Building a Home Made LED
- Thread starter tenthirty
- Start date
No such thing as stupid questions my friend...
even though there is very little heat given off, depending on the strength of the LED's and wave lengths involved as well......bleaching can still take place. If no bleaching is noticed, then by all means lower the unit right on top of the plant...
Also understand lens angles. The tighter the angle the farther the diodes need to be from the canopy. This introduces "Inverse Square Law" problems
1 watt white diodes + wide lens angles + close to canopy (6-12") = WINNING
WHITE POWER
yup...my bad... I forgot about the lenses.....
Wide angle lenses means less penetration. Narow angle -> deep penetration into the canopy but indeed require the light source to be placed higher.
I use 45 degrees lenses on 1W and 3W chips, but the leds are spaced in such a way the light spot from each of them barely overlap. In this way is not necesary to adjust the height.
I use 45 degrees lenses on 1W and 3W chips, but the leds are spaced in such a way the light spot from each of them barely overlap. In this way is not necesary to adjust the height.
tenthirty
Member
repuk, send a link to the specs on the heatsink and I'll do the calcs to see how close we are.
Lenses are a bad idea. The XM-Ls seem to like about 13" above the canopy.
IMHO mixing the light from the different diodes is a bigger problem than using lenses. The cone of light without lenses seems just about right for the XM-L's anyway.
XP-E2s would be a better bet if you could get them instead of the XP-Gs.
Get tapped and drilled HS. Remember HS are forever!
Gluing the diodes down is just a bad idea on so many different levels. (good, fast, cheap pick 2)
Lenses are a bad idea. The XM-Ls seem to like about 13" above the canopy.
IMHO mixing the light from the different diodes is a bigger problem than using lenses. The cone of light without lenses seems just about right for the XM-L's anyway.
XP-E2s would be a better bet if you could get them instead of the XP-Gs.
Get tapped and drilled HS. Remember HS are forever!
Gluing the diodes down is just a bad idea on so many different levels. (good, fast, cheap pick 2)
Whoa whoa whoa...just hold on a sec here...is this above statement coming from the same lazy-ass stoner that I know??!!?
.....ROFLI've always said you're not lazy!!LOL.....you've got a good point bro...good point indeed.
I've looked at pre-drilled and tapped heat sinks and the damn shipping charges are prohibitive and I have no idea how much duty or tariffs I would have to pay bring it across the border to the Great White North and all....
I would love to be able to find a place up here or find a machine shop that wouldn't extort me into paying the equivalent of my right testicle for drilling and tapping a few holes...
tenthirty
Member
I can see me drilling out a busted off SS self taping screw.
Probably would work pretty good though.
I can see me drilling out a busted off SS self taping screw.
Probably would work pretty good though.
Better the screw than the damn tap!
Here they are, 60mm:
100mm:
Not going to use lenses. I've checked angles and coverages and spectrum mix will be fine. Why XP-E2? I have to order everything from a single shop (RL)... I'm about to order everything on monday, I'm still concerned about bleaching with the 10W chips, are your ladies doing fine at 13"?
That would be expensive... I'd do the drilling and taping. Guess rivets could crack the PCBs? that would be faster...
They need a certain amount of pressure to 'pop' of which I'm sure you are aware of and yes, that pressure might be too much for the PCB to take...
rives is onto something with the self-tappers which is something to look into.......my issue is still the fact that I live in a building...not much room to set up a drill press if I wanted to drill and tap the heat sink myself. Now, thinking about it, the bigger problem would be sneaking the thing into the building...ahahah
Good Luck
rives is onto something with the self-tappers which is something to look into.......my issue is still the fact that I live in a building...not much room to set up a drill press if I wanted to drill and tap the heat sink myself. Now, thinking about it, the bigger problem would be sneaking the thing into the building...ahahah
Good Luck
not much fins on the 60mm, forget it
http://led-heatsink.com/LED_heat_sink_calculation_simulation_thermal_design.html
only my test-heatsink has threaded screws.
tape for all led on star-PCB
can be rework without a mess on the heatsink
thermal-glue for emitters without Star-PCB.
http://led-heatsink.com/LED_heat_sink_calculation_simulation_thermal_design.html
only my test-heatsink has threaded screws.
tape for all led on star-PCB
can be rework without a mess on the heatsink
thermal-glue for emitters without Star-PCB.
thanks for the reply....I was thinking along the same lines but needed confirmation..
<edit> I made a stupid comment I needed to remove...too early and just smoked one with my morning coffee....
Last edited:
tenthirty
Member
At 20" long you should be good for something > 57w.
In an air conditioned room with a little breeze, maybe 90+ watts of leds.
The XP-E2's put out 20% more light, which means that much less heat for a given watt.
Send them an email and request XP-E2's and see what they say. Maybe you'll get lucky.
If it were me, I would go 2 ww XP-Gs for each nw XM-L.
In an air conditioned room with a little breeze, maybe 90+ watts of leds.
The XP-E2's put out 20% more light, which means that much less heat for a given watt.
Send them an email and request XP-E2's and see what they say. Maybe you'll get lucky.
If it were me, I would go 2 ww XP-Gs for each nw XM-L.
tenthirty
Member
Back to the thread.
The test group (home made led) are doing noticeably better than the control group (Pro-grow).
Better color, no funkyness with the leaves, bigger buds (same development stage) just overall a notch or 2 better than the control group.
In a few days the test will be reset and a new batch of clones will go under the leds.
So the first test setup is for the first 30 days of flower.
There are about 12 individuals in the test group and a few more in the control group.
This should give a good sample over 3 or 4 cycles.
Next up will be tuning the light......coming soon!!
The test group (home made led) are doing noticeably better than the control group (Pro-grow).
Better color, no funkyness with the leaves, bigger buds (same development stage) just overall a notch or 2 better than the control group.
In a few days the test will be reset and a new batch of clones will go under the leds.
So the first test setup is for the first 30 days of flower.
There are about 12 individuals in the test group and a few more in the control group.
This should give a good sample over 3 or 4 cycles.
Next up will be tuning the light......coming soon!!
Thanks!!! that confirms my "rule of thumb" approximation.
20% vs XP-G or XM-L???
So you think too much WW in the last incarnation? (4 X-ML NW, 2 XM-L CW, 12 XP-G WW)
I could sub some WW XP-G's for NW XP-Gs' with the added benefit of adding 100% spike in the 450nm's (XM-L NW spike is after the 450nm and XM-L CW is before the 450nm mark)
tenthirty
Member
XP-G's The XP-E2's are 1a. diodes. Here is the spec sheet.
http://www.cree.com/led-components-and-modules/products/xlamp/discrete-directional/xlamp-xpe2
IMHO, what you want is to mix the diodes so that the blend creates a nice smooth wave.
I really believe that spikes are a bad thing.
Look at it this way, what do your eyes do when a flash goes off, when you are driving into the sun or snow blindness.
All photo receptors do this.
Weather you want to shift the wave blue or red, who knows????
tenthirty
Member
http://csep10.phys.utk.edu/astr162/lect/sun/spectrum.html
Check out the spectrum in this graph.
Note, that it is a nice smooth wave.
Also, because it is so far away, for lack of a better term, the light is homogenous. (well mixed and smoothed)
Think about this.
For those of you that have party leds like me (pro-grows), I remember the first time I turned them on. I could see all of the little cones of light and thought isn't that cool.
Well it may turn out that it is not.
The whole plant is covered with photo receptors and locality of the receptor may (it seems) be responsible for a number of given responses.
What make using the pro-grows and the like even worse,
as the light progresses down the canopy, the spectrum is going to change at any given location down the plant
because of the non-locality of the sources of different frequencies of light. (red led and blue led respectively)
Talk about unintended consequences.
Test:
When you look at the lights, what is your response, or your little point and shoot camera.
I can tell you this, my response to the pro-grows is not all rainbows and butterflies.
If this is correct, this means that it was a mistake for me to put the 66nm leds on my first light.
Check out the spectrum in this graph.
Note, that it is a nice smooth wave.
Also, because it is so far away, for lack of a better term, the light is homogenous. (well mixed and smoothed)
Think about this.
For those of you that have party leds like me (pro-grows), I remember the first time I turned them on. I could see all of the little cones of light and thought isn't that cool.
Well it may turn out that it is not.
The whole plant is covered with photo receptors and locality of the receptor may (it seems) be responsible for a number of given responses.
What make using the pro-grows and the like even worse,
as the light progresses down the canopy, the spectrum is going to change at any given location down the plant
because of the non-locality of the sources of different frequencies of light. (red led and blue led respectively)
Talk about unintended consequences.
Test:
When you look at the lights, what is your response, or your little point and shoot camera.
I can tell you this, my response to the pro-grows is not all rainbows and butterflies.
If this is correct, this means that it was a mistake for me to put the 66nm leds on my first light.
if the XP-E2's are 1a. diodes, what class is this?
http://www.cree.com/led-components-and-modules/products/xlamp/discrete-directional/xlamp-xpg2
