These Bowden Cables Surprised Me!
Vložit
- čas přidán 31. 01. 2023
- Support Wintergatan:
- Patreon ► / wintergatan
- CZcams membership ► bit.ly/4cQVM7C
Marble Machine Engineering Discord Server:
/ discord
Video edited By Martin and Hannes from the Trainerds CZcams Channel:
/ trainerds
-
PATREON ► / wintergatan
CZcams MEMBERSHIP ► bit.ly/4cQVM7C
WINTERGATAN RECORDS ► www.wintergatan.net/#/shop
SPOTIFY ► bit.ly/2oKxXWd
ITUNES ► apple.co/2ntWNsZ
ENGINEERING DISCORD ► / discord
COMMUNITY DISCORD ► / discord
-
- These Bowden Cables Surprised Me with Spooky Mechanical Action at a distance!
#4k #wintergatan
-
MUSIC DOWNLOADS ► wintergatan.bandcamp.com
WINTERGATAN RECORDS ► www.wintergatan.net/#/shop
SPOTIFY ► bit.ly/2oKxXWd
ITUNES ► apple.co/2ntWNsZ
MERCH ► teespring.com/stores/wintergatan
DISCORD ► / discord
SECOND OFFICIAL CHANNEL:► / wintergatan2021
- - Hudba
As an avid cyclist, here's two pieces of advice in using these cables:
1) Get the expensive stuff. It's MUCH more durable when it comes to rust, kinks, stretching cables, and glide friction. Cheaper cables will eventually betray you in exceptionally frustrating ways.
2) Use the endcaps (ferrules) that come with the cable. They prevent the outer wire from fraying, and stop dust and grime being funneled into the channel on every actuation.
Edit with more ideas from the replies:
3) Do not push on these cables. It'll work for a while, but eventually your cable will buckle. Instead, make sure the inner cable is always under tension. The mute mechanism will fail in this regard, the marble dropper gets it right.
4) Cable stretch will introduce timing errors. You won't measure this in the studio or your tests now, but you'll absolutely know before the end of your world tour. Maybe not introduce this vulnerability into the dropper design? Alternatively, get the good stuff and have adjustments like on a bike.
5) Go to a good Local Bike Store (LBS) for mountainbikers, and explain what you want to do. They will assist you with proper tools, installation and gear choice. I can't think of any bike mechanic that won't be thrilled to help you and save you a lot of time and frustration.
6) @Nevir202 suggests "push-pull-cables" if you wish to avoid tensioning the inner cable.
Edit 2:
This is not a put-down, I have been fascinated with everything you do since MM1! Go Martin!
I second this advice. Also, the cables have a tendency to crush at the attachment points and then subsequently slip through what is clamping them in place. You need to introduce a turn or edge or something the cable is clamped around to help keep it in place otherwise they loosen with time.
The stretching of the cable and loosening over time will not do kind things to your timing I wouldn't think if these cables are being used between the indicator and the gate for the gate to drop marbles.
i highly support that advice as well
I might add that smoothly ground cables are just nice. Less friction.
And instead of using that two plate and for screw, it's possible to just have a block that stop the end of the guide, will save time and parts
Also cyclist and I just wanted to add to your suggestions:
1. I use nice PTFE stuff and grease the cables, grease reduces the friction and makes it shift even smoother IMO.
2. Agree on the endcaps. (Edit: think your referring to the 'housing ferrule' and I'm referring to the 'cable tips' here - use both) Goes from sharp end to sharp and frayed all over and no longer possible to put endcap on easily, do them right away. Use steel wire cutters not side cutters etc. Generic work just as good as park tool IMO. Dremel cut off wheel to cut inner and outer together works as well.
Spooky mechanical Action at a distance you know
So Spooky!
@@Wintergatan_2 Sure it is!
I obviously never knew Einstein personally, but I'd like to think he'd appreciate the mixture of engineering and art you put into this project.
Is it weird that "Spooky Action at a Distance" is the first thing that came to my mind when you first showed the cables?
Are you planning to completely mute the machine except for the instruments you'll be playing or are the marbles gates supposed to be a some sort of musical instrument?
As a bike mechanic from Amsterdam i have a few tips:
Use an end-STOP for the outer cable, this way you don't have to clamp the outer cable which could add friction or indjuce miss-adjustment over time!
Also I would not suggest these inner cables for pushing motions unless you have a spring pulling on the other side!
yes, he should keep the cable in tension either with a spring or self-righting mechanism/feature (counterweight?), I completely agree with you.
Solution for push is use a solid core instead of twisted strands. Has worked for a long time for choke control on engines. Also used for other automotive controls where you need push as well as pull. Like the slider for temperature control.
For such applications the sleeve can be the cheaper style with a spiral of metal (like brake cables) instead of the expensive type with lengthwise stands (for indexed shifters).
@@TEDodd Spiral housing cables will change the length if you bend them like in the video. Braking cables are not a good choise for this project.
@@PhreakDarkSoul spiral housings are fine for this application. The precise positioning need for indexed shifters isn't required.
The marble gate is pull only, just like brakes.
Use a solid core for the push-pull, operation of the other part. It works fine for throttles, chokes, and moving doors to control airflow. My mower and tiller use them. There are 4 for temperature/airflow on my '71 Super Beetle. Most are bare metal sprial housings.
I scrolled down to comments looking for this one :D
I LOVE all this small scale testing! Your streams were incredible brainstorm and modeling sessions but THIS is where the real magic happens! Very refreshing to see how practical and hands-on the MM3 journey has become!
Same. wish he had done small scale programming wheel testing.
All in time, the progression is fantastic, proof that starting when you aren’t prepared but are set on finishing can be beneficial if you ar e paying attention. Love this so much
Most of the designing and cading is now offline, so we only see the more exciting tests
Absolutely lovely use of some off the shelf components. And bike parts, much approved. A few remarks:
* Get yourself a cable cutter, available from a bike shop. These cutters will make a clean cut and do much less damage to the cable.
* The steel cables will fray, always use ferrules on the ends that you cut. I'm sure the bike shop will be happy to show you what these are and how to apply them.
* Add some sort of spring or tensioning mechanism to the remote end of the Bowden cable. You'll want to have the inner cable under tension. Do not rely on the push from the inner cable. This is done correctly for the dropper mechanism, but I fear that the mute mechanism as you currently have it may fail.
I was going to say the same thing. The cables are designed to operate under tension. I wouldn't put too much faith in their ability to consistently operate under compression. Either spring load the cam, or somehow design it so that wants to flop back to a home position some other way.
Solution for push is use a solid core instead of twisted strands. Has worked for a long time for choke control on engines. Also used for other automotive controls where you need push as well as pull. Like the slider for temperature control.
For such applications the sleeve can be the cheaper style with a spiral of metal (like brake cables) instead of the expensive type with lengthwise stands (for indexed shifters).
I have a dual cable system for the throttle on my motorcycle. That way it does not have to rely on the pushing of the cable, it is pulling in both directions, which makes it much more immediate and reliable.
Someone suggested solid wire cables like they use for automotive applications for push/pull situations.
@@kainenable that's pretty good, but it's also double the casings and cables and, presumably, friction.
Hey Martin, one suggestion, bowden cables should only be used for pulling, not pushing. Not only for better performance, the cable can bend on the outside of the tube where the lever is and not push into the tube. To get rid of this issue you can integrate a spring or rotational spring on the other end of the cable (where it's attached to the programming part). Great progress so far, i love it!!
Solution for push is use a solid core instead of twisted strands. Has worked for a long time for choke control on engines. Also used for other automotive controls where you need push as well as pull. Like the slider for temperature control.
For such applications the sleeve can be the cheaper style with a spiral of metal (like brake cables) instead of the expensive type with lengthwise stands (for indexed shifters).
>> "bowden cables should only be used for pulling"
But...an ordinary cable can pull without pushing.
The whole POINT of Bowden cables is that they can push as well as pull.
This is what I was thinking. They are not meant to push. It could be set up to pull the top part up so the hook does not connect, similar to how it looked when he unmuted while it was lifted. This does depend on the flexibility of the 3d printed material and its ability to fall back to its original position reliably.
@@DavidByrden1 that's not right at all. Stranded core cables like that are meant for pulling only. Bowden cables aren't the same as an un sheathed cable, they are designed to pull through indirect directions, without further complications like pulleys etc.
Pushing on a stranded core cable will cause issues. Most likely leading to a jam.
@@DavidByrden1 No, as @TEDodd says, to push you must have a rigid cable. By pushing the cable will take more and more bending on its free part then end up twisting in this free part without the system at the other end moving.
If you have seen systems that give you this impression, it is because you have seen that the joystick side and the other side of the system were in permanent tension with a spring, for example.
When testing the cables make sure to test vastly different lengths.
Make sure that a 2 meter cable doesn't introduce more delay than a 10 cm cable.
I'm sure there will be some point where the amount of play inside the cable will become an issue.
I was thinking the same thing, but the timing critical point in this design is when the programming reader releases and the spring in the ball-dropper pulls the cable back in. There shouldn't be any play because the cable is under tension at this point. My only concern is the springs in the ball-dropper not being strong enough to pull a long cable through the bowden tube. The longer the bowden tube (especially if there are bends), the higher the friction. But how much higher I honestly don't have a feel for.
@@Sembazuru I think he really has to do the test actually in different length conditions and especially the longest length with the biggest twist he wants to use. Honestly the system of bowden cables allows you to transmit an effort but not to be precise to the millisecond. Even if the timing is tight there is no chance that it will release exactly consistently as soon as there is a twist and a slightly longer length. And the worst is that unlike rigid systems, new and used behavior will be extremely different. I think it's the first mmx 3 thing that won't stay or die.
If a long cable has a predictable delay it could be compensated for easily. Delay isn’t an issue necessarily but unpredictability is.
Making the radius of the lever and mute different such that the angles are different has made my mathematical heart glow up like the sun, incredibly genius
If that's what floats your boat, you would love gearboxes.
Take a look on cable operated throttle valve - basically same thing, relatively small angle of throttle pedal is translated to nearly 90° rotation
I love the use of these cables, although I do have one concern that could lead to failure. These cables should always be in tension. Because they are not under a larger load it works in your test but over time the compression stability of the wire could wear out. Although it is an easy fix, on the one movement use the tension in the wire, but for the opposing motion use a spring to add tension from the opposite end.
I'm specifically talking about the mute lever to reader cable. The reader to gate cable already has a spring doing exactly that.
@@michadonald Noticed the same under the mute lever. Its a risk, works now but after friction adds up, cable gets more worn and flexible, longer runs etc. Shouldn't be hard to fit a Torsion spring on the mute cam, or I seen another person suggest using push-pull style cables made for push (not sure how lightweight those come in?).
I think you are right. This design also routes the cable in a curve outside of the housing, and that probably makes the cable even more vulnerable to deform or unwind under compression. Once the cable strands start to unwind, the cable will no longer go smoothly into the housing.
Is there a version of these cables which are designed for compression instead of tension?
I mean there are Pneumatics and hydraulics, but not exactly what we're looking for
Bowden cables generally are not good to push the wire - but rather pull. With a spring pulling in the opposite direction from the other end to get control in both directions. As the wire is prone to buckle when pushing in.
Hej Martin! FYI there are also push/pull bowden cables that will work in both directions if you ever have any issues with the slack. These are normally used on lawn mower hand throttles for ex.
Totally agree, although I'm not sure if they are as flexible as might be required on the MM3. Definitely something to look at.
Not sure if you considered it, but it would probably be beneficial to build a bit of a shoulder into your Bowden cable holder. That will let the casing rest against something and not just rely on the friction fit.
this is purely prototype
Almost as if he should look at how pedal cycles hold the same bowden cables …
@@PabloEdvardo true, but it's still worth pointing out
The ones on bikes get stuck all the time
@@zedudli they really don’t - and of course they’re in a far less mechanically friendly environment as well.
Not gonna lie, I loved the old MMs for their dirty and 'dumb' design. Those were totally unique pieces of art, and every time you talk about them as failures, I die a little inside.
Nonetheless, I'm super excited about those new designs, this is gonna be one phantastic piece of music machinery! Can't wait to hear it play its first tunes!
I know how you feel. I liked the very obvious mechanical designs of them, and i feel they worked, they played music.
Unfortunately Martin wanted consistency and seems to base success on that.
@@benjackson8731 Well, any musical instrument that is to be taken seriously by a musician needs to be consistent, in many ways. So Martin's need for consistency makes sense. Like, for instance, how would you like it if your keyboard erroneously missed a note even though you clearly instructed it to play that note? No musician, amateur or pro, wants that.
There is imprecision induced by the instrument operator himself, and then there's imprecision induced erroneously by the instrument alone. The former is obviously better. If the instrument is precise on its own, the operator will be able to confidently use it :). I'm not even talking about the issue of an instrument going out of tune... that's much less of an issue than the instrument full-on missing notes or erroneously delaying the timing of a note or whatever... things that are not directly induced by the operator.
You should have a look at bike hardware for pinching cables, they would reduce the size of parts dramatically and make it much easier to adjust. Add ferules to the end of your outer too!
In my experience, you can use cycle oil in the outer to improve the friction. just let it pour down over the gear wire. With a lot of channels on the bi-stable mechanism i imagine the friction could add up!
Can you imagine he makes the perfect marble machine and he post all the blue prints for everyone to make at home lol
Hey Martin, just wanted to add a quick little tip for the bike cables and housing. I was a bicycle mechanic for many years, and whenever I'd do cable replacements I would always make sure to put some extra lubricant (I used Tri-Flow, but im sure other similar lubes would be fine). The extra lube does two things: 1. it helps prevent rusting of the cable inside and 2. it lets the cable inside glide much smoother. I would add just a few drops of lube per foot of housing or so.
Love your journey, and if you ever have any questions about bicycles I'd be happy to help :)
Edit:
I saw some other comments mentioning ferrules, and I highly recommend those as well. You can buy them for pennies a piece and will go a long ways in preserving your cable housing.
Also, while cables do stretch a little bit, most of the perceived "stretching" comes from the housing, ferrules, and associated hardware setting into place. An easy way to "pre stretch" your cables is to just use some good ol elbow grease and pull the cable taught and then relax it over and over using your arm strength.
How do I get a job that's that ridiculously specific and make money off of it?
That "no half hits" guarantee is genius and the new simpler system makes it all the more apparent. Everything is coming together so much smarter. I was already impressed with your first two machines. This one is gonna be stunning.
The flat plane prototyping gives me the warm fuzzy nostalgia, harking back to an earlier Wintergatan. I like it 😀 I'm really glad you're keeping the modules from the previous iteration of the Marble Machine that worked for you, and worked well. That gives me a lot more hope that we'll see a working Marble Machine 🥰
I'm also really glad the bowden cables work as well as they do and friction seems not to be an issue! I was really worried about that.
What a pleasure to watch - addressing concurrency issues, low coupling/high cohesion design, constrained but stylish look
Please put the foam on the programming pin so it's easily replaceable when it wears out
True but given the number of programming pins it is significantly more manufacturing operations. Maybe making the finger on the reader easier to replace.
i think it's better to keep it on the reader because you only have to replace it once
You want the wear to be on the reader as to minimize complexity. If any part of the pins can wear the design is inherently flawed. A foam or a soft plastic pad on a few dozen readers is far easier to replace than hundreds of pin pads.
He will be adding and removing programming pins for songs. It's far easier to replace pins as needed. It's the same complexity overall.
These bowden cables on my bike tend to stretch after time. The force used when hitting the brakes is probably much higher than on the marble machine, but it may hurt tightness of playing over time (or may require constant adjustment).
You get similar issues with mechanical linkage, I imagine. Material may expand or contract with changes in temperature or humidity. Maybe the reality is that the MM will need the timings readjusted before every performance.
@@kaasmeester5903 Or else it's a bad solution, there were others. And no: in the introspection on the previous model, he says that he wants to move towards more simplicity and less manipulation, not having to readjust the system of the machine every 1000 notes.
The next beast: cable management
I used to build precision road bikes. I can say that those cables do stretch over time but that can be avoided by prestreching them. I don't think you'll be putting enough tension on them for this to be a problem but I'd prestrech them just in case. Also put a few drops of light chain grease in the bouden cable housing before pushing the cable in. It will reduce a ton of friction.
Another cyclist here, I agree with what many other people are saying about always having the cable in tension, putting ferrules, end caps, and maybe barrel adjusters on your cables to deal with stretch, although I dont think stretch will be as much of an issue given it seems your springs are much weaker than what you see in a bike derailleur. Still leaning on off-the-shelf bike components is a good way to save yourself some engineering and reduce the amount of custom parts you'll need.
Along the lines of weak springs, I was noticing that the dropper seems to be shifting around since the cable can flex a little and the foam compresses & springs back a little when the gate opens, you are probably already working on this but its probably a good idea to retry your timing and accuracy tests. Maybe more rigid foam or stronger springs that bounce less would help
I feel like the black 3D printed parts and their cool clean look really represents the whole ethos of the MM3, feels like we're flying right now with all the progress that's being made!
I love how you made the system resilient to half hits!!!
Aloha Martin, a suggestion for a slightly better design: at 1:55 glue the foam to the stationary part, not the moving part. Less mass to move around with each marble drop and saves wear and tear and friction if the foam happens to rub against something as it travels.
Well Done!!! The small-scale test of the "programming wheel" is a fantastic proof of concept!
As a photographer thank you. I didn’t know what the mechanical cable release was called. And now I do!
Be careful with the measurements of those wires because the length of the core can change depending on the curve of the sheath. I know it can cause some problems when it's use to move something, but some careful measures and cuts should be enough in your case... Good luck for your awesome project !!
There’s about a billion different kinds of Bowden cables (both inner and outer cable) all with slightly different properties. But Have a look at push pull cables, will reduce the need for return springs everywhere and should make the translation of movement tighter.
Great to see you actually build something again.
I like your planning and talking, but the building and seeing effects videos are just more entertaining.
For a while I was worried about the future of the machine but after these short experiments and the brainstorming that has been done for MM3 I feel like this time as long as you stay on track it should be smooth sailing and hopefully you can dodge integration hell and we can all enjoy the fruits of all your hard work on this machine!
I remember suggesting bowen cable / camera release exactly one year ago with a few other people during a live. I'm glad this idea come to life :)
Hey Hannas, I see Martin is showing off his big guns today. I'm so proud, as I know you are too. I'm looking forward to Thorsday tomorrow and seeing how everything is going on your channel also.
Hi Martin, as a bike mechanic I love the use of these cables! Genius! I do advise you to use brake cables instead of shift cables. They are a bit thicker so will have a longer lifespan. Also use outer brake cables. Outer shift cables have a flaw that over time, the inner part comes out and makes for less accuracy. Brake cables don't have this flaw. But please use "not compressing" outer cables with end caps. Also, all metal cables stretch a bit over time so I'd advise you to design a way to adjust for this stretch.
Keep on rocking!!!
Martin have you already taken a closer look at the mechanics of a piano key for inspiration? It's a rather complex but proven design known for excellent control, silent on rebound etc.
Now it's a challenge! You absolutely need to mute-unmute Marble Machine while cooking! :D
and taking a shower. LOL
Cooking stream when?
Two questions I would love to see covered in the next video. 1) how long can the cable be before the force required to push/pull becomes too large? 2) change the top cable to be a long one since you want the drums further apart. Then run the tests for tightness with different length cables.
I love the scientific and engineering approach of this iteration of the marble machine, keep it up and you be successful with this version.
Hi Martin,
2 things spring to mind here with this demonstration.
First: These cables are meant to engage something from a relaxed position. The system should reset itself to the default without interaction of the cable. They are simply not designed to be used as a push-to-enable type. With little force, I don't think it will cause any issues. But be aware that these cables will cause issues when pushing (considerable) weight.
Second, please be advised that when the two interacting hooks are at an angle, not horizontal like in your setup, it may cause issues with timing errors. The hook may lower at a different point/time due to gravity and less force.
Furthermore best of luck with this epic machine project!
I'm so excited for what's to come. All these little tests are super interesting.
Bravo! This is such a smart mechanism! Also, crazy to see how much the assembly has been reduced with that building montage.
His clips of getting knocked out while boxing are among my favorite clips ever.
This test makes me really happy! Especially how the hooks (dis)engage when the mute is switched mid-trigger
Gotta watch out for that cable stretch. Short cables stretch less, and less strain in the cables will stretch less. But they do stretch. I'd also consider barrel adjusters just like bikes have to adjust the tension easily.
gotta love martin's journey in becoming a programmer, using the single responsibility principle, dependency inversion, interface segregation, open-closed and loose coupling in the most literal sense possible to make his designs better
I can't wait for the music vid of Martin playing music in the kitchen while also cooking
Absolutely immaculate design, Martin! Well done!
Martin your design thinking has revolutionized the methods I use as an Engineer. I've realized that I had dumb engineering requirements in some of my projects and I'm working to make simplified new designs!
Your Marble Machine Extreme is coming along great! Keep up the good work!
This is actually pretty cool. You learned the design principles of machinery in the hard way. But that means you'll never forget those!
So far you are relying on gravity for a lot of actions. You shouldn't. I suggest you to introduce elastic members when you need a piece to "fall" into certain position. Aside from the marbles themselves, those are heavy, the really light (and almost massless) bars of your mechanisms can have some trouble of you rely on them falling into place by mere gravity.
Springs, rubber, even little plastic flexible members can do the work and they don't need to be strong. They just need to assure you the pieces follow the path you intended (for example follow a cam). There are inexpensive flex metal springs that you can attach into your design. Like the ones in the scissors of swiss army knives.
Keep these videos coming. They are great!
Your face at 6:07 - that is a face that confidently says "yes. This is very good indeed."
WOO new wintergatan video!!!
If you have the time after a marble machine is built, you should make a one-off setup for a music video or something where you actually go and put muting levers and random instruments in random parts of your house or studio! I think that would be pretty funny!
I'm loving these small scale tests! Extremely cool. It's awesome to see these components start to work together!
In the mute position, I think a small spring on the pin will be necessary. The cable will not always have the force to move the pin in mute position.
I commented the same thing, the marble gate actually does have a spring already, it's just the one end of the muting cable.
Gotta love ALL the regular uploads
Bowden? That’s my name!
Nicely done! You have shown physically how separation of modules with interfaces (Bowden cables) reduces complexity, through avoiding a highly-coupled system. I'll use this video as an example when I teach this principle in computer programming.👏
Bowden cables introduce backlash in the bends, the more bends (and length) the more backlash
huh, I didn’t know that Bowden tube/cable wasn’t an exclusive term for 3d printing. The more you know!
There are a few in your car. Handbrake, gas throttle, clutch, shifting lever, window opener, heater flaps, inner door handle trunk release.......
The Bowden cable was patented on August 23, 1898 by Ernest Monnington Bowden (Patent US609570A).
The 3D printer probably came a little later.
@@sebastianreitebuch2906 no Im pretty sure that they were around back then
@Wintergatan
The next step to reduce friction is hydrolic cables with no actual cable but oil in a tube, that is pushed and cannot be stretched or compressed. It then pushes on a piston on the other end of the tube which can activate whatever you want.
I love it when Martin uses existing concepts and applies it masterfully into his machines. Not only that, he finds a way to use said technology so that when something is accidentally engaged or disengaged that minimal damage is done. These contingencies would help keep the parts, and the machine altogether, safe from breaking down.
Bowden cables have been used for decades in animatronics and puppetring, and if they are used as (bicycle) brakes, they can never go wrong
There’s an engineering saying “you can’t push a rope” but these cables are all “hold my beer”
The problem is you actually can't, even with these cables, if there's any amount of resistance the cable will kink at the end where it's exposed. The fix is to add a spring like there is on the marble gate.
You can when the rope is constrained in a tube. Physics is weird.
Nah. The Mute cam should be pushed by a string not by the cable
@@Mostlyharmless1985 2 issues, 1: the cable is not designed for compression. It will eventually start to wear if there's any resistance.
2: the part of the cable that is exposed is still just a (rather stiff) rope trying to push. It can right now, but if there's wear or resistance it will bend and the motion won't lift the muting pin
A lot of budget and hobbyist 3D printers do exactly this, and with much more resistance than this machine is making the cables push. Sure it may not be the most efficient or optimal use of a bowden-type setup, but does work. Besides, he is putting the higher resistance of engaging the linkages on a pulling motion, while the push is at least partially assisted by the weight of the latch. If he ever reverses that setup or tries to put too many muting levers on a single cable I will be more concerned, but the design he used for the prototype is not completely unsuitable
Amazing design Martin! Keep this up and you will finish Marble Machine 3!
I think there was fear that your engineering improvements might make the machine less mechanically interesting. This video makes it more mechanically interesting than ever. Everything seems so clever.
We decouple software modules all the time in systems engineering to preserve their independent operation and mantainability. It's truly interesting how you are approaching the decoupling process here. Kudos!😋
LOL. Nah, bowden cables will never surprise me. Bowden cables are basically magic. I love them.
Love this. Starting to become good old Wintergarten engineering (of course better every time) 😅
Marble Machine 3: Unconstrained!
Nice one, Martin!!!
Martin I’m so excited right now, you mad a great decision to change your method of building and it shows! Thanks for keeping on with this project, you are becoming an incredible system engineer!!
No one correct Martin on his amazing pronunciation of 'chisel', we need to preserve what little purity is left in this dying world
Now I'm concerned, is that not how it's pronounced?
@@davidgustavsson4000 Martin says the ch sound like a sh sound, so shisel rather than chisel. you can tell he's not a native speaker but it doesn't affect understanding so it's not a big deal
@@mymoomin0952 ah, yeah. That's a known weakness, I'm Swedish too so I can't even hear the difference unless I focus on it.
There's a Swedish phone operator whose mascot is a sheep and whose slogan is "Born to be cheap". It only works because Swedes don't know the difference.
Braziilll
An old fan here - cyclist / designer / musician.
Look (and disasemble if needed) at the parts of an old bicycle friction shifter and derailleur for inspiration - they are tested and proven
1. use shifting, not braking cables. you do not need as much force as is needed for bicycle braking, you will save some weight as well
2. use them in tension only (again look at derailleur)
3. design end stops for bowden casings into printed hardware -this will save you screws and time.
4. I know you said that the system with the brake on the reader worked for you, but I cannot escape the feeling that the whole system can be made simpler - with the lever disengaging the marble gate itself. seems to be more first principles approach... also, it will make the reader>actuator>marble gate system simpler and more robust.
Yes - the marble gate becomes more complex, but you then have 1 zone that needs extra attention in designing and quality control in manufacturing which is a life saviour when scaling up.
Live long and prosper, Cheers
i'm so happy you got back for this video format, instead the live only ones
Brand new gear cables will stretch (unwind slightly) over a few weeks and months of tension so worth keeping an eye out for or adding a tensioner to tune them up again
Now you can even stand close to another instrument during a concert and program MMX-3 :) Awsome!
I'm so happy to see you start using bowden cables, they're a game changer for machinery, very wire-punk.
Just be careful and try adding few springs as resets by the end of them as those cables (or ropes in general) are great in tension but terrible at pushing things as they can bend or kink. Otherwise the cables tend to squish like those chinese finger traps and adding friction.
If you want to make sturdy or custom sized end caps I often put a knot in a metal bead and fill/melt everything together with solder.
Also, lubricate those puppies like there's no tomorrow, good against wear but unlike us you'd probably seek a smoother transition to shave off as much delay as possible.
I think I said "wow! that's really smart!" out loud about 5 times in this video!
amazing work! so exciting to see your tinkering and inventing!
Love the way your engineering this machine. One thing to keep in mind with these cables, you can fix the outer sheath at one end move the inner cable then fix the inner cable at the other end and use the outer sheath as the moving item. You just need slack in the installation so the inner and outer ends move independently. This might overcome the issue mentioned by other of pushing the cable. It might also just give an alternate method of operation in some circumstances.
*@Wintergatan*
With this Bowden cable design, you can even program in a "mute track", so the instruments turns on & off automatically depending on the programming, with all the precision too :D
So, a new track on the programming wheel, controls the On/Off lever, instead of the marble-dropper.
Martin: "I don't have a chisel."
Every wood worker and chef the world over: "Martin no, Martin please, Martin whyyyyyyy?"
I would be so interested in a video describing your philosophy on having a vision and being persistent as well as your experiences of the emotional ups and downs along the way
fun fact I learn once about Bowden cables, some weapon, called bullpups, make use of their principle to move the trigger away from the action. as you want no delay between pressing the trigger and the gun firing, the only solution is often a cable or a steel bar to transmit the force. so, yeah, bowden cables rocks (now that I think of it, it's also why they're used for breaks on so many vehicles - not just bikes - their instantaneous nature is unbeatable.)
Brilliance! Simple and pure :)
A couple of observations from 30 years or riding and wrenching if I may;
1/ Use the cable to pull on the device and a spring to return to normal as demonstrated with the marble gate. Every cable actuated bicycle component works this way. If it aint broke...
2/ Keep cable runs as short and straight as is practicable for each application, every bend and loop will add a little friction which might upset Martins righteous obsession with "tightness" Incorporate barrel adjusters for ultra precise fine tuning
3/ Quality sealed systems with non compressing cable outers will cost more in the beginning but pay off with smoother, "tighter" operation and reduced contamination issues, I use Jagwire Elite Link system on all my bikes and heartily recommend it or a similar system
I Believe!
7:10 I had this question in mind when looking at your experiment 😂 well done Martin!
Love how you go thru the whole process can't wait for the many vids to come and final product review
Great Idea! A point of caution. If you are in any way relying on the cable to work bi-directionally (push and pull) you will eventually experience a failure with the stranded cable. Stranded cable is best for flexibility and works reliably in tension. If is is being asked to push a component the cable will over time develop a kink and fail in that action unless the required force is tiny.
Bi- direction bowden cables should be solid spring wire. These were often used in older cars to actuate HVAC vents and such.
Not as flexible or smooth, but bombproof.
I know the 'felt' under your adjustable lever which gets pushed by your programming cam peg looks like a good idea, but it is a part which will randomly wear. Try testing 2 - 3 bearings in place of the felt. They will provide less friction as the programming peg-module glides under them and all it will do to your mechanism is remove the felt, add 2 screws with small bearings mounted in them. You may find that as the rotation speed of the program wheel and it's pegs might not incur much noise at all.
Love that you could probably also mute and unmute with the reader itself so you can have a channel that could give you parts of a rotation with no human intervention!
Hadn't seen Wilson that happy in a long time!
I can't help but get excited every time i see you put up a new video that implements a new concept. These cables really feel like a GIANT gamechanger if you were to move forward and build a new machine.
Your videos continue to be outstanding. The design process, the engineering process, the creative process - while any of those can (and have) brought you frustration, how fulfilling it must be. I don't know if you are an inventor, a musician, an engineer, or a systems analyst, or all four - but whatever the answer is the end result is a fascinating journey for the rest of us. Thank you so much for the content!
It's so interesting watching each step of the development seeing you go from conceptual design to experimental design, and finally functional design after learning what causes flaws during the experimental design. It will be really amazing to see you go to a design of beautiful function once you understand how to get the basic concept down and figure out how you can add some of the interesting aesthetics stuff that you went for in the second design in a manner where they work in a error controlled environment so that they aren't necessary for the functioning capacity yet function via a small amount of diverted marbles in a manner where it manages excess on any paths if there happens to be any excess... At least I think that would be cool to work towards for a 4th design if you ended up deciding on creating extra functionality like that.
Ps. The crazy piano guy is just jelly... No wonder he was firing shots ;P
Fun thing about that bistable mechanism, light switches have used that since 1884 for quick-break to extend switch life and reduce fire hazards, I never quite realized that the switch lever on a light switch is not mechanically bound to the switch itself, its why when switches get old they don't have the characteristic snap that new ones do, the spring is worn and soft so the quick-break isn't so quick anymore, which also makes old switches a possible fire hazard in regards to airborne flammables. I watched a video on the history of light switches a few days ago, quick-break switches used to be these robust and expensive things and now they're made of mostly plastic and like at most 7 or 8 individual parts.
Of all the tools in your shop, the kitchen knife Chisel is the best! 😆
I love love love your analogy with the cables to "Spooky action at a distance". Mind-blowingly apt, and I think you might have something there!
I love that everything is so precise, then you whip out the pencil that is blunter than a pre schoolers 😂😂
Loving this new fase of "really prototyping and really testing"! MMX3 will be a masterpiece!