Article: Lighting effects for a spaceship filming model

star-art

Sr Member
My article appeared in Volume 30 of Sci-Fi & Fantasy Modeler. It's a shame this magazine went out of business. I'm lucky to have a copy of the issue so I can preserve the article here.

NOTE: This is Part 2 of a 3-part article series. Part 1 can be viewed here:

https://www.therpf.com/showthread.php?t=155323
 

Attachments

  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_01.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_01.png
    671.1 KB · Views: 450
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_02.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_02.png
    2.1 MB · Views: 383
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_03.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_03.png
    1.6 MB · Views: 341
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_04.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_04.png
    1.2 MB · Views: 352
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_05.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_05.png
    1.1 MB · Views: 307
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_06.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_06.png
    702.7 KB · Views: 303
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_09.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_09.png
    405.3 KB · Views: 271
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_08.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_08.png
    1,018.6 KB · Views: 312
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_07.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_07.png
    1.9 MB · Views: 324
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_10.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_10.png
    843.7 KB · Views: 291
  • KestrosIV_part2_pp74-84_proof-7-6-13_Page_11.png
    KestrosIV_part2_pp74-84_proof-7-6-13_Page_11.png
    1.7 MB · Views: 330
Lots of useful information - thanks! Though it's amazing how much the LED situation has totally changed in just a few years...
 
Well, lots of things. I'd say the main ones for modelmakers are affordable dot-addressable LEDs, cheap and easy to use microcontrollers, white LEDs based around phosphors rather than blended RGB light, low-current LEDs, and extremely small SMD LEDs.

Dot-addressable LED modules, such as Neopixels, allow for precise control over colours on a per-LED basis if you want. They can be controlled and sequenced in an animated form via a microcontroller. And there are even RGBW modules, which contain four different LEDs - red, green, blue, white. And a big plus is you can vastly reduce the amount of wiring in a model while still maintaining full control over colour combinations.

Microcontrollers allow for very complex sequencing for animations, and can be used in conjunction with Neopixel-type LEDs to implement all kinds of colour and brightness animation.

The white LEDs you talk about, and diagram, consist of three separate LED dies inside a single housing. The RGB output is seen by the human eye as basically "white". But these aren't so common now. Much more common are the ones with phosphors that produce white light directly. "Warm" and "cool" white LEDs are widely available.

Low-current LEDs are still annoyingly hard to find, but out there. Instead of 15-20 mA, which is fairly typical for an LED, they can sip a mere 1-2 mA of power. In a single stroke you can add 15-20x more LEDs for the same power budget!

SMD LEDs are available in all kinds of sizes, and can be used where traditionally one might of used fibre optics. 0402 LEDs are widely available, and are only 1mm x 0.5mm x 0.5mm in size. You can even get 0201 SMD LEDs as small as 0.65mm x 0.35mm x 0.2 mm these days! Admittedly those are a huge pain to wire up and solder.
 
I'm familiar with many of these, but you're right a lot has changed since the article was written in 2012. Neopixels are something totally new to me, thanks for the mention! Also, I wrote specifically about RGB LEDs in the context of color changing applications -- not using RGB LEDs to make white light. (I used natural-white LEDs for the spotlight effects.) As far as I know, RGB LEDs like the ones I used for the engines in the spaceship model are still available. Only now, you can also get them in LED strip form which I don't think was readily available back then. But, again, you are correct that RGBW is something new since I wrote this. I will need to look into those. . .

Microcontrollers are well outside the scope of the article (and the project as I haven't yet had the chance to play with them). The main emphasis in the article was my need to provide *linear* dimming for all lights. That was a particular challenge as it seems 90% of dimming hardware and applications are based on pulse-width modulation (PWM). As I understand things, microcontrollers are used primarily to create PWM signals. That is how they can create complex lighting effects. As I researched ways to control things in a linear fashion, I had a very hard time finding any good info.

Most modelers probably don't care and/or have no need for linear dimming. It's really only an issue if you want to film your models using a digital camera. PWM output creates an annoying "flicker" effect that can be difficult and annoying to deal with when filming. For the spaceship model featured in the article, that would have been unacceptable to the filmmakers -- except, ironically, they decided they *did* want a flicker effect in at least one scene (the separation sequence) and they actually added it artificially for that shot.

For me, I've also been commissioned by a client who happens to be a cinematographer and wants to film his own models. That means each project I build for that particular client must use linear dimming. So, at least for me, PWM is not an option for some of my projects and therefore I have to use an alternative approach.
 
Last edited:
That's true... for most modelmakers, low-frequency PWM is fine as you can get away with it owing to persistence of vision. You only really notice how bad it flickers if you move your head suddenly looking at the model. Or when making a YouTube build video. :)

Neopixels (Adafruit's rebranding) and other WS2812 devices are pretty cool, but have a pretty objectionably low refresh rate (a measly 400 Hz), which is a shame. Adafruit do also sell "Dotstar" LEDs, which operate at 19.2 KHz. I haven't tried them, but that PWM rate might be high enough for video and film. Maybe?

I'm using a BBC micro:bit for my current project, which is a children's educational toy - a chip on a board with a 5x5 array of SMD LEDs in red. It uses 2mA LEDs and low-freq PWM to power the lights, which is great for battery life, but not so great for avoiding flicker.

By the way - my comment wasn't meant as a criticism. Just that it's interesting how much has changed in a short period of time when it comes to lighting models. The long decades of grain of wheat bulbs, with the occasional halogen light and fibre optic bundle, are long gone!
 
Last edited:
This is all very interesting. I'm interested to know more about the compositing process when shooting the ship against a black screen. I understand why you shoot multiple passes (the matte pass with the white background so you can pull the luma key(?), but why wouldn't you just shoot it on bluescreen instead?

SB
 
No worries nkg. Even though I have a background in electrical engineering, I still learned a lot when lighting that model spaceship. (For example, I had never even considered things like "flicker" when filming.) I'm sure there are many thousands of people out there who don't understand electronics but simply want to put lights in their own projects. I hoped to share some of my experience.

Electronics can be surprisingly complex, and the learning curve can be very steep for some. Even if you master the basics, in order to ensure things operate safely and reliably a certain amount of engineering must go into every project you create. I wonder if this is the part that many model makers don't take into account. For example, it's not enough to simply figure out the correct value of a resistor and what voltage to run to make an LED light up. You also must know *exactly* how much current is flowing through EVERY single wire and then design the circuit so it operates safely.

SB, the filmmakers employed the same technique Doug Trumbull used in Star Trek: The Motion Picture (which, ironically, I've been re-watching lately as I work to recreate the Klingon Cruiser miniature from that film). The model was shot against a black background. Then, a matte pass was created by filming it against a brightly-lit white background. This was used to extract a matte.

Blue screen (now green screen) is a common way to film things like this. That method has advantages and disadvantages. Trumbull didn't seem to like it much, but ILM and Apogee used that technique extensively. The starship Enterprise from ST:TMP could not have been filmed using a blue screen process. The paint job on that model would have reflected the blue light, creating "holes" in the mattes. This is why ILM completely repainted the model when they did the effects for Star Trek II: The Wrath of Khan.

When Apogee filmed the opening sequence for ST:TMP, it was done against blue screen. But, the Klingon Cruiser model had to be painted in such a way that the finish was absolutely flat.

I believe the "C" starship model could easily have been filmed against a green screen. But, the filmmakers wanted to do things "old school" and so they chose to follow in the footsteps of Doug Trumbull. In fact, he actually mentored them on the project.

Part 3 of the article goes into the filming of the model. If there's enough interest, I can post it in the Studio-Scale section.
 

Attachments

  • 601131_448114205256573_393208048_n.jpg
    601131_448114205256573_393208048_n.jpg
    49.8 KB · Views: 169
This thread is more than 6 years old.

Your message may be considered spam for the following reasons:

  1. This thread hasn't been active in some time. A new post in this thread might not contribute constructively to this discussion after so long.
If you wish to reply despite these issues, check the box below before replying.
Be aware that malicious compliance may result in more severe penalties.
Back
Top