad enough to handle anything I want to throw at it, PLUS have room for future expansion. As I was falling in love with many smaller light sources again, and the RGB-CCT led panels, I came to the conclusion that whatever hardware I build, it has to have MANY channels I can use to control and possibly dim all hardware. Just as an idea, I wanted to use three RGB-CCT led panels, and those use 5 channels each for red, green, blue, cold- and warm white. So that is 15 PWM channels at 24V already maxing out an Arduino Mega (which has exactly 15 PWM outputs). So I needed more. Time to grab back to my faithful PCA9586 PWM I2C-based controllers. I would not place them onboard this time, but design them as separate modules that I could either place on top of the Artemis Controller, or optionally somewhere else (just running power to them and the I2C connection).

In the end, this is the huge wishlist I cam up with regarding the automation hardware:

Read more: Artemis v1.22: The new automation module
  • Needs to have a “smart” / web-enabled frontend;
  • Needs to have a “dumb” controller to make sure nothing floods / breaks;
  • Needs at least 3x RGB-CCT (24V) + 10x led 10W powerled (12V) outputs +5 pumps = 30 PWM outputs
  • Needs some digital outputs for control of valves (12V / 24V mix)
  • Needs an ability to control a single NeoPixel led strip (while I am at it – I can use this to draw stars, the moon, sunrise and sunsets);
  • Needs 4x 230VAC relais to switch anything that cannot be lower DC voltage (like aquarium heaters);
  • Needs digital inputs in order to measure water levels;
  • Needs to have multiple I2C channels so I can add SHT-XX humidity/temperature sensors;
  • Needs to have inputs for at least 3 LM35 temperature sensors;
  • Needs to have a very well controlled FAN speed controller for a (set of) 4-pin fan(s);
  • Needs to have an input for a Ph sensor and two more analog inputs for measuring things like water conductivity;
  • RealTime Clock (RTC) added to the hardware.

Can you believe this list!!! Impossible? As it turned out, no… Just a bit complex. I solved the number of I/O’s required by using an Arduino Mega Pro, and extending on the huge amount of PWM channels required with a set of two PCA9685-based expansion modules. One I will run at 24V (delivering 3x RGB-CCT + one spare output), the other will run at 12V (mainly for the 10W “follow the sun” power leds). Arduino’s onboard PWMs will be used for non-lighting things requiring PWM (like DC pumps that require speed control).

Here you can clearly see the design that was made: A large base PCB where an Arduino Mega Pro can be pinned on. Mostly all around the edge of the PCB you’ll find I/O’s. Here the additional boards (Raspberry Pi and two PCA9685 boards) are mounted in a way I can still reach the base PCB. They can also be mounted on top of the base PCB. More compact but harder to troubleshoot 🙂

Rewriting the existing software to work on this new hardware was relatively straight-forward; as the architecture remained mostly the same the majority of work was focusing on the new features. The PCA9685 wasn’t new, neither was the PCA9546A I2C bus-switch chip I already used in an earlier version of the board (1.01). It were the things like the NeoPixel led strip that REALLY took some time to build and get working. But ey, it is all great fun!

On the picture above you can see the base board with an Arduino Mega Pro pinned on top. The additional boards on top are two PCA9685-based boards each carrying 16 PWM outputs, and the third board is the Raspberry Pi 3b+.

Also visible in this picture are two RGB-CCT led tiles (one 30x30cm and one 120x30cm), A 10W power led with a collimator lens 3D printed to it on a heatsink, and a 4-pin fan that can be controlled (PWM) and measured back (actual RPM).

A closeup on one of the PCA9685 modules can be seen here:

Close up on one of the PWM extension modules, based on the PCA9685. These are called “Apollo modules” in honor of the original modules I used for lighting back in 2005 (!!). There are two variants: One with 3×5+1 channel (this one),and there is also one featuring 16×1 channels. All controlled by I2C, all fully opto-isolated.

All of the home-built PCBs were designed in KiCAD and were produced (including SMT mounting) by JLCPCB (Kudo’s to them for always creating perfect PCBs with all SMT part mounted for me; my eyes aren’t getting any younger 😉 ).

If you want to know all the things this board can do, stay tuned! Way more cool stuff coming soon!

Artemis hard- and software = WORKING!

After more testing, more writing code and more debugging I have finally managed to call the code inside the Artemis as COMPLETE. The last few bits and pieces came together, so next step will be to work on the higher level software inside the Raspberry Pi.

Artemis has been put through it paces. It all seems to work OK now, Although I did need to make some changes here and there…

Read more: Artemis hard- and software = WORKING!

NeoPixel Ledstrip: Working

So The “smart” led strips with the WS2812b smart leds on board refused to work initially. After a lot of debugging I figured out that the first NeoPixel, the onboard ws2812b was the culprit. It just would not work properly, even though it would pass the bitbanged stuff to the rest of the led strip in the correct way.

Read more »

The Canopy: Building a wooden hood

In the past I have always had wooden hoods on top of my paludariums which I call “the Canopy”. The latest one did not have a hood, as I placed lighting and fans in the open on top. For this new and big design that will not work, so once again it is off to the store to get wood cut to size.

Modules, modules everywhere.

In the past I used the Canopy to mount everything I needed directly in to. I even had a version that had a full-blown PC power supply mounted on top. Not this time. I want everything to be removable, modules everywhere. So I had plywood sawed to size not just for the Canopy itself, but also for up to 4 modules which slide into the Canopy. As I do not want to remove the canopy whenever I need to make a change, this time I will build a detachable front element which is held in place by magnets.

So far I have only two modules in mind, one that carries a few pumps (rain pump, backdrop moisturization pump), nothing too fancy.

The second module is more interesting. It holds the complete “air conditioning” system. There are two 120mm regulated fans in there, an air heater and two mistmakers. This is quite a complex module, as it will measure air going in (to get an ambient reading), and have multiple connections for osmosis water pumped through (for the mistmakers), connections for the air heater and of course the fans.

These modules will be covered in a later post.

Constructing the Canopy

For the Canopy itself, I just need some holes for feeding through cables, sliders to mount the modules, and a system rail in front where I can mount a series of power leds. Also, I’d like to maximize accessibility. The idea is that I’ll build an “inner layer” of plywood that sits on top of the paludarium, and a thinner “outer layer” that neatly falls over the edges of the glass paludarium, hiding the top two centimeters or so. First it is off to the inner layer construction. Yet it always seems to start with a saw:

Why do things always seem to start with a saw? Anyway, sawing some corner beams to form the shape of the Canopy.

Assembling the inner layer of the Canopy

Using the sawed corner supports it was relatively simple to build the basic construct. As usual the fit was ALMOST right. So I guess there will be some sanding and filling in my near future 😉

I needed to add some holes in the rear of the Canopy for feeding through cables and such. Not quite sure how big the holes should be, so I settled on a slit of 28mm high (just because I had a 28mm drill handy 😉 ). Hopefully all connectors I will ever use will fit through! In line with the other 2 holes I just added some holes in the back (yes, 28mm as well 😉 ).

Slit added to the rear for feeding through cabling and hoses. On top you see the two holes for the 120mm fans which will sit in the “air conditioning” module.

Furthermore I have added fan guards to the 120mm holes on top. The biggest struggle will probably be to get the air conditioning module to align with these holes 😉

Fan guards added to the Canopy. Bring on the module that actually has fans 😉

Finally, a cozy look from the inside of the Canopy that demonstrates the shape of the Canopy a bit more:

Inside view of the Canopy for as far is has been constructed. It clearly shows the cable feedthrough and the fan holes with their fan guards.

What next?

So up next is figuring out a way how to fix the top in a more solid way and not just on the edges. Anyway, I’ll figure that one out soon enough. Then it is on to constructing the outer layer and the sliders for the modules and the lighting rail. Not to mention the lighting rail itself 😉

Stay tuned for more!

Adding the sewer-dump

If you want a paludarium where plants will grow under water, I have learned you need to separate the aquatic water system from the “land” water system. The sewer-dump does just that.

The idea is that any water you add in the aquatic part which is too much will overflow in the sewer-dump. Watering the backdrop for example comes from the sewer dump. Any water falling on the (yet to build) land portion will be dumped here as well.

This accomplishes two things: A steady water level and a clean aquatic water world!

1. A steady water level

Anyone who owns a paludarium will know: The water levels tend to fluctuate constantly. Evaporation, adding water, it just varies all the time. How to solve this? Well, just add water on a regular basis (aka automated!) and make sure the water can overflow from the aquatic part. Like some cool infinity pool that keeps the water level at a constant. Excess water overflows into the sewer-dump section.

2. Clean aquatic water

Over time inside a paludarium soil will form. You can choose to add soil, or to leave it out… But as the paludarium becomes bio active rotting leaves will form soil. Lots of nutrients, but that will simply overpower the aquatic section. Too much nitrates and phosphorites will kill fish, ruin plant life, and if you are unlucky cause algae to massively bloom instead.

How to make sure that won’t happen? Divide the water systems! The sewer-dump recycles any water for the land portion. Any water from the land portion should flow back into the sewer-dump as well. Aquatic water that is “too much” flows in there as well.

What the sewer-dump looks like

So the sewer-dump in itself is pretty basic: It is a sealed-off portion of the paludarium by glass. I used to have a glass “wall” 10cm from the rear for this, but in order to maximize the aquatic water section this time I decided to make a really small sewer-dump:

The sewer-dump: Nothing more than three glass plates separating the aquatic water system from the “landmass” water system

The trick is of course that these pieces of glass have the exact height of the water level you require inside; in my case it is 5mm below the front window vent.

You can see that there are three PVC feedthroughs; one outside the sewer-dump and two inside. The one outside is used to draw water from the aquatic part into the external canister filter (notice the 3D printed filter cap on the right) so it obviously needs to sit in the aquatic part.

The one in the middle is used for the canister filter return line. Water gets pumped in through the long PVC tube you see in the picture above, which ends up in a (yet to construct) waterfall. That waterfall flows back into the aquatic section closing the loop. It is vitally important that no water gets “spilled” into the sewer-dump in this process!

So where does the excess water in the sewer-dump go?

The PVC tube on the left in the picture above is the real trick: Any excess water will flow through this pipe (in the picture the 20mm PVC tube is too short; I need to order more 😉 ). It will flow out of the paludarium eventually into the sewer (the real world sewer that is). I will capture excess water into a small canister under the paludarium which I will empty using a pump with 6mm hose that goes to the sewer (unfortunately the location of the paludarium is in the mancave, meaning it sits lower than the sewer system in the house).

Handiest of DIY paludarium tools: A 3D printer

Nope, I am not a fanboy of any 3D printer brand, it is a tool for me that (hopefully) gets the job done. Some of the most complex shapes and some of the simplest shapes I have printed for the paludarium… It is always nice to have the ability to print things you just cannot buy (and usually are hard to build otherwise).

The simplest: Alignment washer for the PVC feedthroughs

The awesome guys at Blue Lagoon built me a nice glass setup with holes in the glas predrilled. But these holes are drilled at 29mm, while the actual PVC feedthrough only measures 20mm. Not really a problem, there are wide flanges with rubber rings to cope with the sizing difference. Still, I was able to “misplace” the PVC elements in such a way that it would simply leak. “If only I had a washer at the exact right size”… Uhm… I DO have a 3D printer.

Read more »

Ordering the Paludarium itself

After some measuring, more measuring and several conversations with the people at Blue Lagoon (A Dutch Terrarium shop who build custom terrariums / paludariums), I pulled the trigged and put in an order.

The Big Bad Glass Box has arrived!

It took them less than a week to build. I got a phone call that my paludarium was ready for delivery… So I needed to speed up the construction of the cabinet! I had triple verified where the holes should be drilled in the glass… But that stays exiting every time. Will they have drilled in the right place, did they measure from the outer wall or the inner wall? Did *I* measure correctly when I drilled three big ugly holes in the cabinet for the PVC piping to go through? Luckily, in the end it all worked out great.

Read more »

Electronics and Waterworks in a Cabinet How?

The next issue I faced was one I remember well and I think you may actually “feel” as well: You add stuff and add more stuff in the cabinet underneath, and at some point it is just wires, tubes, pipes and devices… Completely inaccessible.

I wanted to do these things differently this time. What did I come up with? Why not install a slider drawer inside the cabinet! I can put all electronics on one side, water-related stuff (osmosis, heating, valves and pumps) on the other side. If I add enough length to all wires and hoses, I’d be able to pull out the entire
“module” to do work on without banging my head ever again! As a picture says a thousand words:

Read more »

The Birth of a new paludarium – Palu2022

Yes, I am back! Ok, ok, I can hear you saying: “Yeah we are in 2023 and you are building a 2022 model”. Well true, but these things take a LONG way before they actually see the light of day, especially given the nature of the paludarium projects I run. Every single time there is more pipes, more sensors, more lights, more stuff. This project was started over a year ago…

WHY? Well, welcome to the age of Zoom and Teams calls

I have come to the realization that things will not change much after the pandemic. Especially in the world of IT where I am still part of (#IworkForDell !) the movement back to how things were are slow or maybe even absent. Where I used to be on a plane like once a week I am now cooked up in my house doing online calls. Also, it doesn’t help that I do have a room where I can work, but it is in the “Mancave” and it has NO outside windows. Yuck!!

So what to do?

Read more »