LEDBrick Project - DIY pendant w/ pucks
- Thread starter theatrus
- Start date
sfsuphysics
Active member
not to shabby at all!
theatrus
100-mile-commuter
Yup.
The stencil is a kapton one from OSHStencils.com. Build a frame for the board (old PCBs, these little acrylic frames, etc), tape it down, align, and then use a putty knife to "print" the stencil by simply applying a squeegee like action. In this case I used a (very expired, like 5 years out of date :hmm5
jar of Kester SAC305+EM907 paste. Lift the stencil up carefully, and then just use tweezers to place parts on top of the paste.I may have one cold thermal pad on my board I may have to touch up after running some thermal tests. More on that later.
theatrus
100-mile-commuter
Running some thermal and output tests using three channels running at design-maximum output (750mA). I'm also monitoring PAR output using a Apogee PAR sensor (multiple: x5). There is no thermal paste or other gap-filler material between the PCB and the heatsink, as a baseline measurement.
PAR sensor, at 6 inches, centered. Don't consider this absolute, its mainly to check for thermal-related output reduction.
Output levels, holding steady for 30 minutes so far:
Thermals.
No fan, two channels running. The top-right royal blue has the highest measurable temperature from the (hacked to 320x240) FLIR-E4. This is a close approximate of die temperature - IR is going to be limited by the silicone lens of course.
Note that shiny aluminum heatsinks are not directly measurable using IR - in later images I've placed pieces of black electrical tape over sections of the heatsink to provide a better radiator. Ambient is approximately 18C in all of these images.
Three channels operating - sorry about the funny angle, I had already setup the PAR test at this point. Low-CFM Panaflo is running at 9V (basically inaudible)
Side of the heatsink with electrical tape to get a temperature read.
PAR sensor, at 6 inches, centered. Don't consider this absolute, its mainly to check for thermal-related output reduction.
Output levels, holding steady for 30 minutes so far:
Thermals.
No fan, two channels running. The top-right royal blue has the highest measurable temperature from the (hacked to 320x240) FLIR-E4. This is a close approximate of die temperature - IR is going to be limited by the silicone lens of course.
Note that shiny aluminum heatsinks are not directly measurable using IR - in later images I've placed pieces of black electrical tape over sections of the heatsink to provide a better radiator. Ambient is approximately 18C in all of these images.
Three channels operating - sorry about the funny angle, I had already setup the PAR test at this point. Low-CFM Panaflo is running at 9V (basically inaudible)
Side of the heatsink with electrical tape to get a temperature read.
theatrus
100-mile-commuter
U1 is an MCP9808 I2C digital temp sensor, C1 is the bypass cap for it.
Been building the lab up over many years, with some very very lucky finds (such as the "shows ERR, must be broken" 3458A which just needed new backup batteries
)
oreo57
Well-known member
Love when that happens..Been building the lab up over many years, with some very very lucky finds (such as the "shows ERR, must be broken" 3458A which just needed new backup batteries
BTW: A toy you need..
I would think it could, w/ the proper programming do "par"
Of course it also needs a remote sensor..
Isn't that sort of like putting a drink in front of an alcoholic?BTW: A toy you need..
Agreed tho - some very nice toys you have there!
Tim
Isn't that sort of like putting a drink in front of an alcoholic?
Agreed tho - some very nice toys you have there!
Tim
I think alcoholics actually have more self restraint than electronics junkies :lolspin: now I need one of those.......
sfsuphysics
Active member
hmmm wonder if someone has a cheaper add on version of that to a cell phone. I mean I know you can add a spectrometer reader to a cell phone, but I'm curious if it can accurate measure the actual energy output.
oreo57
Well-known member
Think I found an O2Surplus project ...
http://publiclab.org/notes/mathew/09-12-2013/open-source-linear-cmos-spectrometer
http://www.tricorderproject.org/blog/sneak-peek-3d-printable-mini-spectrometer/
http://publiclab.org/notes/mathew/09-12-2013/open-source-linear-cmos-spectrometer
http://www.tricorderproject.org/blog/sneak-peek-3d-printable-mini-spectrometer/
theatrus
100-mile-commuter
I've been trying to get my hands on some of these. Sadly, they appear to not be in the business of selling:
http://www.hamamatsu.com/resources/pdf/ssd/c12666ma_kacc1216e.pdf
http://www.hamamatsu.com/resources/pdf/ssd/c12666ma_kacc1216e.pdf
Aqualund
New member
Looks great! I have a thread kicking around here somewhere with the same layout, but I was lazy and used the Lumias. I came to give some advice now that its been about a year since they have been running.
Corrosion.
Anything that was exposed and about 24 inches off regular flowing saltwater, nothing crazy, has produced enough corrosion to make some of the channels inoperable. Especially where the molex connecters go into the boards, and the wires where they go into the connectors.
Of course, anything under 24" was much worse.
My work around has been to cover every exposed connector with liquid electrical tape.
Obviously, if I had built an enclosure with a water tight glass panel this would have been optimal....but I underestimated the effect of the corrosion so high off the water with the molex connectors.
And of course...any leds I didnt have protected behind a lens suffered as the solder immediately went bad.
Corrosion.
Anything that was exposed and about 24 inches off regular flowing saltwater, nothing crazy, has produced enough corrosion to make some of the channels inoperable. Especially where the molex connecters go into the boards, and the wires where they go into the connectors.
Of course, anything under 24" was much worse.
My work around has been to cover every exposed connector with liquid electrical tape.
Obviously, if I had built an enclosure with a water tight glass panel this would have been optimal....but I underestimated the effect of the corrosion so high off the water with the molex connectors.
And of course...any leds I didnt have protected behind a lens suffered as the solder immediately went bad.
