Zds Project Log

One nice side-effect in my LED lighting system is that it makes my plants grow sideways, not upwards. Naturally this is not what everyone wants, but for me it suits very well.

On September I finally had time to cut my yucca. I literally cut it, to four pieces, and sealed the upper ends with plant wax. Now that it has had some time to adjust, the bottommost part has started to make new branches, and not just one, but four!

This is so nice change to the one long upward-reaching trunk with leaves just on the top.

The same effect is seen on my chilis:

The Interweb and people who understand plants told me this has to do with plants sensing if they are neighbouring another plant or a rock, ie. do they have to outgrow their neighbour or not. More specifically the red:infrared balance is what matters; the bigger that ratio is towards the red light, the more the plants will branch and grow sideways.

This is also very nice, as I’d love to have them bush-like, not tree-like. Much rejoicing ensued!

As the UV LEDs were 7-10 times as expensive as the red and blue ones, I switched them off and ran the light with just blue and red LEDs for five days:

Again fluorescent tube is on the left, LEDs on the right. One could argue that the plants turned a bit less this time, but as they grow all the time it is hard to say for sure. Also there naturally was less light output, too, this time as the UV LEDs were lacking and blue and red LED count stayed the same.

Nonetheless, it seems blue and red might be sufficient alone, which cuts more than half off of the LED cost.

This time I dug into territory of plant lighting. Early summer I got some young chilis from IRC acquaintance of mine and, well, three of five were still alive when autumn fell and it was time to check how I should have grown them in the first place.

The most obvious thing lacking was light. During the summer they did ok’ish just with sunlight, but during the rest of the year there’s not much sunlight to speak of here in Finland. Time to go artificial.

I already knew most of the plant lights just produce unnecessary heat and make plants look good, but do not help the plants grow. Quick dig into the Web confirmed this.

From http://www.bonsaifi.net forums I got what I was looking for: some hard data on what wavelengths chlorophyll actually uses (from Botany Online):

To put the long story short, I bought a E27 form factor compact fluorescent tube optimized for plants (info via bonsaifi.net again) and ordered some 400 leds of proper wavelengths from China and it was time for soldering and comparison.

The reference light is branded as “Megaman” and has a spectrum that looks like this:

The leds came from Chinese company branded as HB. They are the only Chinese manufacturer I am yet to find who sell small quantities directly to customers and who have English web page with prices. The delivery was very fast and service was good, earning my warm recommendation.

As UV and deep red frequencies are hard to come by and/or expensive, I took leds in 1:2:2 proportion on wavelengths of 402nm:455nm:639nm. The UV ones (402nm) cost four times the deep blue/red ones, so I took a chance and hoped this combination would work.

The combined electricity usage of the led light is around 20W and this is how it looks in action (never mind the casing, or lack thereof, its just in prototype stage):

The light is actually nowhere close to white, it’s red-blue-violet to a human eye. Whatever white you see there is just camera sensor having difficulties. The plant leaves look dark blue in that light as expected - there is no green light to reflect away, just wavelengths the leaves absorb and use.

And here is a combined image showing the LED light passing the real test with flying colors. On the left is the fluorescent tube light, on the right the led one; time difference between the two shots is five days: