I made a fake CRT TV with a LASER

[@bitluni]

Check out brilliant.org/bitluni/ for a 30 day free trial or a 20% discount for an annual subscription (sponsor)

[@Randomman149]

ONE DAY AGO? It was uploaded 31 mins ago bro

[@Nobe_Oddy]

YES, TWISTING WIRES IS GOOD. when you twist 2 wires together (2 wires of the same circuit... so 1 wires is the positive and the other wire is the negative) they 'magically' shield themselves from outside interference... this is why you see 4 twisted pairs of wires in an ethernet cable.... but you CANNOT have more than 1 circuit in a twisted pair... this will interfere with it self and can ruin both signals... - I JUST explained this SAME THING to another channel about 2 days ago :D - So going to school for Network Administration LEARNED ME SUMTHIN!!!! 🤪😁😳😂

[@francoisleveille409]

May I suggest for our common safety that you forgo the gravitational field project ? A small black hole on earth sweet jeezz!!

[@cheeto4493]

@@Randomman149 uh, released to members early?

[@cheeto4493]

@@Nobe_Oddy You can only have one circuit in twisted pair, cause you only have 2 wires. Otherwise it would be a twisted group, which is not a thing.

[@EddieHart]

I suggest you buy yourself a cheap laser barcode scanner and see how that creates a linear beam from a single laser diode. It typically uses a mirror fixed onto some compliant material/thin plastic hinge (something thin that can bend) so it can wobble along an axis side to side while remaining balanced. The mirror has some kind of light magnet on the back and a voice coil behind it is used to attract/repel the mirror at a specific frequency, causing it to wobble along the hinged axis. Because the mirror is balanced on that axis, you don't need to move the full weight of the mirror meaning you can do it faster than with a motor or the speaker method you used, where you have to move the entire weight of the mirror which is much slower because the system has a higher inertia.

[@jhonbus]

That would make it a raster display instead of a vector display though!

[@MadScientist267]

One thing.. vector displays don't use line by line scanning.

[@EddieHart]

@@jhonbus It would if you oscillated the mirror at a set frequency (the same way the barcode scanner does) and used scanlines. But I meant more that this approach could be investigated to see if it's feasible to move the mirror to an arbitrary angle using the same technique. I don't know how much the oscillating of those mirrors is sustained by simple harmonic motion as opposed to actively being moved by the voice coil - I suspect it's probably more the former, but still worth a try.

[@cheeto4493]

@@MadScientist267 if you varied the voltage, you could vary the deflection. Similar to his first attempt with the speakers as drivers. I don't think Eddie was suggesting a line like in a raster scanner, but the hardware style used to deflect the beam.

[@SmallSpoonBrigade]

@@EddieHart TBH, probably the easiest thing to do is just rotate the laser and block it when it's not oriented towards a part of the visible screen. That way, you don't have to stop the mirror and switch directions. Which just means that you have to raise and lower the assembly as you go through the screen. If you go from top to bottom and then bottom to top, it wouldn't be as authentic, but as long as you mirror every other frame, it probably wouldn't be visible to the viewer.

[@cll1out]

I implore you to find an old laser printer. The laser module in it will have a laser along with a high speed rotating mirror. The idea is when the reflected beam runs off the end of the drum, the next side of the spinning mirror starts reflecting the beam at the opposite end. Get a color laser and you might find four of these laser assemblies! You could build a raster system out of such a setup.

[@MarkZimmermanKicksAss]

This. I made an analog version of this in the 80's to use as a night club lighting effect. Used octagonal prisms (what you're calling a drum, I think) and it could run in vector mode or raster. Each motor had it's own function generator. It was about 20 years before I discovered microcontrollers, sadly.

[@Axel_Andersen]

A rotating mirror does not lend it self to random XY movements very easily.

[@Jackpkmn]

@@Axel_Andersen It doesn't need to. Just like with the assembly shown in this video you use a second mirror assembly so each assembly is handling one component of the XY movement. This is a long solved problem that's why you see so many similar comments about the same kind of mechanism in these comments.

[@Axel_Andersen]

@@Jackpkmn I still do not see how two rotating mirrors make a good XY movement (which to me imply random or vector like drawing). For raster scan yes, but for XY, I don't see it. A rotating mirror needs to accelerate and decelerate relative sizable mass to do random movements. Of course if you consider a galvanometer as rotating mirror then sure, but to me rotating implies more or less continuous motion. I guess it all depends on what is our built in vision when we read a word like 'rotating' mirror.

[@Jackpkmn]

@@Axel_AndersenThe more sides the mirror has the less the speed has to be modulated to adjust the timing. The only actual limitation is that you can only draw in one direction, the rotation direction. So if you wanted to draw two parallel lines you would have to draw them in 2 separate passes or raster scan them in. Drawing them as two separate passes doesn't have to be as much of a problem as that probably sounds at first with a mirror with many sides that spins relatively quickly. Since the number of times you can scan the screen per second is the number of faces the mirror has times the rpm of the motor.

[@redderthanmisty6762]

"Then you have vim... which can only be exited by turning the system off and on again" 🤣

[@039bb]

:wq!

[@DasIllu]

Yep, that's VIM. Can't remember all the shortcuts, sacrifices have to be made 🤪

[@SnakebitSTI]

Esc, :q!

[@Riichrd]

Killed me! 😅🤣

[@thatcatthatalwayseatsyourchees]

@@DasIllu :q

[@Lampe2020]

1:12 "draw permanent lines on your retina" is certainly a way of putting that XD

[@eat_things]

2:12 A sudden electroboom appears...

[@xavicariteu]

@ElectroBOOM, take a look!

[@powertomato]

Beat me to it It was my thought as well

[@jonny11bonk]

Der Elektroboom

[@robertk1701]

double checked the channel name when that happened

[@user-qr4jf4tv2x]

sounds like him at one point

[@Kruglord]

As others have mentioned, a spinning triangular prism with front-surface mirrors on each face is probably the best way to create your horizontal scan lines. That's what's typically used in aerial lidar scanners, and it's very reliable. You could increase the number of faces to reduce the RPMs but increase the scan rate, but doing so also decreases the angle they sweep, so would require a longer base-line for the same size of CRT screen.

[@BolverBlitz]

Love how you're shifting (back) to historical and educational content-such a refreshing change from the complex PCB stuff.

[@Aydenitro]

Soscwna7scsns8sga mama absurd dhsfs ajaf

[@Oli1974]

That is fascinating! When I was like 14 or 15 yo, I built sort of a mechanical oscilloscope, just because I liked constructing things. It had three mirrors stuck onto a rotating three-sided column driven by an electric motor and, because there were still no laser pointers or any other affordable lasers around, I used a small halogene light bulb focused through a lens whose beam was reflected by a small mirror stuck to a speaker and then, after passing another lens, reflected by that fast rotating mirror column to the backside of a matte glass screen. It was blurry but it somehow worked! You could adjust the "trigger" through a potentiometer that controlled the motor speed. I demonstrated it to my physics teacher and she was amazed.

[@leonardomarchesi6169]

When I was younger (35 years ago) I intendend to build a TV with laser, but the main difficult was the beam deflection. At that time my efforts focused on using voltmeter device. The idea was to glue a small light mirror to the axis of the arrow. The higher the voltage the higher would have benn the deflection angle. But the speed required was too high. One full frame in 20 ms (50Hz) must be created by 625 lines, means a sweep (the line) every 32 microseconds. Really too fast. If we used a triangular mirror on a motor, this should have rotated at about 10.000 rps! And with a such precision....

[@Oli1974]

@@leonardomarchesi6169 Mine rotated at about 1500 rpm and it was well sufficient to at least draw one contiguous line. But was incredibly noisy, and the startup of the thing sounded like switching on a mad scientist's machinery haha.

[@James1095]

When I was a teenager in the 90s my friend and I built a number of laser light show devices using head positioning mechanisms from discarded hard drives. They were open loop so they could not generate vector images but they did produce some nice patterns when driven with audio.

[@mikeselectricstuff]

Your first mirror could be a lot smaller - less mass & air resistance

[@bitluni]

not only that: first mirror should rather do the horizontal deflection and the second one the vertical. but don't tell the others 😂

[@AdiGitalEU]

@@bitluni I would also suggest using old hard drive aluminium plate cuts as the mirror. The surface is the reflective layer and should provide better focusing.

[@lorenzo.c]

@@bitluni I wonder if you thought of and then discarded using the voice-coil motor of a mechanical HDD. Those motors, which position the arm carrying the heads at the right spot on the platters, have a coil between permanent magnets, a bit like your speakers at the beginning. It's easy to find examples of people placing a mirror on the axis of the arm, pointing a laser at the mirror and then driving the motor with an audio amplifier playing music. In your case you would have to code a more sophisticated control algorithm. I remember finding the video of someone doing precisely this with an Arduino but I could not find the code. Extraordinarily cool project, by the way 😄

[@mal2ksc]

@@AdiGitalEU Any front-silvered mirror will do for the job, and hard drive platters are kinda hard to cut to size and shape although they're great if you can use them as is.

[@doctorweile]

Not sure how much it would matter, considering the mass of the stepper motor rotor itself. Perhaps a better solution would be to use a mirror on both sides and refrain from using reciprocating motion, but rather to have the mirror spinning continuously.

[@threeMetreJim]

Might be nice if you can program 'Asteroids' for it. That old 1979 arcade machine used a vector display, and that had a unique look.

[@lucasrem]

threeMetreJim Vectrex, the MB Arade system at home, vector CRT system that was, but people hated that it was Black and White only. program ? What you need to program here ? Why not just run it ?

[@threeMetreJim]

@@lucasrem Generally you can't just provide copyrighted code to run on an emulator (is there even an emulator that can draw a vector display that runs on the esp32?); I'm assuming the original Asteroids arcade code is still under copyright, so you will need to create a clone; hence 'program' or find a version that does the vector drawing that is public domain.

[@RichardBetel]

I wanna suggest a rastering mechanism: You get an octagonal thingie with mirrors on each of the 8 faces, and spin it at 225rpm, then shine a laser on it. You'll get a straight line with crisp start and ends, re-painted 30 times a second. That's your vertical sweep. If you build a second spinning octagonal mirror spinning at 1800 rpm and arrange them so that they spin on axis 90 degrees apart, you should be able to get a NTSC image. Now you just have to switch the laser diode on and off fast enough to get a picture! You can manage 350kHz on the diode control, right?

[@mal2ksc]

Wait, aren't you just stealing the guts from a laser printer at that point?

[@Meower68]

@@mal2ksc That's *EXACTLY* what you're doing.

[@threeMetreJim]

The image might be quite dim though, as you have to spread the power of the laser out over the whole screen. I did try it with leds once, for a projected 8 line moving message display and noticed that enough power to spread out to get the brightness was a problem, this was one led per line too. A vector display can sort of 'cheat' that problem.

[@hyperturbotechnomike]

@@mal2ksc I remember watching a video from someone a few months ago, who built a long distance laser projactor using this method. He even got multiple laser beams in colour (RGB)

[@Djhg2000]

​@@threeMetreJim Could be, but judging from the clipped highlights at the beam point there should be enough to at least make a discernible image. The screen seems to have plenty of retention time so I guess you could just run it slower and give the beam more time to penetrate throughout the filament. Maintaining synchronization of the spinning mirrors might be a bigger issue though, I'd go with a mechanical geared solution where the slow mirror is driven by the fast mirror. LEGO gears are cheap and small but to my knowledge are all straight cut which would be noisy at those speeds. Herringbone gears have continuously engaging and disengaging teeth they don't have the same pulsing kind of noise as straight cut gears. 3D printed ones can be made quite large and heavy which smooths out the drive speed over short periods. With a brushed motor a timing wheel with optical sensors to trigger vertical sync would also be needed to compensate for fluctuating mirror speeds over longer time periods, but the easier solution is probably a brushless motor with a good driver that doesn't have noticeable jitter in the rotor logic.

[@ItzTheMrMan]

The Question many want to know now. Can it Run Doom?

[@Vrboy_418]

What I was going to say

[@unlokia]

Who cares. Such a passé cliche.

[@savethedecieved27]

the actual question is can it display doom

[@EyemachineStudios]

THATS WHAT I SAID

[@TheRenegade...]

Run Doom? No, there's no computer conponents. Display Doom? Probably, but not with color

[@KRtekTM]

The performance in 13:29 is awesome! (whole project is amazing, thanks for sharing it :))

[@flash001USA]

Love it. I've thought about this when lasers started hitting the market at affordable prices. Years ago when lasers were not solid state (or cheap) they were still being used in elaborate light shows for well known big bands who could afford the technology drawing shapes on a back drop during a show. Even then I wondered if you could use 3 different color lasers for producing a color pattern with more detail sort of based on a color television design. I had doubts that someone would ever a get fast enough scan rate with a mechanical setup to make a picture or a shape legible then you used the photosensitive film as a resolve and made it work. Nice job.

[@richardzeitz54]

You should try brushless DC motors! They can go VERY fast and stop on a point. They don't need anything like much torque. Maybe not as simple a steppers but they're used by some corporations for very precise work nonetheless. Your initial idea, using speakers, was the best. I'd love to see it scan fast enough to serve as the display for a videogame like Asteroids or a Vectrex title.

[@clockworkspiral]

faster dc motors could probably lend themselves to a more continous-scan type design too, depending on how fast they can drive them. some sort of linkage that directs the horizontal deflection to a set rate, and the vertical deflection at a set rate with some way of resetting it. i wonder how fast the luminance of a laser diode like that can be modulated or switched, would be cool to see some sort of continuous scan rendering by adjusting the brightness of the diode itself like that.

[@mal2ksc]

How about voice coil head stepper assemblies from dead hard drives? Those can move really fast, for obvious reasons. I bet they could move even faster with much of the weight removed (all but one arm snapped off). Also speed is helped by keeping the mirrors as small as possible, so there isn't so much inertia to fight. The first mirror in particular doesn't need to be as big because it is receiving a stationary input.

[@josefnovak7496]

Couldn't the mechanisms from old HDDs that move the heads be used? I think they would be fast enough to move the mirrors.

[@richardzeitz54]

@@mal2ksc awesome idea!

[@richardzeitz54]

@@mal2ksc great!

[@computersales]

I love how the internals started out well laid out then became more chaotic as you added features. 😁😁

[@didierdubos]

next : convert to a Vectrex video game console

[@Eliasdbr]

Yes please, no matter how low the speed could be

[@howdidigethere-5226]

My first thought was hey it looks like my vectrex, crazy

[@FuZZbaLLbee]

Vectrex mini confirmed 😋

[@cjermo]

This was the first thing that came into my head.

[@nathanstein589]

That would be really easy, we already have laser machines that can output vector graphics from vintage arcades and all you’d need to do is use a UV laser.

[@neogeo8267]

Another option is to make a multi-faceted rotating drum. Each facet is at a different angle. As it rotates, it sweeps each angle across. Each sequential angle is slightly higher, so it scans left/right and slowly up/down.

[@ArneChristianRosenfeldt]

But wouldn’t that shift the beam? I was thinking about the vertical mirror after this. Do we need a lens to hit it ? Laser Gaussian beam lens waist on polygon scanner 1:f lens pair waist on galvo large concave mirror waist on pixel

[@redleader7988]

My idea was just to have two drums each with increasing angles rotationally for the x and y axis. No problems with dark areas in the center as with flat prism and mirror solutions because the drum is already curved, and without the problems with inertia that we have with oscillating mirrors. The machining tolerances of the drums would be critical but doable.

[@neogeo8267]

@redleader7988 maybe 3d printable with attached facets? I may not be imagining what you're describing though

[@redleader7988]

@@neogeo8267 I wasn't thinking in terms of facets. I was thinking that if the surface of one of the polished drums is helical, it could deflect the laser beam gradually across one axis (e.g., the horizontal direction, or X-axis) as the drum spins. The idea is that as the drum rotates, the angle of the reflected beam changes smoothly, causing the laser to "sweep" across the screen.

[@neogeo8267]

@redleader7988 interesting! Machining or otherwise manufacturing such a surface would be challenging, unless there is a technique I'm unaware of. I was imagine facets such that each was a flat surface. Each face would be angled vertically more than the previous one. As the drum rotate, the beam would sweep right left, then the next facet would start the sweep over but one line down. You would need the same number of facets as lines which means either a very large diameter drum or a series of lines of facets and a gross-adjust galvo to switch between them. An absolute position sensor and good speed control on the drum as well

[@Jamman88888]

I really loved this project and cant wait for version 2. Would love to see some vector games running on it, battlezone would be incredible.

[@sickness80]

this! :)

[@dogoonubs997]

I like how he calls this a “fake” CRT, even thought it’s actually kinda close. While the method used to generate the light is different, as is the way the light is manipulated to points on the screen, the concept of the pattern of display is almost identical. Like others said, if he added a rastor mechanism it would be almost perfect.

[@ArneChristianRosenfeldt]

No cathode , yes ray , no tube

[@lucasrem]

Replace the Cathode Ray for a Laser. Like others said, play games on it all others say ! Understand it ? Use magnets on the lenses to move the laser beam over the screen !

[@ArneChristianRosenfeldt]

@@lucasrem I know these scanners as others have said. Just I hate that people always dismiss the engineering in CRTs . I studied years to know how to create a low vacuum and to create low work function surfaces. Electron optics. Yet here it all gets dismissed as unimportant .

[@silber7010]

For the horizontal line you can also maybe use for example a hexagonal shaped mirror, then that stepper can rotate at a constant speed to draw multiple horizontal lines. But that works only for TV mode, not oscilloscope!

[@atari7001]

It was think the same. Seems like it would be easier to just modulate the laser.

[@virtuallyaverage9357]

Id wonder how quickly the Lazer can be switched on and off, I'm sure there's a couple ms delay, which would really limit the max speed

[@SmokeyWire56]

I'm basically trying do this. I used a square but the beam was too wide. Im going to try a pentagon, then motor will only have to spin 15,600 ÷ 8 about 2000 rpms. And the other motor is just a brushed motor that shimys back and forth at 60 or 30hz. No fancy computer stuff needed. And the cathode current controls the lazer brightness.

[@BritishEngineer]

6:22

[@BritishEngineer]

@@virtuallyaverage9357​​⁠not lazer, but Light Amplification [by] Simulated Emission [of] Radiation

[@ChannelSho]

This seems like a cool way to revive the feel of CRT without needing an actual CRT.

[@steffenjendrusch3734]

Reminds me of our laserscanners we did 20 years ago with the popelscan software. The solution we found was not to use steppers with snmall steps but steppers with really big steps of appr. 7°. so really corse stepping. Then you hold the rotor with a voltage between 2 steps and then modulate this voltage for scanning. So you get a few degrees of scanning and that doubled as a scan angle. No steps, no jagged lines. But as you already mentioned, there are cheap galvos available now.

[@ThingEngineer]

I love this project and I love that you also included the universally correct way to exit VIM. Great job!!

[@JohnDlugosz]

2:25 ElectroBOOM would be proud.

[@Gambiarte]

kkkkkkkkkkkkkkkkkkkkk good one!

[@WB_OZON_YANDEX_BOX]

Да бро

[@NunoFerreiraX]

😂

[@BBB_bbb_BBB]

Even the accent matches

[@jimmimak]

I think the speaker option was on the right lines, you just needed to play with the mechanics of the mirror to get larger angles of reflection from a smaller speaker displacement. For example, make the distance from the fixed end to the moving end of the mirror smaller. There's also voice coil actuators, which are like speaker coils, but they have no diaphragm or cone, and they have larger displacements. You can make one yourself with a cylindrical magnet with north and south at each end, and tubular coil of wire hooked up to an audio driver. The old CRT screens had a ~15.7kHz horizontal scan rate, within the audio range, so speakers and voice coils should work.

[@Stabby666]

This reminds me of the Tektronix 4050 "storage tube" vector displays. The CRT itself would retain an image, so no display memory was required. It had some obvious limitations - the display had to be "undrawn" to create animation, but it looked really cool and could create beautiful images with relatively simple hardware driving it. I think there was a later version that had 2 colours - one was the static image color, which was retained automatically, and another colour was only temporary, allowing animation over a static background without the CPU needing to worry about handling the background redrawing. There are videos of it on YT - very cool!

[@georgkrahl56]

The Tektronix 4010 (and so on) have been storage tube computer terminals. xterm has still a tek 4010-mode, changing fcolor to green gives back the feeling :-). Mabe one can find still some programs.

[@Stabby666]

@@georgkrahl56 I don't think xterm would work with the 4050 - it's a vector display, so wouldn't scroll, at least not the phosphor-stored image. The immediate mode drawing would though. It's amazing that they had this 4000x3000 (effective) resolution display with no pixel "steps" in the 1970's - with really fast animation though.

[@OoGICU812]

This is fantastic! Years ago, I thought about applying this concept to make an oscilloscope. This is proof of concept, thank you

[@devWeidz]

"And we have Vim, the only way to leave that is by turning off and on again" 😂

[@datachu]

Can relate

[@narrativeless404]

I felt that, despite only ever using Linux to study.

[@sqeekykleen49]

​@@narrativeless404as a beginner, which version of Linux would you recommend to start out with?

[@ericov.o.2399]

If you wanna learn go with the hardest possible, not gentoo because it takes a lot of time to compile some things, try bare debian.

[@shabath]

​@@sqeekykleen49Go hard or go home, linux from scratch. On more serious note, just pick a distro and start messing with it.

[@slim7306]

this concept already exists and is implemented, it's called an LPD (laser phosphor display), the patent is sadly hogged by Prysm which means we don't have it for regular consumers.

[@N4CR]

It doesn't use galvo scanners, so this isn't patented. Galvo scanners are too slow, 60kpps maybe 90 with tiny beam. Prysm uses a solid state scanner (not MEMs).

[@GeminoSmothers]

Patents belong in the trash.

[@msrblonline]

100% feature complete vi. nice job!

[@leftaroundabout]

Woah, no it's not complete! Every proper vi implementation comes with a set of arrow keys. Their purpose is to summon a vim nerd who will then tell you off for using the arrow keys instead of the way superior hjkl keys.

[@davidonguitar]

This is awesome. Struggling with depression lately after quitting both THC and nicotine simultaneously. This just showed up on my homepage and I need something to distract myself with.

[@LiebeJung-c6v]

Keep it up dude!

[@oflasho1]

Any technological DIY toys can make boys forget all their worries~ Come on bro

[@saitamatechno]

that's a great project! A screen without a glass, amazing!

[@imacat643]

8:47 twisting 2 wires together does actually do something! increase cross talk noise from magnetic currents! all tho you most surely wont ever encounter this like at all on such low power, so keep doing it! it does look kinda cool!

[@InfamousSabreMods]

The speaker method was the correct one, just with the wrong implementation. Speakers are simple solenoids. Find some solenoids and use those instead. You'll attach one edge of the mirror to a solenoid and the other to a hinge attached to a fixed point. Do that once for X and once for Y. That will be much faster and more precise than your current method.

[@mal2ksc]

Or if you want to raster scan, one of the two mirrors could be multi-faceted and in constant rotation like a laser printer. Then there's only _one_ solenoid to worry about and it can carry a pretty tiny mirror since the input beam is not yet moving.

[@erlendse]

First that comes to mind is HDD parts aka the head movement part. Second thing that comes to mind is laser galvos or using needle meter movements. You got the right idea.

[@ArneChristianRosenfeldt]

@@erlendseHDDs use voice coil . Galvo scanners have a different magnetic field , but the same coil

[@niko5008]

Oh my god this was running in my mind for YEARS. I was like "huh if a crt scans the screen with a single point like a laser... what happens if you do it with a laser...?" This is so awesome!!!

[@matik09]

The question is, can it run Doom?

[@Napolitaninh0]

Yes it can, very slowly bcuz it uses slow motors but runs

[@FlfLuvr]

this takes an "analog" tv to a whole new level, this feels extra EXTRA analog, feel the moving parts!!

[@ashadowawhisper]

16:03 ElectroBoom moment.

[@BeefJerkey]

Main problem is the mechanical movement of the mirrors. If you could figure out a way to move them with electromagnets, that may work faster, but I'm not sure exactly how you could make that work well. Years ago, I found myself wondering if it would be possible to built a modern recreation of the old Vectrex game console, without the CRT, but a new kind of vector display. This seems like a step in the right direction, but the laser could be risky.

[@JamesTitcombOSwarthoull]

replacing the speakers with the head arm and coil out of a hard drive might be an option? you will get plenty of movement and a good high speed, but there is no feedback and no auto return.

[@illtemperedavians]

Just last week I was dreaming up a display using a single strip of RGB LEDs and oscillating mirror to with magnifier to make scan lines like the Virtual Boy. This week I see this. Amazing prototyping work here!

[@erroxcoder]

I accidentally skipped to this minute 02:18 it reminded me of elector boom XD

[@Jellolotl7468]

This is a really cool concept! I really want to this innovated on and upgrade. It would also be way more effective then actually producing true CRT tubes with a vacuum

[@Simon_Rafferty]

Try replacing the Stepper Motors with regular DC Motors. They behave fairly similarly to a speaker coil, except with more displacement.

[@mdhasiburrahman8806]

Man I am blown away by the end product

[@TheGreenking]

THIS IS THE FUTURE OF CRT'S. I FEEL LIKE THEY WILL HAVE A RETURN!!

[@MsHojat]

Maybe 20 years ago you could say that, but these days that is absurd. We already have amazing OLEDs and LED tech that will just get better with time and is already phenomenal.

[@tnmrvc]

@@MsHojat maybe 20 years ago? you're both being absurd, the CRT's we had, relying on electron guns were the superior ones, while this project is great for illustrating design and operational principles, using a laser with mirrors will always be too slow and imprecise.

[@KrotowX]

Nope. Unfortunately CRT tech has a ton of drawbacks - technical and space considering. Though for cheap projection solution this project would work.

[@tnmrvc]

@@KrotowX bro shut up and dont interject with something you dont know shit about

[@GammaRayTech]

@@MsHojat I legit dont like my oled LOL. There's a reason all my 4K Monitors and my OLED are used for youtube and google tabs and my CRT's are used for gaming or movies, etc.. I've got 3 Panel monitors and 2 CRT Monitors and 4 CRT TVs, got rid of my panel TV legit hate modern displays, they make me feel sick when I can literally see every frame skipping to the next and the disgusting blur. And naw, dont go about black frame insertion, I ain't want darker images to reduce motion blur, I want vivid image with no blur and instant response times. Yeah CRTs aren't perfect, and neither are OLEDS.

[@muf1772]

There are two types of galvanometers: regular galvos and resonant galvos. If you want to attempt to display raster graphics, you want a resonant galvo. It's not as precise as a regular galvo (so it's not suitable for vector graphics), but it's capable of much faster speeds, and should be able to display a 15KHz raster (240 lines). That means you can play DOOM on it! Of course, the next step after being able to display true raster graphics would be to use three lasers for an RGB colour monitor! Looking forward to future development.

[@rayraysss]

"The only way to quit vim is to turn off and on" that seems accurate

[@Tazanthro]

Wow! For the last week Ive been trying to figure out how to make a mini curved crt screen with static. Looks like I'm not the only one trying to replicate mini tvs! Great vid! Got a new sub

[@mr.coolio4321]

I am convinced that you are the most creative man on KZbin.

[@stevenmcmaster]

nice work mate. love the fail and success waves with a detailed view on final version. so inspiring

[@nutzeeer]

Whats the song at 0:22?

[@Sporquill]

It seems to be “Powerup!” By Jeremy Blake :)

[@Viscous_Flow]

@@Sporquill absolute legend you are

[@fiveoneecho]

It’s incredible how many comments are saying, “I know how you can improve the mechanics! You just and it works in the way I understand how pixel displays work. Much better!”

[@derre98]

I always thought it would be really nice to make a VGA-compatible RGB-laser based CRT-equivalent, but never could figure out a way to scan the beam fast enough.

[@UNSCPILOT]

As others have mentioned, something like an pair of octagonal spinning mirrors rotating at high speed would work fine, so long as you align the three color beams together and get the speeds right you just need to modulate the RGB beam powers and you'd have a laser "CRT", it probably wouldn't be compact but it would probably be lighter than a comparable CRT, and potentially use far less power as generating and steering electron beams is power intensive

[@derre98]

@@UNSCPILOT I suppose 236,000 RPM could be borderline doable, but I'm a bit sceptical that would be practical or quiet or stable enough. Some sort of fully electronic solution like voltage dependent refractive index would be better, but I don't recall seeing practical values for such either. Maybe someday.

[@ArneChristianRosenfeldt]

@@derre98electronic solutions can shift the phase by 360° . A mirror does 1000 times the phase shift. A hologram? But the electronic I used needed 10 kV . You could send chirped ultrasound pulses through a crystal.

[@ArneChristianRosenfeldt]

@@UNSCPILOTprism or dichroic mirror? Or a grating?

[@dansw0rkshop]

Use the magnets/arms from hard drives. You have a very fast travel and a lot of movement, way better than a speaker.

[@MD_Builds]

Silly question.... Why not use real galvo motors? Those things can run at the speeds and accuracy you need.

[@janikarkkainen3904]

15:48 "I already ordered an off-the-shelf galvo" ...

[@stephenkeen6044]

You want a pair of simple single coil galvanometers to do the deflection (like you find in old analogue multimeters). Just a simple coil with a magnet mounted on a pivot and spring. Can drive them from simple transistor amps on analogue X and Y signals, or directly with PWM. Can have very smooth movement, although you'll have to take momentum into account if you want high precision. Or if you want a CRT style scanning, can use a pair of rotating multiface mirrors, one at slow speed for Y, one faster for X. The speed is critical on these ones, though, to keep signal synchronised. Very frustrating when youtubers just go at it without doing any research on what other people have done in the past to already solve these problems (and can be found with minimal effort). Reinventing the wheel for "content" is just...

[@theHoIe]

10:10 rotating 3d pringle chip

[@anaveragecanadian5061]

i see the meta quest logo

[@RoofusKit]

There are lots of projects out there for home made SLA printers. I would think that would have been the best place to start as they have already worked out the high speed UV laser deflection and must be very accurate.

[@lukasrgl]

3:11 Oh no we killed this Motor... Scheiße

[@rdqsr]

Random idea but one way you could possibly quickly and accurately move the mirrors could be to gut a hard drive or two and use the mechanism that moves the heads. Since it uses magnets it might be a lot faster than using motors.

[@purpleduggy7680]

Please please please develop this into a working display for analog retro gaming. A 22" Color Laser Phosphorescence Tube would be incredible and would probably have insane refresh rates, not to mention massive brightness and contrast benefits, I would bet you could achieve far higher resolutions on this even.

[@werpu12]

He basically replicated the mechanism of a vector display like the Vectrex or some Arcade games used, in a modern fashion! Great stuff, love it!

[@ArneChristianRosenfeldt]

Why phosphor? I did not notice phosporesence on CRT . High speed cameras show that it lasts not even a scanline.

[@Wyatt_James]

This technology already exists. Look up "laser phosphor display."

[@swedneck]

i was thinking this as well, it honestly feels like this ought to be a very viable product to bring to market! As far as i can figure it should let you make incredibly cheap monitors at the cost of them being way thicker than LED ones and they could only be about this size or larger. Also it ought to be possible to just keep adding lasers, thus enabling waaaaaaay more advanced vector graphics!

[@TheRenegade...]

​​​​​​​​@@ArneChristianRosenfeldtPhosphors are what turn the electrons from a cathode ray into green, amber, or white photons on a monochrome CRT or red, green, and blue photons on a color CRT. While phosphors can be phosphorescent, fluorescent phosphors are what's used in CRTs. Either the original commenter probably confused "phosphor" with "phosphorescent", or they're referring to the high level of persistence this specific device has. A Laser Phosphor Display (LPD) is a real thing though.

[@James1095]

It's a neat idea and a fun project but what you're really looking for is called a closed loop galvanometer. You can buy inexpensive ones from China specifically for laser light shows or if you are ambitious you can build your own. A very talented person named Elm Chan published a product quite a few years ago to build these and the special amplifier for driving them from scratch. There were also a few high end prototype laser raster displays back in the 80s that used a rotating polygon mirror for the horizontal deflection with a galvo for the vertical deflection, I never saw one in person but apparently the image was impressive for the time.

[@perz1val]

Did you notice that the red laser actually works as an eraser? Paint with blue/violet (405nm), erase with red

[@mal2ksc]

Maybe it would be possible to use a red or infrared LED to flash the entire inside of the screen when you WANT the persistence to stop.

[@BRUXXUS]

[@R.B.]

@@perz1val that doesn't seem right. The higher energy violet/UV laser works because it is elevating the energy level of the phosphor, which then emits a lower energy photon which we see when it decays. The red laser doesn't have enough energy to elevate the phosphor, but it almost certainly doesn't have energy to somehow lower the energy level either. The only decay mechanism I'm aware of is time.

[@perz1val]

@@R.B. see, the red laser brings just enough energy to make the phosphorus emit all of it's stored energy at once, but it does not have enough to recharge it. I've played with it and it was almost impossible to notice the phosphorus discharging everything after red laser hit it, because the beam was so much brighter. It just looks like erasing. Also blue/violet is better for charging the phosphorus than UV due to how the electron energy levels work. I guess the red bumps it a bit higher and after the electrons collapse emitting all of their light, because I doubt it being heat related. Warm phosphorus glows stronger (discharge faster), cold glow dimmer but longer, however the tiny red laser I was using was incapable of really heating anything up.

[@Slowly_Going_Mad]

​@@R.B.it's a real and documented thing. Yes you can erase phosphorescence from things with intense long wavelength light. Styropyro has a video on it.

[@samj1012]

Great project. The most smart part is the signal calculations to produce a picture. Tip: you can harvest a high speed rotating mirror/motor from old scanners, or video projectors.

[@EnTilSeksten]

14:04 average vim user

[@bashkirtsevich]

Ahaha, exit vi/vim through pressing reset button

[@kairysisKrantas]

This video is not just sponsored. It is APPROVED by Brilliant. This is brilliant.

[@OptiVR]

BUT... Can it play doom?

[@xephael3485]

Crysis

[@plapix]

yes!

[@abdalsalam6682]

Yes, one frame at a time

[@EWOS412]

@@abdalsalam6682(for a big change)

[@OptiVR]

@@abdalsalam6682 I believe that's how all computers work yep

[@spaceminions]

I had an idea: zinc glow in the dark paint can be quenched by shining red or infrared on it, so you could flash it from the back right before you draw a new screen and have the remnance without the ghosting.

[@BESTvsWORST-vx2dg]

placing a sponsor right after showing how you used chat GPT is like winking while the sponsor read

[@cutterhead13]

Ha using the glow in the dark filament like that is brilliant ! Love it

[@lualdiz]

Interesting, earlier today I was thinking if CRT TVs could be made thinner and lighter by ditching the magnetic deflection part of the electron beam altogether and instead using electron mirrors and/or electron lenses that can reach a larger screen area with little movement, given this demo uses a laser beam, but it's an interesting proof-of-concept, and I would love to see big home CRTs brought back again in an enhanced form through an insight like this, other thing I thought about is using some sort of electron phased array of sorts.

[@andygozzo72]

this has been done, google 'sinclair ftv1', but it never took of as LCDs developed fast

[@ArneChristianRosenfeldt]

TV were quite thin. A big problem is the actual gun.

[@andygozzo72]

its already been done, a gabor tube, google sinclair 'ftv 1'

[@Kim-e4g4w]

On the horizontal axis you should use a DC motor with a hexagon-nut spinning. If the sides on the nut is like mirror then you can spin it really fast. This was done in old laser photocopiers, not sure if todays models are using this method.

[@ArneChristianRosenfeldt]

Some are, others use The Nintendo virtua Boy. But it can only do Red.

[@MarinusMakesStuff]

Interesting, I also use a UV lamp to light up the glow in the dark stickers that we put up on the ceiling above the bed of the little one :) It has a moon, planets, stars, etc. Works like magic!

[@listerdave1240]

For faster response I would suggest using the 'voice coil' mechanism that moves the heads of a hard disk. Just take apart an old HD. Remove the platter and mount the mirror on the head arm. You can get some insanely fast scanning. You just need to apply current to the coil. It takes some work to get it to move in a predictable way as the current through the coil translates to a combination of speed and acceleration but it shouldn't be too difficult to map out.

[@MichaelMathews-q2k]

I can the frustration in getting it to work in a satisfactory way.

[@apathtrampledbydeer8446]

Beautiful work, especially your drawing skills. Never give them up, never let them down!

[@teamllr3137]

Thanks for the knowledge about the uv sensitivity of bambu glow pla

[@b0rd3n]

You are the best coder of the whole town! Never give it up

[@remotepinecone]

try a really fast spinning mirror like in an old laser printer. Now a really fast linear actuator like thing to make the laser go up down fast (not sure if printers had this). Then, time it all to the actuator which is your vert scanning speed.

[@codebeat4192]

Really nice and fun project! To improve the speed you need to use coils and magnets to move the mirrors. Better is to use just one mirror. I believe the same method is used in a laser projection unit that can draw shapes on a wall.

[@TheAgentGray]

The restraint showed in the gravitational pull project cutaway was perfect I legit double taked when I realized there was a black hole there

[@why6212]

I love the idea of a 1960s family sitting down and equipping their safety glasses to watch their laser tv 😂

[@RinoaL]

1:30 I like the gremlin expression lol

[@KafshakTashtak]

For the mirror, you could use a flexure mechanism, with a piezo actuator. That would make it faster to rotate, but the range is very small. So, you gotta amplify that range.

[@insightfulgarbage]

glad youtube recommended this vid, this and the short preview of the mechanical tv earned my sub 👍

[@mal2ksc]

I think I can help a tiny bit with the speed. The laser is not using much of the first mirror, and most of it could be removed to cut down on mass. You're probably only using a region a few millimeters tall and maybe 1 cm wide. It's only the second mirror that has to deal with a beam that is already being swept. Also, if you were to sweep lines in alternate directions like a printer, you would eliminate the "flyback" motion that probably keeps you awake at night, making it a lot easier to do raster scanning. ILDA supports bidirectional scanning for exactly this reason.

[@vanceblosser2155]

Years ago I bought a used laser and power supply at a ham fest that had been removed from a broken laserdisc player. The laser was a tube but the way they moved the beam was with voice coils and mirrors similar to the speakers you used but to get more amplitude the mirrors were hinged on one side and a stiff wire from the voice coil attached to the mirror about halfway between the hinge and the other edge. This made the mirror deflect more than the voice coil moved. Of course it doesn't matter that the laser wasn't a diode, I was just trying to give you an idea of how old it was.

[@rizen3467]

I never knew there were square nuts, these seem so practical for 3d printed stuff like you did there!

[@robertgast5953]

This is really cool. It would be awesome at a larger scale if paired up with a DIY vintage computer made from TTL or 6502/z80 etc.. The biggest issue with using it as a simple display is the beam width, are the traces so thick due to the laser culmination or is it just a property of the PLA screen diffusing to much light? It could defiantly be sped up to trace faster with some BLDC based servos.

[@paulbunyangonewild7596]

Well for one, twisting the wires makes it abundantly clear which ones belong together, and they're usually going to a similar place anyways. And depending on the type of signal being read, it can actually negate interference.

[@Dazdigo]

You can have the horizontal mirror just connected to a spinning motor and then pulse the laser for the pixels. The vertical mirror can then be servo based to control which horizontal line is being drawn.

[@malvoliosf]

You might want to make a raster display, so the two mirrors spin continuously, and cover the screen left-to-right, then top to bottom, and turn the laser off and on when it’s pointed in the right direction. You would probably want double-sided mirrors, so you get a longer duty-cycle.

[@bo-xmusic6940]

This is a really cool project. I was thinking maybe you could improve the rasterizing mechanism by replacing the mirrors with two lenses, one for the vertical field and another for the horizontal, each connected to a motor. The motors would change the angle of the lens relative to the laser and thus change its angle.

[@verebellus]

the fact that CRT tv technology is so old is just amazing to me

[@adamp3988]

That's a sick DIY vector display. I wonder if you could "emulate" a CRT more passively by just having the laser move such that it scrubs across the screen left-to-right and then top-to-bottom. If it can do that fast enough, you could then use the laser pulsing to control the actual image instead of trying to wring a ton of accuracy out of your stepper motors. If you use a full 180-degree rotation, you could even technically have a "v-blank" period where the laser is swinging back around to the top of the image.

[@sunday87]

Wow. This is a great Idea. First I thought: just a rear projection screen with laser projector. But the persistence of vision through the flourescent screen is much closer to classical CRT behavior. Next step: Use an actual screen from an old crt together with the phosphors and the matte. use three lasers for red green and blue just like the three electron guns in the CRT. Use old school analogue video signal as input and render properly the interlaced fields in real time. Perfect CRT replacement for retro gaming.

[@alexanderbeliaev5244]

There is no need to bounce mirrors back and forth, they can be simply rotated + synchronization circuit is needed for the laser.