Corrosion is less an issue if you are forcing cool dry air into the hood/fixture. In aqualunds case it's in a shed with how many thousand gallons? and no cool, dry, totally salt free air available as would be the case with a display in the home.........still splash guards are a must IMHO.
LEDBrick Project - DIY pendant w/ pucks
- Thread starter theatrus
- Start date
theatrus
100-mile-commuter
I can see both a PAR meter and a spectrophotometer all in one.
Sadly its not characterized for absolute intensity, but I imagine it would be stable and could be calibrated in.
I don't have the equipment to do that properly though
theatrus
100-mile-commuter
I've been running T5s and bare (HASL, ENIG, etc) PCBs with exposed components in my open back canopy for quite some time. Splashing is an issue (especially when your giant derasa clam gets irritated), but I have not had ambient humidity be an issue for a bunch of exposed stuff. With the T5s there is forced cooling, and house humidity is generally in the 20-40% RH range.
Aqualund
New member
I too was going to use acrylic shield, But I know that especially for the high energy spectrum (uv to blue) you can run into problems with diffraction and "burning" the acrylic. basically the separation between the molecules of acrylic aren't the proper distance to prevent the light waves from running into them...heating the material, and darkening it.
I've never found that to be the case. Acrylic is designed and intended for exposure to the sun in windows and such and that is much lower (down to sub 300nm UV radiation and far more intense) Reef Uv's should never go below 405nm and should never cause an issue for acrylic. Heat build up "could" melt or burn it if intense enough (i.e.with in less than an inch off a 50w emitter with no air flow, I have done that but it is more due to the trapped air being super heated not the acrylic ASFAIK and not the UV directly impacting the material). I've never seen or heard of that happening in a diy build, even those using lumia chips or even more powerful emitter arrays.
I have 5 of these
http://reefll.com/index.php?route=product/product&path=59_60&product_id=51
on a 36" makers heatskink for the past 1 year shining through the standard makers acrylic shield. There are 0 visible changes in that acrylic.
http://reefll.com/index.php?route=product/product&path=59_60&product_id=51
on a 36" makers heatskink for the past 1 year shining through the standard makers acrylic shield. There are 0 visible changes in that acrylic.
theatrus
100-mile-commuter
Good to know pwreef!
I'm the midst of laying out the LM3414 based driver.
Would anyone be interested in the source design files for these boards? I'll be happy to put them up on GitHub or similar, however its not a freeware CAD package so I'm not sure how applicable it will be for everyone.
I'm the midst of laying out the LM3414 based driver.
Would anyone be interested in the source design files for these boards? I'll be happy to put them up on GitHub or similar, however its not a freeware CAD package so I'm not sure how applicable it will be for everyone.
I'm interested. What software are you using to lay out the board?
theatrus
100-mile-commuter
Using an older version of Altium. But I'll happy farm out a Protel compatible set of files, and gerbers, under a CERN license.
Time ran out on the driver board last week, will be touching it up this weekend.
theatrus
100-mile-commuter
I've started populating all of the stuff in this GitHub repo under the CERN Open Hardware license.
https://github.com/theatrus/ledbrick
Gerbers, PDFs, drills and a netlist file for the emitter are included. I can provide several CAD export formats (no Eagle, sorry) - let me know what you want and I may be able to provide
https://github.com/theatrus/ledbrick
Gerbers, PDFs, drills and a netlist file for the emitter are included. I can provide several CAD export formats (no Eagle, sorry) - let me know what you want and I may be able to provide
Electrobes
Montipora type guy
I may not understand all of what it going on, but this is a fantastic thread for the visuals alone. Thanks for sharing, I love DIY and this is a fun read!
theatrus
100-mile-commuter
The metal core PCBs were here after about 7 calendar days - pricing was by sq-inches (increments of 50/100/150 sqinch) - the more you order the cheaper it gets per unit of course.
The driver boards are from OSHpark, which takes a bit longer (about 1.5-2 weeks) but is a great way to get 3 boards at $2/sqinch.
The driver boards are from OSHpark, which takes a bit longer (about 1.5-2 weeks) but is a great way to get 3 boards at $2/sqinch.
theatrus
100-mile-commuter
I finally found time to build a single copy of the driver board and start doing validation tests.
I used a similar paste + reflow construction technique due to the higher part count (I usually manually assemble one-offs, but this was faster in the end):
Starting to populate parts
Several minutes later, magic, a complete board!
The average current output regulation is spot on (set to 500mA, within 1%). The switcher is running at 300kHz, which allows a lot of flexibility in the frequency of the PWM dimming signal. The driver appropriately cuts out when the PWM signal is brought to GND. Dimming works great without an output capacitor - I'm experimenting what the realistic ripple currents and ringing end up at. I haven't performed an efficiency calculation yet.
I used a similar paste + reflow construction technique due to the higher part count (I usually manually assemble one-offs, but this was faster in the end):
Starting to populate parts
Several minutes later, magic, a complete board!
The average current output regulation is spot on (set to 500mA, within 1%). The switcher is running at 300kHz, which allows a lot of flexibility in the frequency of the PWM dimming signal. The driver appropriately cuts out when the PWM signal is brought to GND. Dimming works great without an output capacitor - I'm experimenting what the realistic ripple currents and ringing end up at. I haven't performed an efficiency calculation yet.
