Friday, 16 September 2016

Hydra - Your Plant Care Companion Part 2

So the last post left us with the question of how to deliver different amounts of water reliably to multiple pots of plants, given that gravity plays a major role in how water flows through a network of tubes. There are a few indoor watering systems out there that try to solve the gravity problem in different ways:

1. Capillary Action Systems:
  • Very simple.
  • Low cost.
  • No electricity required.
  • Usually requires one water source per pot (messy, not good for many plants).
  • Soil is kept moist around the clock (plants prefer a wet-dry-wet-dry cycle).
  • Size of water source is limited (most of these systems come with the water container attached).
2. Gravity Drip Systems:
  • Water source needs to be installed on top of the plants (not discreet, refill & maintenance troublesome).
  • Size of water source somewhat limited (how much water are you going to put up there?)
  • Systems on the market ran on batteries and gave the same amount of water to each plant.
3. Pressure Drip Systems:
  • Requires powerful pump or connection to the water mains.
  • Some systems had trouble with the tubing popping out.
  • Watering volume is controlled by varying the aperture size of the outlet, which is not very precise.
So how do we create an indoor watering system that can deliver varying amounts of water accurately to multiple potted plants? The only solution was to either have a separate pump for each plant or a solenoid or pinch valve array to control the water flow from one pump.

After a lot of digging, I found a manufacturer of (relatively) cheap, but strong pumps. I won't reveal exactly what pumps they are, but despite their small size and price, these pumps could:
  • Prime themselves,
  • Draw water up to over 2.25m in height,
  • Ingest small particles,
  • Survive 2 accelerated life tests (5 & 10 years) with no problems.
Best of all, the pumps used common aquarium air tubing that is small, discreet, cheap and readily available. In comparison, a solenoid array system would have cost slightly more and is prone to failure if the sole pump breaks down. These beauties were definitely the way to go!

Now, the pumps need to be controlled, and for that, I decided to employ an Arduino based system. Arduino is an open source electronic prototyping platform based on a microcontroller and a simplified programming language. Although the platform is simple enough to be understood by children, it is very powerful when taken to its full potential. The electronics it is based on is also solid technology.

An Arduino Uno

To give a brief overview of how an Arduino works, the microcontroller has multiple 'pins'. These pins can either act as 'input' or 'output' pins. An input pin can 'measure' the voltage it is 'receiving' (which is then interpreted by the program to mean something, e.g. a button press) whiles an output pin can be instructed by the program to deliver a specific voltage (which may then trigger something else, e.g. an LED turning on). By using these two types of pins as 'building blocks', one can create complex systems to do a wide variety of tasks.

Since the pumps required much more energy than the Arduino pins could supply, the prototype utilized mechanical relays to control the pumps. The switches basically have small electromagnets (controlled by the Arduino) that close the main circuit when activated, thus turning the pumps on. However, these made quite a bit of noise, so the decision was made to switch to a silent alternative for the final product (which I won't reveal).

Mechanical Relay Array (Prototype)

A custom printed circuit board (PCB) was designed for the final product. It was based off the Arduino microcontroller and incorporated the microcontroller chip, buttons, a 16x2 LCD screen, electronic switches for the pumps, protective circuits and the ICSP all on one board.

Electronic Circuit Prototyping - The Messy Truth!
A standard Arduino Uno board is not that hard to understand if you break it down into sections:
  1. Power Input - Contains protection components such as decoupling capacitors, a diode for polarity protection and a resettable fuse for overcurrent protection (the Hydra PCB contains all of these components) as well as a MOSFET to regulate the voltage coming in.
  2. Microcontroller to USB - Enables the microcontroller to communicate via USB. This involves a separate controller chip to mediate and a few resistors in between.
  3. Microcontroller to ICSP - Enables the microcontroller to be programmed via In Circuit Serial Programming (ICSP). This is basically a simple 6-pin connection that is an alternative to the USB. Think of the USB as a grand entrance to a building and the ICSP as a backdoor.
  4. Microcontroller Circuit - Contains the actual microcontroller and its crystal (a separate component which the microcontroller uses to measure time).
  5. LED - All Arduino boards come with a controllable on-board LED.
An Arduino Uno being programmed via ICSP

I won't go into details about the actual PCB design here, but for those of you who want to try your hand at it, there are two great programs out there, EAGLE and KiCad. I built the circuit in EAGLE at first, but had to switch over to KiCad as the PCB was too large for EAGLE - I had to pay for a premium version of EAGLE to work on a larger PCB; KiCad on the other hand is totally free.

Hydra's Custom PCB
That will be all for this post, look out for the next one on injection molding and supplier sourcing :)

Tuesday, 30 August 2016

Hydra - Your Plant Care Companion Part 1

This is my first Kickstarter project featuring an automated indoor plant care system :D Since there is a lot of content to blog about, this will be split into a few posts spanning the length of the Kickstarter campaign, which starts today! The topic of this first post will cover my motivations for starting this project and challenges encountered when building the alpha prototype.

For those of you interested viewing the project on Kickstarter, here is the link:

Also, if you guys find this interesting, I would be sincerely happy if y'all help spread the project via our Facebook page:

Now, I started this project as architecture school was getting really busy and my plants were getting neglected as a result. As I could not find a suitable system in the market to take care of my 20+ plants, I decided to come up with some watering automation of my own. The idea to put it on Kickstarter was on the back of my mind, but I needed to get the device to work first.

The aim of the game was to create a simple, compact watering system that could be installed anywhere. The water source would be a bucket on the ground and there would be a pump that sucks it up to distribute many potted plants, so I ordered a few peristaltic pumps to test the concept out.

For those of you who don't know, peristaltic pumps mimic the motion of peristalsis (the action that makes food move down our gullet and through our intestines) to move liquid through a flexible tube. Compared to the common centrifugal pumps, peristaltic pumps have no problem switching on and off frequently and are much more precise, which makes them more suitable for laboratory and medical applications.

However, I was soon to learn about gravity! It turns out that branching the outlet of the peristaltic pump into many different tubes to water many plants will not cut it as the all water will flow out of the tube outlet closest to the ground, no two ways about it!

One option would have been to make the outlets very small (basically pinhole) and use pressure in the tubes to even things out. However, we found that even with this, the bottom-most outlet was still favored and it would have been very hard to quantify the amount of water each plant received. In addition, one watering system in the market did exactly this and reviews did mention tubes popping out due to the pressure involved!

To find out how I solved this problem, stay tuned for the next blog post!


Fish Tank Stand

Hi, I just finished architecture school (with a masters) so I FINALLY have time to update this blog. I did a few projects since my last post so do look out for DIY fish tank lights, a cello with a broken neck and my next series of posts (right after this one) that will feature my first Kickstarter project!

Now, onto this project, it's a 2' x 1.5' fish tank stand (height: 3') made out of Kapur dimensional lumber, plywood and blockboard. 

The Kapur lumber was recycled from 2 previous fish tank stands I made (they were one of my first woodworking projects, even before the plywood bookshelves). Those creations were wayy over-engineered so I took them apart and trimmed the salvaged lumber down to size for use in this project. The plywood and lumber were spare pieces found in my school's fabrication lab so this a 100% reused wood project!

For those of you who don't know blockboard, it is a much lighter alternative to plywood (and particle board or MDF for that matter), although it does not have the strips at the side which I like to exploit in plywood for visual effect. It is not as strong as plywood, but is excellent for bracing in projects like these where the stress exerted on the material is not too high.

Structurally, the tank is held up by 4 'pillars' of solid kapur wood. The top structure is made out of kapur wood 'beams', one along each side and one across the middle of the rectangle to support the span of plywood in between. The entire structure is braced on all sides (except the bottom) with either plywood or blockboard.

Being a 2' x 1.5' tank, I would expect this stand to hold up a 2' x 1.5' x 2' fish tank, which should weigh in at around 200 - 225 kg with the weight of the glass and substrate factored in.

The stand was designed with a double leaf door and 2 openings at the side for wires, air tubes and pipes to find their way in. The openings also serve to ventilate the interior of the cabinet if a sump (a secondary tank that connects the primary aquarium, a sump serves to hold aquarium equipment and misbehaving fish) were to be placed there.

The stand has held a 1' x 1' x 1' nano reef tank for the past year or so and serves as a hub for the power chords all the rest of my aquariums use. Now I am finished with architecture school, I may consider upgrading the reef tank in the near future :)

Friday, 21 March 2014



I recently participated in the IKEA Young Designer's Award 2014 ( and got into the finals! Here is a glimpse into the concept and prototype of IKEA TEGEL:

IKEA TEGEL is a modular furniture concept. Users stack rectangular TEGEL Bocks, which come in various sizes, on top of each other to form anything from a bookshelf to a TV stand to a shoe rack in any size they want. Furniture designed with TEGEL can also curve around corners and funny edges, depending on how you orient the blocks. The smaller TEGEL blocks are also sturdy enough to be abused as stools and foot rests!

Each block is made out of 18mm melamine coated plywood and finished off with white PVC edge banding. The blocks all have a standardized depth of 25 cm (enough for most books), but vary in height and length.

Each TEGEL bock is simple to put together. A simple butt joint connects the top, bottom and sides of the block (2 screws per joint) and is reinforced with a flat L bracket that sits flush against the surface. The L bracket should ideally be made out of steel but I settled for laser cut 5mm acrylic for the prototype instead as flat L brackets are hard to find in the right size and expensive.

Altogether, I made 21 TEGEL blocks of varying sizes, cutting 366 grooves for the L brackets, drilling 1176 holes and screwing 504 screws. The entire manufacturing process from cutting the plywood (supplied in long planks) to edge banding and cleaning took 7 days and cost about $250.

To bring the idea of TEGEL further, I also conceptualize 'special' blocks that would include features such as lighting (for plants, a small fish tank or for display pieces) and a glass cabinet door so that people can really play around with the concept of modular furniture! What's even better about TEGEL is that it can be rearranged, making it highly reusable and environmentally friendly.

The TEGEL blocks are on exhibition at IKEA Tampines (2nd Level) from the 21st of March to the 3rd of April 2014. Please VOTE for the best design!

To Vote:

Video Link:

Courtesy of IKEA Singapore

Courtesy of IKEA Singapore

IKEA TEGEL. Design your own furniture.

Saturday, 15 February 2014

Cardboard Box-On-Wheels

Refuse, Reduce, Reuse, Recycle – four things that most architecture students subscribe to, but find very hard to put to practice in the architecture studio.

This box-on-wheels was a personal experiment to see how far one could push the term 'reuse' in the context of the studio. Most of our models are built using cardboard parts that are precision cut by a laser. However, weird geometries combined with time pressure limit how much cardboard we can save by arranging the parts efficiently.

Inspired by the use of cardboard as structural element in product design (furniture, bicycles), I came up with the idea of laminating/gluing multiple sheets of scrap, lasercut cardboard sheets together to form a relatively strong board which could be reused. In addition, I also believed that the interesting shapes and uniqueness of each board would lend an aesthetic touch to the final product.

The box would comprise of an aluminum frame with wheels and the sides and bottom of the box would be made out of up to 8 layers of laminated scrap cardboard. Scrap cardboard was plentiful in the studio and I chose pieces that still had a relatively large amount of cardboard on them. The boards with more interesting designs were earmarked to form the front and back of the laminated board. Gluing the boards was a messy business that drew the curiosity of my studio mates!

The final product was rather heavy. The laminated cardboard turned out to be relatively solid as a material, and I believe that the back and sides will hold up to anything that is not too heavy (3kg on each side, 40kg spread out on the bottom, which has bracing). As of now, the box would be used as a protective container to store my architecture posters!

Thursday, 13 February 2014

Cello Repair 11/2013

This cello belongs to a friend who played it in his school's Chinese Orchestra when he was younger. It had survived what I believe was a relatively major fall that saw the fingerboard take the brunt of the impact.

Fortunately, most of the force of the fall was used to break the glue holding the fingerboard to the neck, resulting in the fingerboard breaking off and taking a small chunk of the neck it it. For those who don't know, violins are held together by hide/animal glue, which is designed to give way in some situations before the wood cracks.

Other than having a small chunk of its neck torn off, the instrument suffered no other forms of structural damage Gluing the fingerboard back on is (and was) a messy and frustrating business as it tends to slip and slide, smearing glue all over the place (in any case, dried hide glue is easily washed off with hot water).

One other complaint my friend gave me was that he did not like the sound of his instrument (before the fall)! Besides new strings (which the instrument was in dire need of), I noticed that his bridge was also thicker than normal. Violin shops sometimes leave the bridge (and the entire violin for that matter) a bit thicker than recommended on student instruments to negate the lower quality of the wood used as well as to allow the instrument to better survive repeated abuse.

However, a thick bridge muffles the sound of an instrument as it cannot efficiently transfer string vibrations to the body of the instrument. On the flip side of the coin, an overly thin bridge also does detriment to the tone of the instrument and will also warp with time (due to pressure from the strings).

Thinning the bridge (I used a belt sander) tested my patience as bridge thickness is measured to an accuracy of 0.1mm (using digital calipers) and it is very easy to over-thin the bridge. Overall, I had to thin the thickest parts of the bridge by around 1.2mm and the thinner parts by as little as 0.3mm. The results were quite satisfactory as constant checking ensured that I did not shave off too much (a lesson learnt from carving a violin bridge for the first time).

Putting everything back together, the cello spoke again after many years in storage. A final verdict on the tone (normal student grade tone, rounded, slightly bright, a little thin) however, will have to wait for new strings!

Violin Repair 11/2013

This is a Suzuki (Japanese) violin that belonged to my friend's grandfather. It came in the condition that any old instrument will have come in – in need of repair!

Overall, the instrument had suffered no real structural damage. However, its seams were very loose in many places as the glue had dried out and cracked over the years. In addition, the previous owner had attempted his own repairs by gluing some of the then-loose seams shut with gum! The violin had lost its bridge and had to have its fittings and strings changed/reworked.

All seams that were not held together with gum were broken and glued shut. One point to note is that gum used was effective to the point where it was impossible to remove from the instrument! Not the proper way to repair an instrument (you can't open the instrument after that without damaging it) but functional nonetheless.

A new bridge was carved; this is the first time I have ever done this or any other fine wood carving so I selected a $16 Aubert Mirecourt bridge (second lowest grade, from France) The bridge comes with its rough shape and detail already in. A violin maker will then thin the bridge to specifications, fit the bridge feet to the curved body of the instrument and further carve out the details on the bridge. This is all very skilled work as the maple bridge will crack/break if too much force is applied at the wrong points.

All in all, luck was with me most of the way. The bridge I made was overly thin in some places due to my overzealous and impatient sanding (which will lead to it warping with time and affect the tone of the instrument). However, I was able to do everything else without damaging the bridge or cutting myself (I used a very sharp carving knife). The only other complaint about the bridge was that the angle between the A and E string was too shallow, making it easy to accidentally hit one when playing the other.

Other than that, the tuning pegs were trimmed as the ends were sticking out. One peg turned out particularly loose. A new soundpost was carved and fitted in. A new tailpiece, chinrest and strings were then installed and the instrument was given a wipe down.

The tonal quality of the restored instrument was bright. It carried a modicum of projecting power but sounded somewhat thin, possibly due to the thin bridge. I believe that adjustments to the soundpost (which was a hair too long on hindsight) and bridge position would produce a perceivable improvement to the sound of the instrument in both tone and projecting power.