I wonder what the power / light output difference would be from hooking up a dimmer to dim them vs just cutting a few off completely with a switch or jumper?

 

LEDs don't work that way, we get a lot of queries about just such a problem. A conventional dimmer won't work with LEDs, they are very different than incandescent lamps. With LEDs a PWM scheme is pretty much a must.

To the OP:
I was starting to modify a conventional drawing, and it occurred to me, what power supply are you using? This dictates a lot of different parameters.

So, what is the power supply you are going to use? If it is 12V you will probably need around 7 amps. With 24VDC you will need around 4 amps, and will use less parts. The design will work around what you decide to go with.

 

What ever power supply you deem to be appropriate, and i'll buy it

I am in the UK so will be powering it from UK mains supply.

 

Find it first, then I'll design around it. I recommend a 24VDC power supply, the unit I showed early on was a 6½A model at an extremely reasonable price. $15 is approximately £10. It does seem you folks over the pond pay a bit more, and have more trouble getting basic parts than we do.

 

Higher amp models seem to be a lot more expensive, what is the minimum amp I should look for?

eBay has 5A models for £24 and 10A models for £31

 

I just said, 6 amps, more is better. You don't want to cut this too close, one of the universal principals in electronics is you need headroom, it works better and lasts longer. That £10 unit isn't too expensive either, try finding it's equivalent. You might even be able to buy from the source I quoted, last time I talked to Tammy (one of the two permanent folks at B.G. Micro) she mentioned they do a lot of overseas orders.

BTW, you will need some air flow into this room. The electronics won't be that hot, but if it doesn't have a place to go heat builds, much like an oven.

 

Would a standard ATX PC power suppy work as I have a few of these lying around (just a RANDOM thought!!). They have 12V rails and i've just checked the net and I believe these are 30A.

Aside from that as that would be messy anyway, I've just checked the B.G.Micro website and found the 6.5A power supply you mentioned... quite a price reduction

It quotes just under $20 shipping which isn't much more than I pay from within the UK these days anyway - very reasonable.

If that power supply would suffice then I can order it from there no problems.

I imagine some other parts may also come from this site?

 

OK, I'll start work on the schematic. It'll take a long while for the package to arrive, you can test the circuits with a couple of 12V batteries in series if you get to that point. I'll post the schematic on this post when I get it done.

 

Cheers pal I really can't thank you enough for helping =)

 

I am new in the field of electronics though I’m compelled to contribute first in form of a question: is there possibility of constructing LED growing light using transistors; where voltage required to light a given LED is varied through an appropriate set of Darlington’s pair with the LEDs in question as the load…?
Recall: VBE acts as a diode. I’m trying it out and will give my stand very soon…

 

Yes, I've modified some of my old illustrations I used for LED drivers. There is one twist to what you're describing, by adding a emitter resistor the transistors become a constant current source, which is ideal for driving LEDs. Without the emitter resistors shown there is no control. While VBE acts like a diode, it isn't how a transistor works. It is the current through the BE junction that controls the collector current, the emitter resistor adds an element of feedback for refined control.

 

Like bill said LED dimming does not rely on limiting the voltage through a VR.
You need to PWM the drive voltage.
Like u said an old ATX supply will have enough power for your need, why buy what you already have, right?
12 V is more than sufficient.
All you need to make three PWM drivers using 555 timers and a power transistor or a FET.
Use a limiting resistor in series with each LED and use all the LED's in parallel for each color.
This way each led will get enough drive voltage and the brilliance can be adjusted with the 555 timer.

But power dissipation will be there in the power drivers, but so will the LED's need to be mounted on heat sinks or they will burn out indefinitely.

Paralleling the LED's will keep the rest lit even one led burns out. A series combination will definitely black out if one fails.

Just a thought. If you like this method I can schema it. It's really simple

Regards

Rifaa

 

I thought of using PWM, then realized I was adding to the complexity for no good reason. The knobs on these LEDs need not be linear, as long as they have the full range. Saved me a few hours in drawing time, and I had already put in over 4 hours drawing this schematic.

 

I thought of using PWM, then realized I was adding to the complexity for no good reason. The knobs on these LEDs need not be linear, as long as they have the full range. Saved me a few hours in drawing time, and I had already put in over 4 hours drawing this schematic.

Good for you .

But don't you think PWM is a better way of controlling the brightness.
Still I like to see what you are drawing, I'll hold back for now.

Cheers

Rifaa

 

PWM is linear, so it is usually preferred. Simplicity is also good. PWM would have involved another 12V regulator chip, a LM339, and possibly a 555. Instead I replaced it with a resistor and 2 diodes. I was all set to go that route when I looked at what I had already drawn, and realized it looked a lot like this (which I had drawn over a year ago)...

I have several other concepts similar to that in my electronics #2 album.

I do a lot of concept drawings.

Just reread your post, all of these components will need heatsinked.

 

I like it.
Nice and simple approach.

Rifaa

 

Just curious if Blue and Red are the right LED's to use. Typically when your talking horticulture "Blue" means soft or cool white, and "Red" means warm white. I'm interested in this too so I'm going to do some research. Back in the day when people we trying to get by not using HID's they would go get the Cool and Warm lights, I've never seen anyone actually using blue and red lights.

 

I've done fairly extensive research on the subject as well as some minor experimentation. I want to re-do it with pictures next time since I was working on developing my own system for potential sale. Blues and reds are pretty much all a plant wants. To inject any other major color is to reduce the optomization of the system. White LEDs are merely blues with a second coating, if its a cool white bulb, then its not the worst thing to use, however, Im just not keen on it personally.

Also, reflected light is an extremely ineffecient method to rely on as opposed to better focus, imo. However, using melamine shelving panels for your enclosure (with a couple computer fans exhausting the top) would be ideal. Foil and other seemingly shiny things are bunk, for many reasons.

If you are growing tomoatoes (seemingly most popular) you don't have to grow them upright, get them moving laterally, threading through some 2x2" metal grating. The other great thing about LEDs is that you don't have to stick to the same old rectangle box convention.

Just some food for thought...





oh yeah, I found a link from before that might help illustrate

http://www.botany.uwc.ac.za/ecotree/photosynthesis/spectrum.htm#light

 

OK I've been doing a lot of research this morning on this and there seem to be some "over-unity" type stuff floating around. 1st off I found 90% of the LED lights being sold are bogus. The ones that work seem to be going for around $700 for 100w. 100w also seems to be the minimum to actually fruit a tomato or other fruit. Also the square foot being required seems to be at about 50% rather then 15%-25% as claimed. But even with a 50% in light power reduction would open up a world of commercial applications for going indoors. Right now its just not realistic to go indoor. Another interesting thing I found is the high dollar commercial systems seem to be putting off a blue white light, that could almost be mistaken for a MH. I'm going to keep researching it when I get free time and if I see anything else I'll post it.