Modern Light Diffusion and Model Starships by Jan Madsen part 2

Posted on by

back to Part 1

Experiments.

I had cut a strip of all the layers (with the exception of the plain white reflective layer) to use as a demonstration piece for this article. I replaced the white plastic reflective layer with a piece of aluminum foil glued to the back of the dispersion layer with Aleene’s Original Tacky Glue. Why this glue? Because it was in the closet where the Elmer’s glue should have been. The bottle read “all-purpose” and “dries clear”. I expected to get pictures similar to the ones above. What I got was this:

Gradated Dispersion Layer with Aleen’s glue and no diffusers.

Gradated Dispersion Layer with Aleen’s glue and all diffusers.

I was intrigued with the unintended optical properties of the glue, although a little disappointed with the bubbles.

At this point I thought that I had discovered something useful. However, after a couple of weeks, the glue went through a rather quick drying process resulting in the following:

Dispersion strip with Aleen’s glue dried

Dispersion strip with Aleen’s glue dried & lighted

Dispersion strip with Aleen’s glue dried & lighted with diffusers

While this result is of little use in nacelle lighting, it might have other applications.

After taking the above photos, I peeled the foil off the back of the Dispersion strip and removed the glue using water and alcohol.

I then decided to use the strip for some LED mounting experiments. I drilled a hole in the strip in the middle of the bottom edge (resulting in breakthrough due to the nearly equal dimensions). I filed a slot in the strip about ¼” from one edge (lower edge in the photo). I then placed an led in each of these positions and also an LED about ¼” in from the other edge to compare the effects. Results are shown below.

We can see that the light from the filed slot is dimmer than the light from the unmodified edge, and that the light from the drilled hole is dimmer still.

I then polished the light-catching edge of the filed slot using 400 grit paper, 600 grit paper, Novus #3 Polish, and Novus #2 polish in that order. This produced the following result:

We can see that the light from the slot is almost equal to the light from the unmodified edge.

I then “filled” the space between the middle LED and the drilled hole with Future floor polish. I use this as clear-coat and decal blender when required. The intent was to minimize the roughness of the drilled hole and approximate a polished edge. The result is below.

We can see that the light from the drilled-and-filled hole and the filed-and-polished slot are now nearly equal.

The next photo is the above unit with the diffusers applied.

Observations and Recommendations

  • All of the backlight layers with the exception of the rear reflective layer are directional, both front-to-back and top-to-bottom. Attaching the LEDs to any edge other than the intended edge produces unsatisfactory results. Using the wrong side of the Dispersion layer as the emitting surface is also disappointing. The diffusers don’t seem to be overly order sensitive, but you are probably better off keeping them in their factory arrangement. I recommend that as you take a backlight apart, you place a number on each layer in a consistent corner with a Sharpie. Light transmission through the dispersion layer is from bottom to top. As seen in the strip experiment, the reflective layer can be replaced with aluminum foil. Without the diffuser layers, any wrinkles will be very visible, which may be your intended result.
  • The Gradated Dispersion Layer is not of a uniform thickness. It tapers from bottom to top. I recommend that lighting units be cut to width from the edge of the backlight, and then cut to length removing material from the thin edge of the strip so that the light mounting area includes the widest part of the plastic. The optical clarity of the bottom, light insertion edge is important. If it is modified for LED mounting or other considerations, it should be polished to the greatest degree practical at the point the light enters the plastic.
  • Orientation of the LEDs is important. If they are misaligned (horizontally or vertically), the light bands will still be in the same direction but will be of less intensity
  • Clarity of the LED emitting surface is important. I tried filing the end of a couple of the LEDs on the string flat to see if I could achieve better light coupling. I’m not sure if the important factor was the removing of the lens effect or the removing of the polished surface, but the result was to reduce the light coupling.
  • The thickness of the light-insertion point may be important. The first thing I did was attempt to make a small, two sided light strip (~3/16ths x 2 ½”) to go in the nacelles of a 1/1000 NX-01, using small strips cut from the top of the backlight. This experiment was a total failure. I suspect that in attempting to make the unit very thin, I did not leave enough light-capture area at the end of the strip. At that point I didn’t know about the edge polishing, which didn’t help.
  • Some of the diffusion layers can produce interesting optical effects in and of themselves.

Experimenter’s Notes:

  • The glue experiment strip is 1” x 6 ½” and is what was left after cutting the pieces for the (failed) experimental mini-strips.
  • I purchased a lot of ten 9” x 12” surplus backlights from eBay for $50.00 including shipping. I bought that quantity to minimize the shipping cost per panel and to have enough to cover experimental errors. It should be noted that each backlight came with a 12” really thin and fragile CCFL tube mounted in a metal reflector (see disassembly picture). While the whole point of this article was to get AWAY from high voltage lights, inverters that run on 12 VDC to power the tube run about $15.00.

Addendum: It appears I just lucked out when I looked for surplus backlights. I haven’t seen any on eBay or anywhere else lately. However, there are a lot of cheap broken laptops out there. If you don’t have a local electronics surplus or recycling store, you can find them on eBay under Electronics/Computers/Laptops & Netbooks/parts for not working. Today, they start at $0.99 in auction and $12.99 (including shipping) in buy it now. The parts I bought were for 13” display.

Pictures were taken with a Canon SX40HS digital camera. The camera was used in full auto mode, shortest zoom setting, and no flash. A fluorescent ceiling light in the area was on to provide a little background illumination.

Application Note:

I am presenting this information before actually placing a lighting unit into a model. I am doing this ‘cause I’m really slow, and if I waited you might not see this article for several years. I think some of the more ambitious among you will put this information into practice long before I do. However, the end result might look something like this:

As a final note, I looked into the availability of Aleen’s Original Tacky Glue. Since it was something my wife had bought and stuck in the closet, I was afraid that it was obsolete and unobtainable. To my surprise I discovered that not only is it still available, you can also buy Aleene’s Clear Gel Tacky Glue, Aleene’s Easy Flow Tacky Glue, and a multitude of other variants (I found it at Amazon). The optical properties of Aleene’s Tacky Glue are not specified by the manufacturer and are therefor not guaranteed.

Results obtained by the experimenter and documented in this paper constitute a sample of one and are not statistically significant. Use at your own risk. Your mileage may vary.

Happy modeling.

Jan Madsen