Mechanically Multiplexed Marble Selector (Electromagnetic) - Marble Clock Seconds Display Pt.2

[@thermidorthelobster4645]

If there’s one thing KZbin has taught us, it’s that you can’t just make a marble machine and expect it to work. It’s a lifetime commitment.

[@helvettefaensatan]

I'm having flashbacks. Do anybody know how Martin is doing?

[@chucklebutt4470]

Lol I hope you like dreaming about marbles!

[@dragontriketv]

You know that this will make you as famous as Martin.

[@dragontriketv]

@@helvettefaensatan He is reassessing his life choices 😏

[@roderik1990]

@@helvettefaensatanJust stuck in the ever increasing scope creep.

[@armoredtitan54]

So glad you found a better way to turn the balls black. That's gotta be such a relief

[@V0S1N0]

I was bouncing off the walls seeing a single ball being done one at a time with a propane torch. How in the heck... ?!

[@MadScientist267]

Protip: It attacks the stainless because stainless is primarily iron and chrome, the latter being the metal being eaten off of the marbles.

[@billkeithchannel]

He is kicking himself in the groin that he did not figure this out earlier.

[@DrKlausTrophobie]

Marbles with very different surfaces. I'm waiting for the things to come... 😁

[@brandyraccoon1473]

@@DrKlausTrophobieAre you thinking the color difference is causing of the giant cascade of new problems near the end? He was testing on only one color before all that.

[@MegaTechnomaster]

Came for the marble clock, stayed for Ivan's slow descent into madness

[@lifeprojects1]

Feel his pain 8:04

[@JaceDraccus]

One second at a time.

[@delphicdescant]

Wintergaten Vietnam Flashbacks Intensify

[@PatrickGunderson]

Ivan made a marble machine music maker in one episode

[@andy4an]

at least he eventually came up with a desk barrier to stop the marbles! that was cathartic

[@AaA-ru6rr]

Eres más español que los toros, porque no haces algún vídeo en tu idioma? A los españoles nos gustaría 😉

[@AaA-ru6rr]

Puedes probar a usar imanes para la recolección de canicas.

[@KaneOtonashi]

Ah . i knew I would find some fellow Marble Fans

[@davidchidester5463]

KZbin has taught me never to build a marble machine because it will drive me completely insane. Lol.

[@sealpiercing8476]

It may or may not be possible to build a marble machine without going insane, but as soon as you try to make it more reliable, for every problem you fix you will find two more and eventually even the stuff that initially worked will break in new and interesting ways.

[@Duskydog419]

😂 I Wholeheartedly agree

[@DrKlausTrophobie]

It's fun! As long as you are watching from a far distance someone else going bananas... 🤪

[@AceKaci]

Would you say it’d make you lose your marbles?

[@sealpiercing8476]

@@AceKaci Losing your marbles is the price of entry.

[@lwo7736]

That troubleshooting at the end really really made me feel good about the "Everything was working, now nothing is" feeling I get with most of my projects haha

[@EEEEEEEE]

E‎‎‎‎‎‎‎

[@TechnoidProduction]

I love that this project embodies the idea of "Make, Fail, Repeat". Love to see the rapid iterations!

[@EEEEEEEE]

E‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎

[@SiblingCreature]

So what I'm wondering... What's the difference between the diameter of the black balls and the diameter of the chrome balls, and how much rougher are they after they have been treated?

[@ledocteur7701]

surface treatments such has used here only penetrate the first few microns of the surface (100 microns is the width of a human hair), roughness could be an issue but diameter shouldn't be at all.

[@Pystro]

I would have tested it with *all back* marbles before doing the test with mixed marbles. That would make sure that *everything* works as reliably as the silver ones and cuts out the sorting step in case it doesn't.

[@SuperCuriousFox]

@@ledocteur7701Wikipedia’s page on Hair’s Breadth has a source stating that human hair ranges from 17 to 181 micron in diameter. For fine machinery, dimensional tolerances can be below 20 micron.

[@PKMartin]

5 microns is the width of a human hair? Try 100 microns. You might be thinking of thou - an average hair is about 4 or 5 thousandths of an inch thick

[@ledocteur7701]

@@PKMartin Just checked on Wikipedia, and.. well now I'm wondering where I got that 5 micron information from, cause as an industrial product designer, I was pretty dang confident about that information. And I definitely wasn't thinking of thou, since I'm European.

[@pokerpig9069]

I couldn’t care less about a marble-based clock. But I’m riveted to these videos because I learn so much about the challenges of design and manufacturing. I have a whole new level of appreciation for it. And the designer!

[@Aussiesnrg]

It's amazing the level of trial and error that is required to make... basically anything

[@SmilingDevil]

As a designer I cannot express enough of how much I appreciate that leveling insight you have reached

[@karthikeyajavangula8423]

I'm getting dangerously close to building this.

[@joshuavincent7884]

🧢

[@DonsArtnGames]

It's too late for me... I've already started 😅

[@jamescollier3]

I'm going to build a shed

[@Fanatic4500]

Ivan crawled so we could walk. Also, if my wife ever leaves me I think "Babe I want to try to make this" will be high up on the list of grounds for divorce.

[@machieltipo]

@@joshuavincent7884definitely a guy who doesn’t have confidence in himself 😂 If you did you would’ve logically analyzed the fact that Ivan is a person who also has the responsibility of filming this process which in turn makes the project take twice as long, and a person without that responsibility could possibly be more productive. Also, he has the advantage of seeing Ivan’s mistakes, so no, it’s not cap that he could make one. It’s just a matter of persistence. Even an idiot can build an empire with enough persistence, never forget it.

[@Simple_But_Expensive]

Perhaps you could create a ball sorter based on reflecting a light off the balls to a phototransistor. The chrome balls would reflect more light, activating the transistor. The black balls would not reflect enough light to activate the transistor. The output of the transistor could be used to activate a mechanical gate. The gate would direct the balls to one of two tracks, one for directing chrome balls, and one for directing black balls. I have seen a similar setup used to sort balls on a Lego ball machine.

[@lifefindsaway7875]

That’s how the first version of the marble clock worked. He just needs to transfer that mechanism to a sorting gizmo

[@johnfunk4313]

I have a few suggestions. 1. put a "fence" around the work table to contain the marbles, and put a little slant so they all go to one corner. 2. put a bucket there that feeds a sorting device that can divide into light and dark. 3. set to run whenever a marble is there. That way you will always have buckets of the right marbles when you need them.

[@GingerNingerGames]

Nothing like a project to solve the problem created by working on your project. Not discrediting the idea, it's exactly what he should do, and what I'd do.

[@aryst0krat]

the issue is that if everything works the first time, or even the tenth time, you've probably wasted more time than you've saved - especially given this would require some kind of electronics/software and you always want to believe it'll work before it becomes worth it to do something like that 😅 I think the rope fence is a decent compromise

[@GingerNingerGames]

@@aryst0kratI mean, he's going to have to make it eventually, he's going to have to sort those balls after they get ejected from the seconds display, so might as well do it now and incorporate it into the desk so he can be testing two functions at the same time, and saving himself a lot of time at the same time. If one or two are wrong after it's sorted 100, that's a lot of time saved

[@johndorian4078]

every solution introduces it's own problems............

[@jarkkoaitti287]

from the fine game of pool, i learned some things about rolling ball physics and energy transfer. when a ball in a newton's cradle bumb into another, it stops. in a pool table (straight shots), it is depended of cue ball rotation. normally without any up or down spin, a slow or moderate hit speed, cue ball tends to pick up a forward rolling motion from table (instead of sliding). the target ball still bounces off with the same speed as the cue ball hits it and cue ball stops, but just for a moment, after which it picks up forward speed again from the rolling motion. in a marble track, you can add to this rotational energy for example by making the track so narrow that marble contact points are on sides near rolling axle, making it spin much faster. that spin is hard to see in a smooth ball, so maybe it could be used to make a magical ramp where balls move uphill :)

[@Ice_Karma]

I'm a software engineer. I can cope with the insane issues that come up sometimes while developing software. The physical world is _way_ too squidgy and imprecise and weird, and it regularly drives me crazy and defeats me. I don't know how you do it! 😻

[@musicalracket]

I'm confused why you are cycling all the marbles at the same time and not just the ones that need to change the color

[@RonParker]

You need a magnetic floor sweeper. It's basically a magnetic bar with wheels on the ends, and a long handle in the center you can use to push it around like a broom. Your back will thank you. Though you might also need to demagnetize the marbles after picking them up.

[@stevebabiak6997]

An old CRT degaussing coil might just be appropriate for that other task.

[@DriftEngineer]

You should use V-channels. The marbles will run much more reliably than on flat channels. On flat channels, the marbles tend to spin while rolling.

[@_adamalfath]

Start a side project and build an automatic marble sorter :)

[@RealGrouchy]

Agreed. Won't he need one anyway once this part is done?

[@NiksSofa]

Sad to see another talented youtuber succumb to the curse of the marble machines. RIP Wintergatan, and now RIP Ivan. :)

[@makers_lab]

I don't think so. Martin, bless him, was doomed from the start, but Ivan's getting there.

[@rainmannoodles]

I'm starting to get some MMX vibes with this "seconds display" - it's complex, requires precise adjustment, and may never be completely reliable. It may make sense to think about a redesign. Here's a rough description of where I would start. Consider having two marble layers at the face. The front of the display is A (current) and then B (next digit) is staged while the display is static. Until the swap, layer A is retained by a frame that positively locks it into place, and layer B is resting against layer A. To eject layer A, an actuator would shift the frame (up/down/sideways, doesn't really matter) which would have a track shape that would forcibly eject layer A, while simultaneously shifting layer B forward partway, trapping it in place. The return stroke of the frame would then shift layer B the rest of the way in, locking it into place at the front of the frame. Effectively the mechanism would "step" the marbles forward in a controlled manner by moving part of the frame. Depending on the angles and friction, it would also be possible to use two actuators and two frame pieces to make it more robust. I haven't quite worked out the ideal geometry for the single actuator case (it's theoretically simpler) but the geometry is also trickier.

[@lifefindsaway7875]

I like this. Abandoning the ‘Newton’s Cradle’ ejection method would avoid a lot of uncertainty

[@vctrsigma]

A "doublebuffer" approach to rendering? IMO It's a lot less satisfying if the marbles are pushed by an actuator at any point. Get 10 buffers and you could mount them all on a giant wheel like a flip clock. Gravity and inertia should always be the primary way that marbels move (with a lift to "recharge" them).

[@2009dudeman]

This is probably the best way. Kind of like the gating he already has at the top, just put another at the bottom only simpler as you describe it's just a single axis frame for all lanes. You still have the top to sort them light and dark, now you just have one at the bottom for A and B.

[@evanackroyd6833]

I’ve never commented on videos so I don’t think I successfully “@“ you in a comment I added to my recent post, but if you feel like scrolling through to find mine I feel like there might be some similarity in our thought process

[@KptnAutismus]

i love how multi-material parts look, i hope this kind of printer becomes more widespread.

[@mrnlce7939]

You want/need to make a marble sorting machine also several magnets on a stick for picking them up off the floor. Just thinking of your poor back. Great video. Keep up the good work.

[@RealGrouchy]

He'd want to keep magnets far away from those marbles, lest they get magnetized and stick to each other

[@thingswelike]

Even Martin conceded in the end that magnets and marbles don't mix. Everyone told him from day one, but he persevered anyway...

[@brunomcleod]

6:00 I don’t know why but something about that company seems so wholesome and good like if I needed a PCB prototype for some reason I would buy it from them

[@enyoc3d]

"Laziness is the mother of invention." Loving this series. Your iterative approach to problem solving is fascinating.

[@nadukkerensky392]

"i am making a marble clock so i built a marble sorting machine that identifies colours" - Ivan Miranda "i am making so much work for myself by having to manually sort so many marbles" -Ivan Miranda o.O

[@JokerStudios69]

2:57 "I remember when we were kids we used this to make bom-"

[@nowadays5826]

The whole thing can possibly be a bit simplified using the fact that you Do not have to replace all marbles every second. It is enough to replace those that need to change color. That itself will only reduce the number of marbels used. The big advantage comes, by loading alternating marbels into the feeding channel: You no longer need two separate feeding channels (black and white) plus you also no longer need a marble selector! That 1) saves room and material (half the feeding channels) 2) requires half the solanoids (no more marble selection, just the release) 3) uses fewer marbles (less of a mess, more time to reload) Hope this helps!

[@sebbes333]

*@Ivan Miranda* 8:35 The obvious solution is to print each layer above some millimeters further forward, stagger each layer.

[@cameronmcdaniel105]

You can feel how close this man is to greatest and insanity at the same time. Must watch on the edge of my seat. Glad you are always improving the loading and manufacturing processes.

[@eestaashottentotti2242]

2-colored magnetic marbles. Electromagnets behind. Switch polarity, if needed, every second. No lost marbles. A marble fountain behind for assurance.

[@ArrakisMusicOfficial]

"After uploading the last video about the marble clock that can update every second, I realized that I didn't explain how it works" * proceeds to not explain how it works * 😂🤦‍♂

[@tin2001]

Yes, but he now is aware he didn't... He's taking the channel in a whole new direction of not explaining things. 😂

[@HenryLoenwind]

I mean ... it doesn't work, so there's nothing to explain? ;)

[@stevebabiak6997]

@@HenryLoenwind - correct, until it works any explanation is potentially incorrect

[@marsrover001]

Just remember, lots of us could badly model something in fusion360. But it takes a brilliant and experienced person to even attempt such a complex project. Take as many videos as you need to make it happen, it's all part of the learning process.

[@chromosundrift]

So many things to love about you but the best must be how you demonstrate the super power of PERSISTENCE! Most outcomes are driven by skill and luck but persistence brings ever more of both.

[@Repkord]

Clearly pimping it out with racing stripes and proper labeling makes the whole thing at least 42% better. Well done. 👏

[@Timeyy]

It seems like marble machines are really the 'final boss' of engineering lol, great video!

[@TheCardq]

I love the custom heatsinks, that being said, I'm curious if you did any extra finishing work on the contact surface like smoothing the surface or using a thermal couple like thermal paste or a thermal pad.

[@leosbagoftricks3732]

An electrical solution to the solenoid heating problem would be much easier to implement. Solenoids can be held-in with a fraction of the current required for the initial pull-in, there are many ways to accomplish this. The power dissipation can be dramatically reduced, with no performance sacrifice.

[@Induku]

marble machines are the dark path to madness

[@Ice_Karma]

I'm a little tiny bit concerned about what you did with the dissolved chrome, although I can't imagine that the small quantity would be much of a problem. Your adventures in clock-making are insane and I love them. I'd suggest watching Wintergatan's adventures with steel ball bearings, because I seem to recall he faced significant issues due to them gradually getting magnetized. I'd also like to amplify the comments about the balls picking up spin, leading to erratic momentum transfer.

[@claimeant7098]

I'ts yeeears since i watch your videos, and one thing I can say is "How patient you are !!!" Thanks for your teaching, always with the smile 😁

[@oftenbetterthanone]

The marble clock series is fascinating! Me encanta! One small suggestion to save you time... Can you invent a quick marble sorter that will read the color of a marble and dump it into a silver marble bucket or a black marble bucket? This way you can dump all the marbles into a hopper and don't have to do it manually (and it will also make a cool segment for a video!)

[@Vendavalez]

Little advice that I heard on the Wintergattan chanel. If you want to save your back from picking all those marbles off of the floor, get a stick and attach a magnet to the end. That way you don't have to keep bending over.

[@stevebabiak6997]

Or use a dog poop scooper

[@DragonBenessere]

Hi Ivan, I produce steel marbles and I have some professional advise to share with you. First of all, I’m Italian so please forgive eventual misspelling. Then, I haven’t no knowledge on electronics but I work in a factory and I produce steel marbles so I can suggest you something. Sorry in advance for the length. We produce various tipe of quality marbles, depending on the costumer request, and the 2 most important traits marking quality are the geometrical form (perfection) and the surface smoothness. The most perfect form and the most smoothness means higher quality, meaning the marble will roll and will bounce easier; that way the black marbles you made are forced to “lose quality” so will roll and bounce worse then the silver one. Also, the more shining is a marble, the more smooth is the surface, BUT it is not necessarily a sign for a better geometry perfection (this can be checked with expensive tools I use at work). Keep in mind that every time a marble bounces from a 20-30cm height it will be slightly “deformed”, you can’t see it only with your eyes but you can with a microscope. What I mean is that if apparently some marble are identical in aspect, they still may roll and bounce differently. So if with your many tries you still notice some marble acting strange, try to keep them apart. And be aware that black marbles will work slightly different (worse) then the silver ones. I could have some other advice but for now, I already wrote a lot, I hope you’ll read those tips and I’m at your disposal for further requests. Thanks for your contents!

[@ludditzthrell]

I find it kind of hilarious how this mirrors my own way of doing things, though not on such a huge scale. For a certain game I love to play (Shadows of Brimstone; I know, to anyone who follows my comments, I NEVER SHUT UP ABOUT IT!) I've been making small quality of life improvements. Starting small with sorting trays, card trays, mostly getting rid of the original boxes to make room for more stuff. Honestly, it's become more of a lifestyle than a game. Fix one problem, six more crop up.. just.. damn. Also, I love it: laziness is designing a new device to make a task easier. You put in work to be lazy. You.. I can't.. I can't tell if you're a mad genius or if you're fueled by the desire to be lazy while not knowing how, fully. Love your stuff, Ivan, and I especially love watching the madness that is this clock of pure evil and pain.

[@urboyluke_]

Ivan, I owe you a huge thanks. The orchestral song you played at the start of the video, I heard about 6 years ago. I loved it but I was never able to find it, until today. So thank you lol. Cool machine btw :D

[@mazack00]

This saga has been thoroughly enjoyable to follow. Crazy how they got stuck.. heat soak deforming the plastic?

[@timothybuchser]

This all seems very unnecessary as I notice you have a watch.

[@xvii_au]

But it doesn't tell marble time.

[@adamoconnor8958]

He has done the reverse of losing his marbles.

[@SmilingDevil]

Well… you know the definition of a „Hobby“ don‘t you…?

[@timehunter9467]

@@SmilingDevilYou know the definition of a “joke” don’t you?

[@EEEEEEEE]

E‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎‎

[@Carthybp]

This clock is the best fucking journey on the internet.

[@fabo1593]

Absolutley amazing work! I have always loved watching this series and the design iterations. I know I'm might get some pushbackfor this in the 3D printing community, but you should consider printing the entire clock with a different plastic other than PLA. It is difficult to reliablly print in another plastic other than PLA, a heated enclosure is a MUST, however the benefits of a higher temperature plastic are worth the pain. In my opinion it's a shame that the entire world of consumer 3D printers is centered around and optimized for such an impratical material as PLA. Even PETG would be a better option and your printer can definetley do that. Keep up the great work :)

[@inviz88]

I'm so confused why you just don't raise the machine and have the marbles fall back into the bucket? lol

[@tin2001]

Which might have solved the rebounding double ejection problem too - they were bouncing back off balls already on the floor. Extra falling height would have solved that.

[@DjoshDePosh]

yesssss! part 2!!!! Love your videos!

[@GaymerJenn]

I know, I clicked so fast 😂

[@polo141088]

Your resilience to “failures” is inspirational. Like Edison and the lightbulb 😅. Can’t wait for next video, thank you !

[@swilson42]

Love this project! One suggestion on the exit though. Relying on the inertia of a new ball to push an old one out is a novel idea, but seems to be way more sensitive than it’s worth. I think you should switch to a gate that stops the balls from falling out that is opened to let old ones fall and then closes to catch the new ones coming in. Don’t try to time it with electronics either, let the new balls coming in trip a lever that spits the old balls out of the way by releasing the gate and puts the gate back so it’s back in place when the new ball gets to the end of the lane. That way, even if the balls aren’t rolling just the right speed, it won’t matter. Can’t wait to see this work reliably! Also, for the love of all that is holy, put that thing on a box and let the balls fall into the tub! Stop chasing them all over the table and floor! 😂😂🙃

[@anjarwicaksono229]

Suggestion for you: First, no need to replace all marbles in 5x3 matrix. Just do the diff. Transitioning from 0->1: 7 balls replaced from 1->2: 8 balls replaced from 2->3: 2 balls replaced from 4->5: 5 balls replaced from 5->6: 1 ball replaced from 6->7: 9 balls replaced from 7->8: 6 balls replaced from 8->9: 3 balls replaced rom 9->0: 2 balls replaced from 5->0: 3 balls replaced Second, Since you're using metallic ball, could it be possible to control the flow of a ball is by using a temporary magnet from a coil at the gutter. It can be controlled individually per pixel. I also believe that since it's less mechanical, it should be more compact. To avoid magnetic interference in between the gutters, you can find any good magnetic insulator. Btw you only need 13 coils since the grid [1,1] and [1,4] ([0,0] is the topleft corner) are always black. Third: in case you want to have hh:mm:ss format, and that's gonna be 6x13 = 78 it's more convenient to sort black and silver marble by the time they fall to the sink.

[@LowTechFarm]

I'm so glad I happened across this project. It's maddening, and when I have to make a thing at work that is a one off to fix a problem that doesn't occur in nature, I feel less alone in the land of solving an improbable puzzle.

[@Remington53]

I wonder if the sudden issues are caused by introducing black marbles. It looks like the acid-finished surface of the black marbles is way rougher than the chromed ones, which probably introduces a lot of friction that wasn't there when you were testing with only chromed marbles.

[@ivanmirandawastaken]

That's where I'd put my bet too

[@BradHouser]

@ivanmirandawastaken: Here's an idea: How about no Black Balls? You really don't need them. The plastic background is black enough. (0:31 is a good example.) Create a release mechanism for each ball. For each second, only change pixels that need to change. Three things can happen for each pixel: (1.) Do nothing, (2.) Release the ball, or (3.) Add a ball. Benefits: No need to sort by color, fewer balls needed overall, release is more orderly. Drawback: a new release mechanism is needed for the display marbles that doesn't destroy the looks. If adding individual releases is too complex or bulky, then a sliding release for all marbles could be used, then the next second all new marbles (and blanks!) will be dropped in.

[@oldmanbluntz]

Your videos are always the inspiration I need to keep me pushing past those issues in a design, be it hardware or code, that drive some people bonkers to the point of quitting. Thank you for that inspiration.

[@ianmontgomery2862]

was just thinking about your saga! Can't wait to see more!

[@murdokiwi]

On the earlier video, I thought "why not load the digit patterns on a constrained track?" - no sorting required. But love the overengineering of this misadventure

[@TSGilly]

I'm not sure if someone has said this before but why change every marble, every time? You could save a lot of marbles by only releasing a marble when a change is warranted. For example, when a marble is silver and needs to be changed to black or vice versa. If it is silver and the next digit calls for the next marble to be silver in that same lane, simply don't release a marble. This would save you a tremendous amount of loading and could be accomplished with only programming changes. I would think you could build a simple device to sort marbles into 2 different chutes to be recirculated back to the appropriate channel using a photo eye device that can detect the difference between the 2 colors. Maybe something where it starts loading lanes from the top and when that lane is full, the marbles are directed to the next lower level.

[@tin2001]

One reason to replace all pixels each time is to keep the sound consistent. It will be the same sound each second, instead of sometimes being loud and sometimes being quiet. Replacing the entire digit each time also means if one pixel is incorrect for some reason, it won't be stuck wrong for multiple cycles. I'm pretty sure the way the colour selection currently works also requires it to drop a ball in each slot. So to make it only do changed colours would require a complete redesign of the ball selector mechanism.

[@pedente_ente]

Most enjoyable Content on YT rn. Keep it going. p.s. you should build an marbel sort machine

[@dragonis003]

Now that you have switched from glass to metal, you really should make an electro-magnet on a broom to help pick those up off the floor, save your back

[@okfj]

Two things. "The power of lazy" should be on a shirt.. :) The power of lazy should design a marble color sorter so you don't have to manually sort the marbles. I'm sure the sorter would be worthy its own video.

[@feuby8480]

This is what internate was made for. I love you dude, your commitment to this useless thing is what made me suscribe. Especially since you show us the fails.

[@methodermis]

I love your extremely complicated and LOUD clock! Thanks for posting these progress vids, this is a fun project to follow. You do a great job editing and presenting your progress!

[@Hiro_Trevelyan]

I'm glad I found your channel. You have this "insane engineer" vibe, like "crazy scientist" but in a fun, informative way. Love your energy !

[@HeadlessOne]

I am really enjoying the progress on this, i think a great idea would be to have a quick way to tilt your table(even just lift one end by hand) and have a collection funnel at the other end so all the marbles collect themselves off the table i think it will save you alot of time in the long run.

[@jasonkrise495]

I wonder if adding a small disk magnet to the end of each channel, from the bottom, would be just enough hole to grab the second marble and prevent it from popping off. You could then increase the travel distance again to get some more speed to make sure the first falls of the track. Getting the right magnet strength could be an issue, but possible a simpler / easier one than what you have now????

[@esser50k]

this feels much more like my project developments. 1 full month of non-stop work to just to end up where you started haha you're inspiring me to show more of my failures along the way

[@AquaMarineFBVA]

Those black marbles are not the same size as the silver ones... idk if its paint or anodized but they look to be slightly bigger and i am almost certain thats why everything is sticking now when it worked with just the silver marbles... But Keep at it!

[@629Justme]

After this fun and super complicated version, mk3/4 could be made using prepared drops with just the numbers needed. Like an analog 7 segment display, drop prepared marbles relating to the ten digits. Second 1, drop the marbles for one, two the same and on until zero where the preparation starts all over. Might be even faster since the sort is already done. Idea?

[@greenbean7355]

11:04 Ivan: Why is no work? The Marble Clock: I gotta poop

[@TheRealAlpha2]

Portrait of a man losing his marbles... but he's still brilliant enough to know where they all are at any given moment.

[@DragonsAndDragons777]

Your dedication to this project is wonderful

[@kevtris]

you can use a resistor in series with the solenoids to keep them pulled in but not draw so much current, to reduce the heating of same. you still need the full current to pull it in, but once it's pulled in, the current demands are much less. this does increase the complexity a little of the driver, though.

[@JoelSapp]

What a glutton for punishment, and always with a smile!

[@Sciman101]

At this point I feel like you need to make a separate marble sorting machine

[@lady_draguliana784]

Yay! turning the channels perpendicular was one of my first suggestions! (not that you even saw it, just stoked I got it right!)

[@dennisdecoene]

I cannot understand why I find this project the most interesting thing since sliced bread. I can't look away!

[@teardowndan5364]

Instead of kinetics to bump marbles off, I'd go with clocking gates to let the queue move up one position at a time and use the chute to store prepared digits. Instead of each marble position needing its own black/chrome queue, the sorting system would directly send marbles directly to the next queue position that needs the color or into a stack channel if no position requires that color at the time.

[@dspaniard]

Your patience is astounding!

[@javacipher]

You could also print in ABS instead of PLA, which has a much higher temperature resistance (around 100°C or so) and strength. As others have already mentioned, the mechanism seems to work fine with the silver balls, indicating that something could be different about the black ones.

[@jangriesel5056]

I dont know how you stay so calm

[@pip360]

So excited for this!!!! Can't believe I managed to catch it within the first couple hours of release! I love your videos!

[@So.Me.]

I'm quite sure you are one of the most positively freaky guy on the world! Your films are like hitchcock: in the beginnig there is an earthquake, and then the tension increases. I love you and keep my fingers crossed for your marble clock! 😍😍😍

[@finndo]

side projects: - convert a roomba into a marble-on-the-floor-begone - marble sorting machine (takes a bucket of mixed marbles in, color-sorts & outputs into 2 buckets)

[@gdipaolo]

The amount of work you put in this project is insane, keep on !

[@robinmaurer2645]

You could also use some thermalpaste between the solenoids and the aluminum heatsinks

[@JoeMakes]

So glad you are making "science experiments" 😂

[@yosemitebandit]

I hope this series never ends!

[@PTEC3D]

Ivan - is there any reason you can't have rotating cams placed one marble space back from the edge, all the cams geared and turning together? That way you can just send the correct arrays of marbles down until the ramps are full, then let the cam push the old set of marbles off with the next digit, and you'll know that a) the new digit will get placed at the exact same time, b) the old digit marbles will all get pushed off by the same amount of force, and c) you could make a taller stop ridge right at the end because the cam arrangement would be strong enough to push the old digit marbles off. The main advantage would be that you'll have precise control over the display. Just a thought...

[@thelefthandedshooter381]

The most patient man in the world!

[@gccchc2449]

Ivan is my all time best Crazy Inventer.

[@DavidJohnsonFromSeattle]

This video is the best demonstration I have for why I leaned away from experimental physics and into theoretical...

[@vwlssnvwls3262]

I am always excited when I see a new chapter to the marble clock story! 😃

[@MadRS]

I got to say. Learning about how this clock comes together while pissing my pants laughing is a great way to pass the time (pun intended). Your video presentation is first class.

[@Specter0420]

If the black marbles are small enough after the acid, you may be able to sort them physically by size. A little bumper that knocks the white marbles off a track to the lower track, but allows the blacks to pass since they're not big enough to hit the bumper. Just a thought, I know they may not be consistent enough in size so you may need to switch to ball bearing balls or something besides marbles.

[@jorgeserralucas4724]

You can make a marble sorter, maybe it saves some time, so you dont have to think about that 4:00

[@Menown7]

Man I love your humor and dedication to this project! just one small piece of feedback, the first half of the video i wasent sure what was going wrong, maybe point out more clearly what the marbles are doing wrong:)