Another Blade Runner thread: The Portable Voight-Kampff Scanner

As I think about it, the simplest solution for that front scanning light is probably the best. I spent ages poring through tech catalogues for something obscure, but I actually think at this point it's something as simple as a Neopixel LED. The larger discrete units are 5mm x 5mm, emit RGBW and would therefore be capable of doing red-blue-white without issue, and more importantly they actually match the wiring arrangement. Neopixels have 4 solder points - 2 to a side.



As it happens the front scanning light also has 4 wire leads connecting to it - 2 to a side.



The wiring arrangement matches. Probably an Occam's Razor situation here and it's just a Neopixel or comparable item in some kind of housing to change the shape from square to round.

EDIT: Or, even more likely. I forgot they make Neopixels already mounted to MiniPCBs. These even have the round shape, and are about 8mm in diameter, which fits that space perfectly.
 
Last edited:
Now that I've thought about it and studied it a bit. the red could just be an illusion. Something to do with the shadow his hand is casting onto his face.
 
Now that I've thought about it and studied it a bit. the red could just be an illusion. Something to do with the shadow his hand is casting onto his face.

Might be the case. In the scene where he's got his thumb in front of it, the top of his thumb is illuminated red, but the red never hits the underside of the pop-out component. If the red were sourced from the light, surely it would be casting on the underside of the pop-out... I only see red when his thumb is close enough to in-line with the light. He triggers it again with his thumb out of the way and you can't see it. Might just be the white LED being stupid bright enough to glow through his skin and come out red.

On the other hand, I don't see any skin in the way between the light and his face on the scene where he's holding it up to his eye. You still never see the red light hit the underside of the pop-out component, but it only does the bottom half of his face.
 
The LED on the scanner only has 2 visible connections, which would mean it can only be a single colour LED (possibly a high-powered flash LED from a phone or digital camera I suspect)
The red/purple bloom visible in the video might be an artefact of the camera's image sensor. High powered blue and white LEDs (white LEDs are blue with a phosphor coating) do emit some IR wavelength radiation as a result of heat from the diode junction, which the red part of the camera's sensor are sensitive to and will add a red tint to the blue - see the examples in this thread on the Canon camera forums of trying to take photos of blue LEDs.
http://community.usa.canon.com/t5/EOS/Blue-LED-Lighting-showing-up-as-purple/td-p/51017


 
The LED on the scanner only has 2 visible connections, which would mean it can only be a single colour LED (possibly a high-powered flash LED from a phone or digital camera I suspect)

This isn't the case, though - see one of my posts above. It has 4 wires, 2 on each side, which would be typical either for a standard RGB LED or something addressable like a Neopixel.

 
This isn't the case, though - see one of my posts above. It has 4 wires, 2 on each side, which would be typical either for a standard RGB LED or something addressable like a Neopixel.

http://i.imgur.com/lGIxqrml.png

It's also common for high-power surface mount LED packages to have 4 or 6 solder pads, allowing multiple connections for the cathode and anode to allow enough current flow, e.g:
High-Power-LEDs-Single-Color-UV-365nm.jpg
Those wires are pretty thin - I could see them doubling them up to get the current to get an LED bright enough for the camera with on-set lighting..
It's just a thought as I's seen that purple/red tint come up before when blue LEDs have been photographed.
 
That's pretty neat! I would have honestly assumed the additional solder pads would be more for the purpose of chaining them in parallel, rather than anything pertaining to current.

Those high-power LEDs are probably very comparable to the blue ones I was testing above, though, and they are blinding. I'm all for realism, but even a second of exposure to the little suckers I bought is enough to have you seeing spots. I'm inclined to use something a bit more humane for a prop so that I don't leave people incapacitated while I'm demonstrating it.
 
That's pretty neat! I would have honestly assumed the additional solder pads would be more for the purpose of chaining them in parallel, rather than anything pertaining to current.

Those high-power LEDs are probably very comparable to the blue ones I was testing above, though, and they are blinding. I'm all for realism, but even a second of exposure to the little suckers I bought is enough to have you seeing spots. I'm inclined to use something a bit more humane for a prop so that I don't leave people incapacitated while I'm demonstrating it.
I expect that's why it looks like it's strobing as it''s being driven by a PWM output from a microcontroller. It would also give an easy way to adjust the brightness on-set by using a potentiometer on an analogue input of the controller and convert it to a value to vary the PWM duty cycle of the LED driver output (which the propmaster alludes to in the video) through a transistor.

One other thing - in your earlier tests of the blue LEDs were you driving them directly off 3V with no current limiting resistor? That would explain why they were very bright and got very hot. Without a current limiting resistor you are relying on thermal conductivity of the diode junction (it conducts less as it gets hottter) and the internal resistance of the battery to regulate current. Eventually the LED will burn up as it's operating well past it's rated current. For blue or white LEDs you need to either use a constant current LED driver (which can be obtained to generate 3V from a 1.5V battery) that regulates the current to a safe level (10 to 30mA for normal LEDs, or higher currents for power LEDs), or use a higher voltage (e.g. 4.5V) with a suitable value and wattage rating current limiting resistor.
 
Last edited:
In the Tested video the prop master mentioned being able to change colors and brightness of one of the props, which made me think addressable light. It would have to be something like the Adafruit dotstar led.
 
Love it! Love your work! Love this thread!

So much love.


Sent from my iPhone using Tapatalk
 
I can't promise this will work yet, but I'm playing with an idea. Sneak peek:



I'm currently hardcoding the graphical elements so I can forego the use of an SD card, which means I think I can cut the PCB sticking out above the screen off entirely... which means this might fit into the device.
 
This thread is more than 4 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