Blade Runner 2049 Voight Kampff Eye Scanner Tutorial / Buildup

Thanks for the kind words everyone. When I started this, I really had no idea how complex it would turn out to be. Feels like I'm doing tiny tweaks all the time! I've now gone through at least 5 revisions of the main shell base... some of them major reworks.

The last render I teased above? Yeah... I redid a lot of things since then. Sooo.... time for another update.


First, some notes on the "popup armature".
1603650925356.png

This is actually an outdated version (by several versions- the latest is much refined) but it shows a direction I'm pursuing. I'm trying to have the arm made out of as much metal as possible. Getting things machined or laser-cut at this level are budget-busting in terms of cost, but I'm chasing options. I think I might be able to have the 0.5mm thin outer plates acid etched out of stainless steel. The central arm (around 2.5mm) is a bigger problem. It's too fragile to machine at a reasonable cost, so I'm looking at either having it metal printed or laser-cut. If it can be printed successfully, I can do it in one piece. If it's laser-cut, I'll have to cut only the flat portion (pictured) and "pin" it to a 3d printed (PLA plastic) "sliding block" (pictured to the left). If my budget candle take handle it, or of there is some other technical problem, I'll print these parts in PLA. As you can see, they came out rather nice! My trusty old Ultimaker 2 can still do magic with the proper coaxing. (Note: The parts in the photo have not been trimmed or cleaned up either.)

Here's a shot with a V3 (I think) body, showing the micro-switch:
1603651082978.png


Test-printing the top camera head. (Also an older version. The final one will have a separate lens, I think.) Note the texture I managed to print into the actual model itself in the midsection!

1603651125144.png


1603651198204.png


Several older shells, all different in small ways.
1603651260302.png




Here's a video showing the first big redesign of the latch mechanism. As Doc Brown says... "It WOOOOOORKS!"


I've been conducting some 'stress tests' of the mechanisms, carrying around the unit just taped together and working the popup as much as I can. (It's almost like a stress ball, haha.) Here's a fairly heavy test I did, trying to activate it as fast as I could. I was happy to note that it worked almost flawlessly. The butt sometimes catches when it's been pressed, but almost always springs back when the top is pushed down again. But there were still a lot of tweaks to be made, and... as with the other tests, there were no bearings or lubrication to help the parts move... that was very encouraging!


Another big problem I was anticipating was how to deal with the wires for the LEDs going from the head, down into the shell. Due to the real screen and other things taking up space, I will not be able to draw the cables absolutely screen-accurate, but I will get close. To make the cables move up and down, in the very cramped space, I had to build in a 'guiding canal' down into the body. But all the electronics would probably interfere with the cables, with the risk of entanglement. That's why I made a little 'central space' right under the arm where the wires could move more freely. Six wires take up a lot more space than you imagine, so again, every micron was at a premium!

Here's how I figured it working:
1603652434203.png


It sorta worked when I tried the popup, but it needed tweaking. The wires had a lot of trouble moving smoothly in the canal. So, i went back and redid that entire section:
1603652525746.png

I widened the canal and squeezed a little more space inside the shell by making the walls a few microns thinner. I also moved the recharge port and power 3-5V power converter to the other side and lowered the supports a little. The wires move a LOT easier. (I'll try to do a video of that later.)


Recharge port! Sadly, it has to be pretty big to accomodate different USB cables. Not accurate to the prop, but what are ya gonna do? Maybe a little cap can be printed or something to hide it a little... we'll see.
1603652634342.png



Here I'm testing how everything fits.
1603653266125.png


I'm hopeful that I'll be able to fit a small piezo speaker (round black thing in the middle) in there too. It doesn't really fit right now, but in the next revision I'll try to shave a few hundred microns of space so I can push it up into the rectangular space above it. Here you can also see a printed dummy of the Teensy computer. (I also have to leave space for all the other wiring, resistors and whatnot.

1603653308259.png


Testing the fit of a redesigned top camera:
1603653577220.png

(Also the first time I did a test-assembly of the complete popup arm. (It's not really cleaned up here either, just quickly glued together.) This design is close to final. I hope!

More soon... the next revision is printing as I write this!
 
Thanks for the kind words everyone. When I started this, I really had no idea how complex it would turn out to be. Feels like I'm doing tiny tweaks all the time! I've now gone through at least 5 revisions of the main shell base... some of them major reworks.

The last render I teased above? Yeah... I redid a lot of things since then. Sooo.... time for another update.


First, some notes on the "popup armature".
View attachment 1361829
This is actually an outdated version (by several versions- the latest is much refined) but it shows a direction I'm pursuing. I'm trying to have the arm made out of as much metal as possible. Getting things machined or laser-cut at this level are budget-busting in terms of cost, but I'm chasing options. I think I might be able to have the 0.5mm thin outer plates acid etched out of stainless steel. The central arm (around 2.5mm) is a bigger problem. It's too fragile to machine at a reasonable cost, so I'm looking at either having it metal printed or laser-cut. If it can be printed successfully, I can do it in one piece. If it's laser-cut, I'll have to cut only the flat portion (pictured) and "pin" it to a 3d printed (PLA plastic) "sliding block" (pictured to the left). If my budget candle take handle it, or of there is some other technical problem, I'll print these parts in PLA. As you can see, they came out rather nice! My trusty old Ultimaker 2 can still do magic with the proper coaxing. (Note: The parts in the photo have not been trimmed or cleaned up either.)

Here's a shot with a V3 (I think) body, showing the micro-switch:
View attachment 1361830

Test-printing the top camera head. (Also an older version. The final one will have a separate lens, I think.) Note the texture I managed to print into the actual model itself in the midsection!

View attachment 1361831

View attachment 1361834

Several older shells, all different in small ways.
View attachment 1361836



Here's a video showing the first big redesign of the latch mechanism. As Doc Brown says... "It WOOOOOORKS!"


I've been conducting some 'stress tests' of the mechanisms, carrying around the unit just taped together and working the popup as much as I can. (It's almost like a stress ball, haha.) Here's a fairly heavy test I did, trying to activate it as fast as I could. I was happy to note that it worked almost flawlessly. The butt sometimes catches when it's been pressed, but almost always springs back when the top is pushed down again. But there were still a lot of tweaks to be made, and... as with the other tests, there were no bearings or lubrication to help the parts move... that was very encouraging!


Another big problem I was anticipating was how to deal with the wires for the LEDs going from the head, down into the shell. Due to the real screen and other things taking up space, I will not be able to draw the cables absolutely screen-accurate, but I will get close. To make the cables move up and down, in the very cramped space, I had to build in a 'guiding canal' down into the body. But all the electronics would probably interfere with the cables, with the risk of entanglement. That's why I made a little 'central space' right under the arm where the wires could move more freely. Six wires take up a lot more space than you imagine, so again, every micron was at a premium!

Here's how I figured it working:
View attachment 1361842

It sorta worked when I tried the popup, but it needed tweaking. The wires had a lot of trouble moving smoothly in the canal. So, i went back and redid that entire section:
View attachment 1361844
I widened the canal and squeezed a little more space inside the shell by making the walls a few microns thinner. I also moved the recharge port and power 3-5V power converter to the other side and lowered the supports a little. The wires move a LOT easier. (I'll try to do a video of that later.)


Recharge port! Sadly, it has to be pretty big to accomodate different USB cables. Not accurate to the prop, but what are ya gonna do? Maybe a little cap can be printed or something to hide it a little... we'll see.
View attachment 1361846


Here I'm testing how everything fits.
View attachment 1361851

I'm hopeful that I'll be able to fit a small piezo speaker (round black thing in the middle) in there too. It doesn't really fit right now, but in the next revision I'll try to shave a few hundred microns of space so I can push it up into the rectangular space above it. Here you can also see a printed dummy of the Teensy computer. (I also have to leave space for all the other wiring, resistors and whatnot.

View attachment 1361852

Testing the fit of a redesigned top camera:
View attachment 1361855
(Also the first time I did a test-assembly of the complete popup arm. (It's not really cleaned up here either, just quickly glued together.) This design is close to final. I hope!

More soon... the next revision is printing as I write this!
Wow it’s really coming along nicely.....now I’m getting really excited!! Well done sir!
 
As promised... a video showing how the wires move inside the shell:

(The camera top isn't properly adjusted, hence the gap.)

Did yet another revision of the casing last night, and a couple other things. Also making some changes to the outer plates of the popup arm. the entire assembly might end up being five layers instead of three, depending on how it gets set up.

Still looking into having some parts etched and/or laser cut. It's not looking too good though, budget-wise, due to the small number of parts. We'll see if it happens.

Here you can see the layers of the arm, with two more thin plates on the opposite side.

1603728423434.png


The red plate is very small, 1-2mm wide. I'm a little afraid that it'll be too delicate to do. If I end up printing the plates, I can do the red and blue as one piece.

Next up is modeling the little circuit board.
 
Lunch hour hijinks... detailing the circuit board.

1604076882879.png



Here's something interesting. A test print of the central arm in metal (before 'baking' where the binder is fused with the metal). The folks where I get a lot of my supplies are checking to see if that piece can be printed on their machine. (It's a pretty big IF. I'm guessing that the cost might end up being too high, plus there are apparently issues with warping due to the small size of the part.) If all the planets and stars align... maaaaaybe but it's just an experiment right now.

1604077245495.png



Also... a new version of the case. Managed to get the Piezo speaker to fit!
1604077386914.png



Printed out a small bezel for the main LED flasher. I think that they used some kind of panel-LED on the real prop, but I'll been looking through all the big electronics sites and have yet to find a good match. So I went for a regular RPG led instead. (I actually had to saw off a few mm from the rounded tip to get it to fit in the case!
1604077524766.png


1604077540727.png


1604077548663.png
 
I was wondering if the printed metal piece you mentioned could be made slightly larger, or with some sprues that would be taken off after baking like a heat sink to help prevent warpage? Unless its a thickness issue...
 
I was wondering if the printed metal piece you mentioned could be made slightly larger, or with some sprues that would be taken off after baking like a heat sink to help prevent warpage? Unless its a thickness issue...
It's apparently a cooling issue (after actual printing is done). They are trying some things and I've sent a slightly thicker part to test, but there is basically zero leeway in terms of size. I'm counting tenths of millimeters in fitting everything. Adding sprues would probably not help, and come with another set of problems. I suspect though, due to a pretty hefty startup cost+materials, it will end up costing too much to have this piece printed. (I also don't know yet how the metal looks when it's done... it might actually look wrong. I'm also checking into having it laser cut, with the block holding the spring to be printed in PLA (more durable than resin for this particular piece) since it's not seen. But even that may cost too much in the end. Metal plates would be awesome though, so we'll see where it ends up!
 
Lunch hour hijinks... detailing the circuit board.

View attachment 1363438


Here's something interesting. A test print of the central arm in metal (before 'baking' where the binder is fused with the metal). The folks where I get a lot of my supplies are checking to see if that piece can be printed on their machine. (It's a pretty big IF. I'm guessing that the cost might end up being too high, plus there are apparently issues with warping due to the small size of the part.) If all the planets and stars align... maaaaaybe but it's just an experiment right now.

View attachment 1363445


Also... a new version of the case. Managed to get the Piezo speaker to fit!
View attachment 1363446


Printed out a small bezel for the main LED flasher. I think that they used some kind of panel-LED on the real prop, but I'll been looking through all the big electronics sites and have yet to find a good match. So I went for a regular RPG led instead. (I actually had to saw off a few mm from the rounded tip to get it to fit in the case!
View attachment 1363448

View attachment 1363449

View attachment 1363450


Just a quick question regarding the LED's. On the original prop the main LED appears to flicker between blue and white with a static red on the side. Is this something you are also trying to replicate?
 

Attachments

  • rpQakfk (1).gif
    rpQakfk (1).gif
    1.3 MB · Views: 105
Just a quick question regarding the LED's. On the original prop the main LED appears to flicker between blue and white with a static red on the side. Is this something you are also trying to replicate?
Absolutely. It's actually more important than having a working screen with animations.

Due to the low frame rate of film, it's hard to know exactly what the LED strobe sequence is, but it looks like it goes "white-red-white-blue" and repeats, since you can also see the red flashing. It's actually possible that it only flashes white, but that they used an RGB LED, because since they essentially contain three separate LEDs, you can sometimes see the three colors separate inside them, even when you want it to shine white. Though, I'm leaning more towards the alternating color sequence. We'll see how it looks once there's some more coding done.

The tiny red LED is static and always glowing when the top is up. On the real prop, it's almost certainly a surface-mounted led, because it's VERY small, but since they are not easily soldered with the kind of soldering irons most of us have at home (i.e. you need a precision instrument for working on circuit boards), I might be using a flat 2mm regular led instead. It's still kinda TBD.

The electronics/animation sequence is still being ironed out and not final, but the aim is for the screen to flash once when it pops up (like it does in the film), at which time the red LED will also light up. IF sound is possible (see the note on memory below) there will be something like a 'camera flash whine' when you start it up. When you press the trigger, the 'scanning' LED will flash in sync along with the screen, while the red LED remains static, then when you release the trigger, the unit will "think" for a couple seconds, then display suspect info along with some animations and some kinds of 'sound blips'. If you press the trigger again, the screen goes black and the entire scanning sequence restarts. And when you push the top down, it'll just go dark and silent. In aperfect world, the unit would always be on 'standby' but since the Teensy computer needs a second or two to load the operating system, we could not have the screen accurate 'flash once' unless the unit is already powered up, so I'm trying to sneak in an extra power on/off switch somewhere, probably under the charging point, so that it can be turned on with a toothpick or something. Regarding audio: Sounds are still very TBD because it's all down to how much memory there is left. My programmer friend is in charge of that and, from what I've been told, just getting the graphics onscreen has eaten more than 60% of it already.
 
Absolutely. It's actually more important than having a working screen with animations.

Due to the low frame rate of film, it's hard to know exactly what the LED strobe sequence is, but it looks like it goes "white-red-white-blue" and repeats, since you can also see the red flashing. It's actually possible that it only flashes white, but that they used an RGB LED, because since they essentially contain three separate LEDs, you can sometimes see the three colors separate inside them, even when you want it to shine white. Though, I'm leaning more towards the alternating color sequence. We'll see how it looks once there's some more coding done.

The tiny red LED is static and always glowing when the top is up. On the real prop, it's almost certainly a surface-mounted led, because it's VERY small, but since they are not easily soldered with the kind of soldering irons most of us have at home (i.e. you need a precision instrument for working on circuit boards), I might be using a flat 2mm regular led instead. It's still kinda TBD.

The electronics/animation sequence is still being ironed out and not final, but the aim is for the screen to flash once when it pops up (like it does in the film), at which time the red LED will also light up. IF sound is possible (see the note on memory below) there will be something like a 'camera flash whine' when you start it up. When you press the trigger, the 'scanning' LED will flash in sync along with the screen, while the red LED remains static, then when you release the trigger, the unit will "think" for a couple seconds, then display suspect info along with some animations and some kinds of 'sound blips'. If you press the trigger again, the screen goes black and the entire scanning sequence restarts. And when you push the top down, it'll just go dark and silent. In aperfect world, the unit would always be on 'standby' but since the Teensy computer needs a second or two to load the operating system, we could not have the screen accurate 'flash once' unless the unit is already powered up, so I'm trying to sneak in an extra power on/off switch somewhere, probably under the charging point, so that it can be turned on with a toothpick or something. Regarding audio: Sounds are still very TBD because it's all down to how much memory there is left. My programmer friend is in charge of that and, from what I've been told, just getting the graphics onscreen has eaten more than 60% of it already.

Thanks for confirming all that :) I can't tell you how pumped up I am about this...thanks again for doing it :)
 
Update time. Now, I originally never envisioned actually having the popup assembly in real metal. Sure, it was a 'would be awesome' thing to have, but I didn't really believe it would happen. However, I'm getting close to having some true, viable options for it. If printing the center arm in metal doesn't work or is too expensive, I'm still thinking it can be done in laser-cut aluminum. Still, I'm not counting any chickens yet.

I was able to have some test-pieces laser cut to test and look at. Here is a metal cut outer plate (sloppily) glued in 2 layers to a plastic center arm to check functionality. The little circuit board is a 3d print that I brush-painted really quickly. (For the metal circuit paths, I used a Molotow marker under a magnifying glass!
1604529777405.png


Here's a comparison between a laser cut plate (on the left) and a PLA plastic 3d print covered in graphite powder. (Neither is the absolutely final design.)
1604530145632.png



And here's a closeup of the little circuit board. Depending on how big your monitor is, the photo probably shows it at twice the actual size. The thing is TINY! (Fun tidbit: I'm pretty sure I know where the original screen circuit board and popup arm came from... a $1000 Sony camera. So it's not like I'm going to buy one to take apart and check!)
1604530184140.png




Finally, here's a render of the almost-final body!

1604530069849.png
 
Great mechanism test video on YouTube
Thanks! I was intending to post it here with the next update (didn't realize I had it set to public, haha) but here it is:


I'm hoping the programmer will be able to send a short video of where he is on the graphics/animation soon. I recently delivered some updated, cleaned up, graphics files to him.
 
I should add that in the video above, I have inserted the two ball bearings to make the movement more stable due to the added weight of the metal plates. It doesn't make a huge difference really, but I'm glad to have them there. I initially planned for three bearings, but am finding that it's overkill. I also added a second spring to the popup activator for more stability. The button would on rare occasions stick a little if you pushed on the bottom part of it. (That might've righted itself with some lubricant, which the test unit still has none of. The resin I use is very good- especially for paint to adhere, but the matte surface also makes it stick a little when two parts rub together, so it needs a little help. That the mechanism is working without any lubricant is a very good sign!)
 

Your message may be considered spam for the following reasons:

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