Which Marble Track Is Best?

22 feb 2021
241 034 visningar

Check out the new SEworld channel Wintergatan 2: seworld.info
After a successful trial run on the second channel we are now moving back into the main channel for the coverage of the Marble Machine X!
Video edited by Hannes Knutsson
Support the Marble Machine X Project:
► www.patreon.com/wintergatan
► teespring.com/stores/wintergatan
► seworld.infojoin
#marble #machine #4K
MUSIC DOWNLOADS ► wintergatan.bandcamp.com
WINTERGATAN RECORDS ► www.wintergatan.net/#/shop
SPOTIFY ► bit.ly/2oKxXWd
ITUNES ► apple.co/2ntWNsZ
MERCH ► teespring.com/stores/wintergatan
SUBSCRIBE ► seworld.info
INSTAGRAM ► instagram.com/wintergatan2000
REDDIT ► www.reddit.com/r/MarbleMachineX/
DISCORD ► discord.gg/wintergatan
Thanks to the Marble Machine X Team 2019
►Alex Füßl | Alex CNC | Munich, Germany
►Carlos Montoro | Anjuda Guitars | Madrid, Spain
►Chris Nadon | Aix-en-Provence France
►Dwayne Myers | Safety | Phoenixville, PA, USA
►Ecco Pierce | All Things Ecco | Oakland, California, USA
►Florian Hu | fhuable | Wellington, NZ
►Jacob LaRocca | Rocket Props | Somerville, Massachusetts
►James Passmore | CAD9 Design LLC | Worcester MA
►Marius Hirn | Munich, Germany
►Nick Householder | Florida, USA
►Philip Brown (solo banjo) Winston Salem, NC USA
►Rainfall projects
►Richard Southall | String HeArt | Devon, UK
►Sebastiaan Jansen | Leuven, Belgium (BE)
►Tim Keller | Inventopia | Davis California USA
►Tim Mitchell | CNC Machinist | Nashville USA
►Tobias Gutmann | Tobias Gutmann Prototyping | Munich, Germany
►Tommaso D'Amico | Vancouver, Canada
►Will Francis | Machine Thinking | San Francisco Bay Area, USA
►Blue Sparkle and her great team for this video's subtitles (and many others).
►Grant Lansdell: Great Idea Provider ;)
►Viktor Stenberg, Calle Guldstrand, Gustaf Törner | ProtoCut
►Liam Ward
►David Lewis
►Tobias Smidebrant
►Erik Holke
►CNC Routerparts
►Nathan Skalsky
►Karin & Olof Eneroth
►Matterhackers - www.matterhackers.com/
►This Old Tony - seworld.info
►Thomas Pilot - seworld.info/tv/JpCQcfOPQghHGcfecxEEVA.html
►Roys Sheds Perth Australia
►Supermagnete.de - www.supermagnete.de/

  • Your test is very different compared to your actual 3D implementation. Friction for sticking depends on the force (grams in your units) and surface area. Your "2D" test designs have smaller contact area than your 3D designs which have contoured profiles which grip with the same force but increased contact area, hence more likely to get stuck. Good luck.

    Jon PrevostJon Prevost45 minuter sedan
  • "I have a small scale here[...]" for small scale experiments? bdum-tsss

    hoiy vinosahoiy vinosa3 timmar sedan
  • It is not Marbe Machine X anymore, It is Meme Machine X now! ;D

    Жора KadrЖора Kadr22 timmar sedan
  • All we have is time, you’re having fun and learning lot’s, your dream is awesome enjoy the journey. You feed Wilson through the nose...😳

    vern evern e23 timmar sedan
  • So if some of the gates tilting backward is causing a problem it sounds like them not being vertical is a problem unless it is 100% required.

    seiom jvonyseiom jvonyDag sedan
    • put them there to Level the machine to a level "standard" before each performance. One Design! One Angle!

      hoiy vinosahoiy vinosa3 timmar sedan
  • New pressure design for marble machine, new haircut for Martin 😁

    Larissa AmurovLarissa AmurovDag sedan
  • Nice. He could have had this z-solution in November 2020. The priciple was posted 8th November last year by Jens#6124 on the Wintergatan Discord channel into the mmx community section.

    Tres SalchichasTres SalchichasDag sedan
  • Have you tried (or do you have space for) a spiral track?

    MedEightyMedEightyDag sedan
    • Use some grqphite

      seiom jvonyseiom jvonyDag sedan
  • 9:14

    Aaron AlquizaAaron AlquizaDag sedan
  • You have been testing the gates out of MMX context, ie on the bench. Stuck marbles seem to free themselves with the smallest jog, in one instance a very light tap. When installed in the MMX surely the overall vibration of the MMX would provide the required jog and prevent them getting stuck in the first place? Just a thought :-)

    Richard Paul-JonesRichard Paul-JonesDag sedan
  • Someday, when I am old and gray, I will say: "I was there when Martin started the Marble Machine X". It still won't be finished but that's not really the goal anymore; it's now about ingenuity and experientation.

    Ever EnigmaticEver EnigmaticDag sedan
  • I think the 'slow descent into madness' subplot is what I really enjoy the most about your show.

    bilishu alissbilishu aliss2 dagar sedan
  • This SEworld journey for me is so enjoyable - part of me asks, "Does this have to be finished?". Looking forward to completion, but will surely miss the journey that got us there.

    Scoopy PigeonScoopy Pigeon2 dagar sedan
    • Sir, I salute your engineering skills, bravo!

      bilishu alissbilishu aliss2 dagar sedan
  • The problem is gonna be that eventually the hard steel bearings will cause indents into the much softer wood, which leads to the decrease in output force of the marbles , but also causes them to jam much more often.

    Gabriel KrivianGabriel Krivian2 dagar sedan
  • I'm still mad at you that it took me almost one month to accidentally find the second channel

    SuppenfischeintopfSuppenfischeintopf2 dagar sedan
  • Martin: Weekly videos is too much! I need to do fewer videos Martin 2 months later: DAILY VIDEOS IS PRETTY DOPE

    Andrew BAndrew B2 dagar sedan
  • Why not attach cam behind the track that will shake the marbles loose if they get stuck?

    Lord SandwichLord Sandwich2 dagar sedan
  • Have you also considered trying different materials? I think aluminum or some sort of plastic have different friction coefficients to plywood and might also have a significantly influence to the marble flow... Especially for the last question mark design with nylon or similar you might have other results...

    01739485012017394850122 dagar sedan
  • And what about dust or other garbage in marble pressure gates?

    Руслан ГрызуновРуслан Грызунов2 dagar sedan
  • One Design! One Angle! Chant it from the roof top people! ! You keep saying you want One gate and track that is the same across the the whole machine. That way you don't have to make a custom one for each track. One gate would be great. One design leave you with only needing a few spares around on hand for a tour. But ... some of the gates leans back at an angle. There is your Mole you designed in. Keep all gate the same angle, just direct the marbles using the wire tracks to hit the mark.. One Design! One Angle! Chant it from the roof top people! ! Ps. You put levelers on the machine. Do you plan to set the Machine at a 30 degree backwards angle one day and 22.5 forward the next day, No you put them there to Level the machine to a level "standard" before each performance. One Design! One Angle!

    Junkboy888Junkboy8882 dagar sedan
  • чëткий вариант! круто. жду новых видосов.

    An IvAn Iv2 dagar sedan
  • what you wshould be mesuring is the amount of pressure per ball added

    cheese tastycheese tasty2 dagar sedan
  • Use some grqphite

    Charlie HaugCharlie Haug2 dagar sedan
  • What if you introduce some kind of vibration or oscillation on purpose, to overcome the issue with the marbles getting stuck? just to prevent them sticking nad maintain minimal amount of kinetic energy at all times

    Artur KłonieckiArtur Kłoniecki2 dagar sedan
  • 9:16 --> nerve break down

    Simon JöhrSimon Jöhr2 dagar sedan
  • PMA hodler? Are you into cryptocurrency, Martin? (I am...)

    NemozoliNemozoli2 dagar sedan
  • You sound swedish. Or norsk (I either don't know or don't remember what "norsk" is in english, but if I'm even close to right, you know what I mean). edit: The name of your channel further proves my theory.

    Illogical ThinkingIllogical Thinking2 dagar sedan
  • Did you just say. "Beta Test"

    BetaTester704BetaTester7043 dagar sedan
  • "Marbles are evil, lets do 30" lolol

    Justin M.Justin M.3 dagar sedan
  • Sir, I salute your engineering skills, bravo!

    Diego MartinezDiego Martinez3 dagar sedan
  • great work. Resist the urge to over-optimize. Go with #4--less than 100g of weight to resist, and doesn't get stuck.

    Brandon HicksBrandon Hicks3 dagar sedan
  • I have an idea ! Tu veux utiliser des lames ressort (+ feutre ?) pour ralentir les billes, réglable par vis. D'une manière générale, une buté mobile. Plus une bille arrive vite sur la lame, plus sont énergie sera absorbé, donc ça devrait, aussi, réguler la vitesse.

    aerisaeris3 dagar sedan
  • @wintergatan you have dragged on this for too long. Tried many solutions and they all have the same issue: marbles getting stuck under pressure. Have you considered adding vibration to the marble gate system? The marbles usually get going by a slight nudge from another marble. You can probably try putting some sort of mechanical vibrator that can be powered with the machine drive train or even some sort of hammer that strikes each channel when the marble gate closes. I worked at a pharmaceutical plant where they had a similar issue feeding small tablets into a container using a queue not unlike the marble channels in the MMX. The solution was to add a small and simple pneumatic hammer that was in sync with the machine feeding cycle. Each time the machine fed tablets to the container, the hammer stroke several times in quick succession to make sure that tablets did not get stuck. That solution would not work in the MMX due to noise concerns but the idea is the same, use vibration. This would also work for the pickup section of the fish stair. Hope this helps.

    Luis RLuis R3 dagar sedan
  • seworld.info/will/qGrbtrPTjJyXrKM/video

    Alguém por aíAlguém por aí3 dagar sedan
  • Random question, could you have a "shaker"? something that repetitively hits or moves the channels? Either a bar that taps the back of the channel or shifts it to cause some vibration.

    The Man With No NameThe Man With No Name3 dagar sedan
  • O long time no see

    HZStudioHZStudio3 dagar sedan
  • MARIO!!

    Landen HoefsLanden Hoefs3 dagar sedan
  • ever thought of a graphite lubricator mecnanism that continually applies graphite as a dry lube to the marbles???

    robert kennionrobert kennion3 dagar sedan
  • If there is some sort of solidworks for these designs, that would be great for testing before CNBC.

    Pedro PereiraPedro Pereira3 dagar sedan
  • Don’t worry about it taking so long. I believe many of us are makers at heart and love to learn with you. Of course I want to see the final machine one day but that would mean the journey would kind of be over. Sad day 🤣 While we‘re at it, I am so so so impressed how your skills and personality have evolved over the course of that project. Oh, and can you make some music on the machine for us?

    Michel FeldheimMichel Feldheim3 dagar sedan
  • I have just had this idea... Maybe it can help. When being played, the MMX constantly vibrates and moves around. Whenever a marble gets stuck in these gates it really doesn't need much to make it fall through. So maybe, and that is a big maybe, you don't really have to care about them being stuck just a little and while they are built into the machine everything runs smoothly. So I think you should do a test with the gates built into the machine. If you are lucky none in a couple thousand will get stuck...

    Orlando B00mOrlando B00m3 dagar sedan
  • You didn't tilt the two question-mark-paths. If you tilt them in the way you mount them on the machine, the marbles will actually have to go *upwards* which will get them always stuck, if the marble lane feeding it is running dry!

    Martin WMartin W3 dagar sedan
  • don't dent or mar as easily. Wood will need some finishing done to deal with the same conditions.

    vbddfy euuytvbddfy euuyt3 dagar sedan
  • 9:14 Gunde Svan plays Mario?

    swerasnymswerasnym3 dagar sedan
  • I believe tin the ?! The vertical path on the right should not be completely vertical. The marbles gets stuck when the center of the bottom rightmost marble, is pushed further right, than the center of the top rightmost marble. If the path forces the bottom rightmost marble's center to always be further left, than the top one, then it should never get stuck (as long as the width of the track doesn't allow for a similar problem. But try to add a small wedge for the bottom marble, forcing it a little bit further left.

    Martin FyhnMartin Fyhn3 dagar sedan
  • just and idea. why do you need the zig zag vertical. culd it not just aswell be horizontal. that way the slope would be the same but the pressure bleedof wuld still work. and you could make all the gates in one piece of playwood. seams like it would work

    KanonOCKanonOC3 dagar sedan
    • 14:05 "Wilson, food time!" was the most adorable thing I've seen this week!

      vbddfy euuytvbddfy euuyt3 dagar sedan
  • marbles are evil

    Ramón CorreaRamón Correa3 dagar sedan
  • Hi Martin, can we that don’t have engineering experience help in any kind of way? (except being member of the channel of course)

    Oskar DampcOskar Dampc3 dagar sedan
  • You need lateral force, and force ball rotation, also consider changing wood type for more friction. The you need a Peogressive zig zag design thats get closed in a logarithmic way.

    juap!juap!3 dagar sedan
  • Have you tried making the wooden surface smoother? Reducing friction should work.

    See SharpSee Sharp3 dagar sedan
  • Why dont u use another material with less friction than wood, cause this is the main reason they get stuck

    Reinhard EderReinhard Eder3 dagar sedan
  • Marbles = Evil --- > Evil Machine X ?

    misterbalrogmisterbalrog3 dagar sedan
  • I found you because of the second channel. Thankful for it!

    Russell HopkinsRussell Hopkins3 dagar sedan
  • I have a question, why didnt you tried to angle the other designs beside the first one

    Gabriele SerraGabriele Serra3 dagar sedan
  • Discordians are the followers of the Principia Discordia, not users of Discord!

    Archie HalliwellArchie Halliwell3 dagar sedan
  • cool, but why is this on my recomended?

    ApiApi3 dagar sedan
  • I was wondering where the daily updates went! I wasn't subscribed here. Life is good!

    Wizard VegaWizard Vega3 dagar sedan
  • PMMA Hodler? Are you trying to tell us something? ;-)

    Anthony HAnthony H4 dagar sedan
  • still waiting for that monster stand up and eat you....

    aldo baldo b4 dagar sedan
  • po, de boa, ja virou um negocio muito lucrativo rpa ele isso. toda a hora ele "cria" alguma coisa e atrasa mais ainda ....está me cheirando marmelada

    David Lima gamesDavid Lima games4 dagar sedan
  • 14:05 "Wilson, food time!" was the most adorable thing I've seen this week!

    BentroenBentroen4 dagar sedan
  • "Marbles are evil." -"Marble-Machine" Martin, 2021

    Alpinwolf5Alpinwolf54 dagar sedan
  • You know Martin, if you're willing to use math, you can calculate the exact amount of pressure in a very satisfying and not overly-complicated way. Then you can call yourself an engineer with no reservation.

    ArcKArcK4 dagar sedan
  • New haircut :)

    ChaomaticChaomatic4 dagar sedan
  • Martin, the vertical exit, though pleasing to look at, might provide too much of a pressure drop at the very end. The pressure has to go somewhere and I think it’s amplifying the friction against the 90 degree wall, even worse when the gate is angled. There might be a sweet spot, less than vertical, where the pressure is being reduced, but not enough to cause the friction to overcome gravity. If, of course, we still believe in physics. 🙂 Love this content!

    Larison JohnsonLarison Johnson4 dagar sedan
  • why dont you place magnets around the edge of your test table so they dont keep rolling on the floor?

    James DanielsJames Daniels4 dagar sedan
  • You seem very happy ^-^

    Lennart KrügerLennart Krüger4 dagar sedan
  • You also have to take into consideration the amount of vibration going through the gates during use, and how much vibration it takes to unstick the ball bearings. I'm fairly certain the machien will be vibrating enough to prevent them from sticking during operation. Part of the sticking is probably also related to the the compressibility and friction of the wood, where the ball bearings are pressing and sticking into the end grain of the wood. If you can get a block of HDPE, try making a sample gate out of it, instead of plywood. You could also try coating the bearing path with some kind of lubricant like a spray-on lithium grease or silicon grease/oil

    Ian AdamsIan Adams4 dagar sedan
  • on part 5, change the angle where it get stuck. Friction on the wood is to big

    M. JeanneM. Jeanne4 dagar sedan
  • Have you ever tried with a material with less friction?

    Michael RodinMichael Rodin4 dagar sedan
  • couldnt you do the marble channel in a different material?

    Michael WilkinsonMichael Wilkinson4 dagar sedan
  • 9:13 are you okay?

    EckoesEckoes4 dagar sedan
  • How about a double spiral channels?

    Kamil KulińskiKamil Kuliński4 dagar sedan
  • Where did his hair go

    Oliver WesleyOliver Wesley4 dagar sedan
  • seems like the reason the question mark is getting stuck is the static friction between the wood and marble, since the pressure from above is high enough to make the marble stick. If friction is the issue, some graphite lubrication might solve the problem nicely. another potential solution would be to make this part out of clear acrylic (like the conveyor belt) which at a glance seems like it would have a lower coefficient of static friction with the marbles than wood, and offers some aesthetic appeal

    Geoff NGeoff N4 dagar sedan
  • HAHAHA xD I love your humor! The Mario reference was AMAZING!

    CikayCikay4 dagar sedan
  • I love your song star machine 2000

    de a-animatorde a-animator4 dagar sedan
  • I can't wait for the new machine to be finished!

    SkeletorSkeletor4 dagar sedan
  • 9:14 looks like... he lost his... marbles

    LiLENZO 101LiLENZO 1014 dagar sedan
  • happen in normal operation. You need to measure successive falls of the marbles ... and you might need a sideways escape in your testing to enable this.

    kolim jonekolim jone4 dagar sedan
  • I love the new format 😍

    Elias RadauerElias Radauer4 dagar sedan
  • im finally able to watch this video now that ive binged everything from winergatan 2

    Andrew HawrylukAndrew Hawryluk4 dagar sedan
    • 4:16 I see you are a fellow investor...

      kolim jonekolim jone4 dagar sedan
  • If you increase the first bend of the question mark one with the radius of the marble it might eliminate the marbles getting stuck! Because it seems the two marbles fit fairly perfectly in that turn! I think that the marble that gets stuck finds enough surface area to get stuck on in the second turn with the additional pressure of the second marble which find enough surface area against the back wall with the pressure from the weight of the remaining marbles, causing that first marble to get stuck in the first place. I think that if there was more area between those two turns then they wouldn't get stuck as easily.

    Lòriethalion Isilgor MithredelLòriethalion Isilgor Mithredel4 dagar sedan
  • I was wondering where the daily video was from the second channel :P

    sabata2sabata24 dagar sedan
  • Perhaps the lack of pressure is what causes it to get stuck

    HmmmmmLemmeThinkNoHmmmmmLemmeThinkNo4 dagar sedan
  • Angled table seems too realistic, what about a table with a bunch of marble size holes that empty to a funnel that pumps marbles into a container nearby?

    Samuel SkalaSamuel Skala4 dagar sedan
  • me: "you can't convince me. the best youtube moment is--*" martin: ****9:14****

    idan shemeshidan shemesh4 dagar sedan
  • The haircut looks great! :D

    HmmmmmLemmeThinkNoHmmmmmLemmeThinkNo4 dagar sedan
  • 5:45 Now you make it like this: |_ |_ |_ |_ what about this: / \ / \ the right angles are for me not that simple. But with the second version you will even by mounting it 20° rotated not have a hill in that the marbels have to climb if the bottom side as in your design is near horizontal it can easy be tilted back so the marbles (with nearly no pressure) have to climb.

    TechnikchaotTechnikchaot4 dagar sedan
  • make little hammers that give the gates a little knock every beat? :)

    LouisLouis4 dagar sedan
  • Martin once you go beyond the "vertical" you get back into the same issues! The '?' design should not be better that the 'not-?' ones

    pinco pallinopinco pallino4 dagar sedan
  • Go with the tracks that have the highest pressure below 100. You don't want to go with the lowest pressure, those are the ones most likely to get stuck. You said anything below 100 was acceptable, so why not take anything below that threshold? Imagine that you had a rocket engine that you wanted to keep below 500 celsius. You have one orcket engine that maxes out at 400C. Another one barely goes above 50C. If you disregard all other factors, the 50C engine would be best. But then you'll realize that the 50C engine isn't firing correctly and you didn't account for that because you were only looking at temperature. Same goes for the tracks. You need it to make sure that the marbles will come out more than anything else. Keep the pressure within acceptable limits, but not too low, or you're losing your guarantee that a marble will always come out.

    trevdaktrevdak4 dagar sedan
  • The more pressure gets relieved, the more pressure there will be between the marble and one side of the path, the more likely a marble is to get stuck against the side. I see a trade-off :)

    Janne PeltonenJanne Peltonen4 dagar sedan
  • I saw once, a manufacturer that produced steel ball bearings like yours. The process is incredibly simple and reminded me of rock companies that filter aggregate using different sized screens. The hot steel balls roll down a ramp of two steel rails. The gap between the two rails changes and "filters" out different sized ball bearings. They don't all make it into the same baskets. Some that are border line do make it. The companies are more accepting of certain tolerances so they can remain profitable. Your machine is most likely more strict about the sizes that fit and those that don't. I probably didn't see you talk about this already, however I wanted to ask, have you checked the size of each of your marbles? The infrequent occurrence makes me think there is a random marble that is a different size. It keeps coming back around at random times and that accounts for the sticking marble. Keep going!

    J. C.J. C.4 dagar sedan
  • Dammit, i thought this was a wintergatan 2 video and i was like "oo more testing!!!"

    Atomicwinter 31Atomicwinter 314 dagar sedan
  • Did you say hodler?

    TheGreatDane3TheGreatDane34 dagar sedan
  • For marble track #5, you need to make the vertical exterior wall (the side on which the number is written) be tilted towards the pipe. This will make it so that the marble that gets stuck cannot be pushed into the wall by the marble above.

    TheGreenLumpTheGreenLump4 dagar sedan
  • 4:16 I see you are a fellow investor...

    Matoro342Matoro3424 dagar sedan
  • If the marbles are unstuck with a little tap.. Would the general vibrations of the MMX not cover those outlier cases?

    bryanarbybryanarby4 dagar sedan
  • This guy is a genius and he is nice to shoot videos of himself explaining his craft - his country should make him a national treasure!

    Jasen ChambersJasen Chambers4 dagar sedan