Everything (else) I learned about knitting machines while designing one of my own

[@scaredyfish]

I’m a programmer with a 3d printer, and my wife does machine knitting, so this is the perfect intersection of our interests.

[@galaxythree4653]

I would suggest you take a look at the Passap Vario plastic bed knitting machine. Here you can see another way of making the needle spacing plates without the screws. This machine knits both 5mm and 10mm seperatly or at the same time on different sides of the bed. I would love to explore this with you in a bit of conversation. You can google this machine to find the manual as well as 2 youtube videos about it. Also as to needle background. Early on needles were more rigid than on many machines today. They did not use a sponge bar as the needle retainer. The needle retainer could also be more durable. The sponge bar used today is for more flexible metal needles that actually bend slightly for lace knitting performance. This is way too much to text into a comment box. Even in the much older home knitting machines there are several different construction designs. The practicality is based on the desired use. Most all machine can do the same things just differently. A study of the cam systems of different brands is also interesting. This is for the locks, AKA carriage or basically the needles paths. The Bond system used easy to drop in key plates of different configurations. Some machens have disc's like older sewing machines, still others have a series of buttons, levers and timing belts to produce different needle paths. 🫠🤗

[@nobodybirthday]

I have a mild obsession with knitting machines this video series so far has been absolutely fascinating

[@DeLaMangaShoes]

Your presentation is very well spoken. I enjoyed hearing about your journey. As a custom shoemaker that uses knit uppers, I just bought numerous machines in different widths for the different yarns. I'm not a computer person so a lot of the technical talk was lost on me, though your clear and amusing presentation was definitely engaging! Thank you very much for this.

[@SpencerPaire]

Holy cow! The amount of talent here to jump into new ideas, designs and tools, balanced with the humility to use existing designs where appropriate, is truly singular. I'm a full-time engineer, and the work I see in the project rivals the output of many of my coworkers, no disparagement to any of them. Fantastic effort and video! I can't wait to see where it goes!

[@MartinMakesThings]

Your long and detailed videos are a true joy to watch. It’s always fascinating to see how people approach the design of machines like this. Thank you so much for sharing!

[@nysaea]

what an adventure, thank you for taking us along with you!

[@SquiddyKy]

Thank you for the brilliant deep-dive into your process! It's fantastic to see all the hurdles, stumbles, and triumphs you've had through this whole thing, and I can't wait to see how future iterations of your machine turn out. If you get around to making a ribber, I am *absolutely* going to alter and print one to pair with my own Knitmaster 360K machine.

[@OnreinKalfje]

I love that you used openSCAD!

[@bruceconnervo2142]

As a former Agile trainer, I just want to thank you for going into depth about your process. Don't be too hard on yourself about that first stumble, it obviously helped you find out what you needed to nail down to effectively move forward (especially re: organization - it's just not as fun as all the coding and building, is it? 😀) Anyway, I've 3D printed some sock machines already, and was having vague thoughts about a flatbed design when you popped up in my feed. Thanks for doing all this pioneering work! You know *exactly* how much time you've saved me. Looking forward to watching the design grow and improve and printing my own soon. Not to stir the pot (much!), but have you considered applying compliant mechanisms rather than pivots/springs/etc. to the design? 3D prints work pretty well for those if you don't ask too much of the material.

[@SparksCuriosity]

Haha, totally, I learned so much from doing it badly first, I'm glad I got the chance to get that out of my system early! And yes, I have been looking into compliant mechanisms and I've got a few ideas but haven't had a chance to test anything yet. Fingers crossed!

[@scaredyfish]

Yes, compliant mechanisms were my thought too. It solves the parts availability issue, and could end up simplifying the carriage as well.

[@dnd9865]

I'm SO excited for this

[@blenderintraining]

For the cam springs, look into 3d printing springs. Slant 3D recently made a video on a few springs here. kzbin.info/www/bejne/raHVmnqGaZqIpNk To me, it looks like a spiral spring would work. You could print the spring separate from the cam, and interface the two with a square shaft or similar. (I have no idea how to do a spiral spring in OpenSCAD. Maybe a vector that's extruded?) Also, for your cam position controls, perhaps look into bistable mechanisms. Wintergaten has a decent video on it here. kzbin.info/www/bejne/aaeYqZmNoNVsj9U This Old Tony has a good video on Over Center Mechanisms here kzbin.info/www/bejne/f5KWepybms-snZo I'm not sure how practical a bistable mechanism would be, but if you can manage to package it onto the carriage, it would make for a very satisfying UX.

[@zenengineer5803]

Something I find interesting in these sort of videos is: what can you do with these 3d printed solutions that you wouldn't do normally. Currently you can slot together sections with different needle spacing to make a weave with thicker sections (e.g. a blanket with a fluffier border). You could 3d print a bed section where needles get progressively closer for an effect you couldn't easily get in a different way. Of course I don't know if this makes sense from a fashion design point of view but it's interesting what you can accomplish. On the software side I'd be curious what you could do with a servo (or a few) and programmable patterns. You could lay out an encoder across the beds for positioning (like they use for CNC mil conversions) and have accurate positioning data for your carriage. With some servos on the carriage you could move needles in and out of the path as they pass by without relying on timing and such. Or fo course just count the needles as they pass by or some such.

[@SparksCuriosity]

> You could 3d print a bed section where needles get progressively closer for an effect you couldn't easily get in a different way. Yesssss that's the sort of weird experimental niche I think this could really be cool for. You get it. Re: needle selection, I've been kicking around some ideas since early stages and we might be thinking along similar lines, though I have zero experience with servos so I'm uncertain what's realistic to expect in terms of speed/responsiveness, i.e. could I make the actuator gtf out of the way before the next needle at typical carriage operating speeds.

[@JaneConnorEmbellisher]

I really appreciate all the work you are doing to achieve a make-it-yourself knitting machine. I have a basic knitting machine that have very light weight springs for the flippers. I had to disassemble, partially, my carriage to reconnect one of the springs. The spring was a fine coil with hook connected at each end. Are there other springs you can explore? I will probably watch this again while looking at my carriage to better get a handle on how the carriage works. What was the roll of sponge you were using called?

[@lvikng57]

can't wait to see the code for this, I have a lot of experience with CAD and writing software but just can't seem to connect them together

[@matthewdoane4781]

You don't really need to use torsion springs in this application. A linear spring (or a rubberband) combined with a lever arm wold work just fine. Torsion springs are usually only necessary when space is a concern and you have the entire carriage to work with, in some respects. This would also allow you to easily adjust the force of the cams by varying the lever length or linear spring. Also, this is an awesome idea and I am looking forward to more content on this.

[@SnakebitSTI]

A linear spring (or rubber band) could even be arranged as part of an over-center mechanism, so it could hold the gate in its open or closed position.

[@Thepaintedtiger]

You've got Brother ribber needles, and the ribbers do not pattern (unless you do it manually, or you can possibly count the tiny bit of EON patterning that Brother ribbers can do if they have lIlI buttons/knobs for double bed jacquard (however that patterning is triggered by the carriage itself, totally different from the main bed with punchcard or electronics). I believe that leg at the back is longer for general stability and leverage, as quite a bit of weight is often used when knitting ribbing, and there's also potential for other pressures like racking, cabling, quite tight tension, etc. You can extend the needle fully into E (holding) and the very back of the leg is still under the bottom of the bed, so held by the spongebar or needle retaining bar (depending on model) at the top working end, and the bed of the ribber at the bottom, making for very straight travel preventing metal on metal screeching if the needle butt could twist and touch the slots. People aren't generally going to be trying to apply such force to the LK150 (and indeed can't without breaking the plastic bits), plus the price point calls for reducing the more expensive bits, so less steel makes sense (and of course, they don't need a leg to push from behind with any type of automatic pattern device).

[@yttube4865]

respect the subtitles - effort shown

[@John-gw3mj]

I'm not a knitter but I'm glad KZbin suggested this to me, thanks for sharing such detail about an amazing project! Two thoughts that may or may not be helpful: Do you need torsion springs? Could you use a linear spring on top of the carriage instead? You might end up with some awkward binding but it might allow you to use a fully printed cam with a wider diameter (rather than the pin/rivet) Could you move the front rail screws behind the sponge strip? If you had "ears" on the back side of the rail, could you move the screws behind the sponge strip? This might interfere with the needle butts so it might need to be screwed in from the underside.

[@tfpanda9050]

I'm very happy to know that someone is working on this. I appreciate how detailed these videos are as well. Thank you for your hard work. I will be following this project with excitement.

[@michelefumagalli81]

I know that this could be a little of stretch but you could look into the needle that are made for industrial knitting machines made by groz beckert, they are much strurdier and reliable (also still in production) than the ones made for domestic machines. Plus they have an integrated System that allows the transfer of loops between two beds, and there are some variants that present a slide System that is much much more reliable than the traditional needle. Also for the gauge System you could research on the multi-gauge System of Stoll machines. Also if you need i have at my diaposal three different ShimaSeiki machines with their part catalogues if you need to take a look at the industrial machines for inspiration.

[@stevenfaber3896]

fork weights for the hair clips Great design! I always have trouble with flow on my USM for the wheel you could make it in the center and perhaps use the butts to push it along? but it's supposed to work without it, orrrr you can put the wheel at the back and have it be a combination with the conical or yarn ball design yarn holder, where moving the carriage spins out yarn at a rate from the ball and hits as a good row counter location

[@Falney]

I am in the process of developing a computerized Intarzia and Lace knitting machine. I am forgoing Cams for pnumatic or electric servos in the carriage to get over a lot of headaches involving cams and needle size. You just put the needle pitch into the computer and a bit of automatic math and step counting later, you can use the same carriage on any size/needle pitch bed.

[@bluephoenixcreations]

Incredibly well explained. Fabulous explanation of how/what the parts of the machine are for. I am also very impressed with your engineering/design explanations. Have you considered (some day) making a basic programming video? With this machine as the example we would definitely would have a reference for where and how the code does its thing. Thank you for sharing all of your trials and tribulations. I look forward to mark 4.

[@grott3n0lm]

I love how you're sourcing the torsion springs from hairclips, like McGyver would have done! I build the csm created by Steve Turner. As i didn't wanna order the springs from america to be shipped to europe, i wound them out of stainless steel heating wire from my e-cigarette. The look kinda strange and shurely don't have the same technical specifications, but work great. If i knew about your hack, i probably would have bought some hairclips and just bent them in kinda the right shape.

[@quasistoic-all-in]

I really love this project, and I so much appreciate all the detail you give on your design, development, and refinement processes. I’m a former software engineer by trade, but a 3d designer/printer, knitter, crafter, and tinkerer by choice/nature, and designing and building a flatbed machine knitter has been on my todo list as well. Even if you chose to never release your designs themselves, the lessons you’ve been imparting through these videos are invaluable.

[@MrGigs94]

Wow I found this series yesterday and then I see you uploaded less than 30 minutes ago. Ive been loosely obsessed with kntting machines for the last 2 years but havent been able to justify the cost of getting into them. I originally found knitting on a whim through loom knitting almost 4 years ago now.

[@sleepydrJ]

COVID crafter syndrome.

[@shoofle]

if you're looking for configurable spring solutions, you might look into using flexures, wherein you print thin bits of material and use the springy properties of the plastic to generate bendable and spring-like behaviors - they might be a great fit for switchable gates! you could also look into winding your own springs out of piano wire, which might be out of scope for this project but could be a solution. and could be useful for researching what kind of springs you want.

[@glenhawksworth5601]

I came here to say something similar. Try looking into "compliant mechanisms". Could be just what you need. kzbin.info/www/bejne/roeWZ3yKh8-DlZYsi=NQOusM4nIoZHHxfp Teaching Tech is a great resource (and the only channel I am currently subscribed to, other than yours). PS: my mum taught me how to knit as a tween, haven't knitted since but found your channel from a desire to return to knitting from a technical perspective. I really enjoy your videos from a design process perspective. I am a high-school IT/maths/science teacher and your process is a great example of what I've been trying to teach my students... I wonder if I could get them engaged with videos about knitting machines 🤔😂

[@doughntworry]

Love following your process.

[@SparksCuriosity]

I'm so glad! Thank you!

[@migsy1]

38:06 i think the captions have a bit of unused script left in. Amazing video series, I really love watching all that you’ve made so far!!!

[@SparksCuriosity]

Ah, good catch, thank you!

[@kristien2010]

I did not fully understand parts of this video, but watching you figure things out & beeak them down is so fascinating. I have a Bond USM that I love, but am excitedly following this project to see if I can print a slightly more advanced machine. Good luck with all the tinkering.

[@etienne6916]

super interesting! :) can't wait to try your machine out! I have a singer standard gauge and printed the Mr Roboto sock machine, so now I'm ready to try out a flat bed machine print haha

[@jacobdunning8373]

Mc-Master Carr can be a great resource for small parts like the springs and pins. If being able to handle them and decide is important, ACE hardware might be an option depending on where you are in Canada

[@nancymoore8949]

I really enjoy your videos and can sometimes even follow the technical explanations but will probably never build my own machine. But keep it coming! I am definitely curious.

[@DaveLeCompte]

yay, stubborn enthusiasm

[@tomweinstein]

For small deflections, torsion springs should operate the same in either direction as long as they have enough space for the wound section to expand/contract.

[@quirkydad6083]

Continuing to repeat in my head “we’re not going to start knitting our own socks and we don’t need any more hobbies…”

[@awwkaw9996]

I was wondering. As far as I can tell your cam was designed entirely by drawing. Did you end up also making the 3D version of it depend on the needle variables? It seems like that would be a great way of making the machine compatible with a wider selection of needles? Beautiful work, it has been really interesting watching so far!

[@joshmyer9]

13:03 If the carriage couples to the bed via a conical cylinder, you could get infinitely variable pitch by moving the point of contact along the cone (see 507 mechanical movements #8, but make the "belt" be a ridge on the needlebed, it's a slightly janky infinitely variable pulley). (This is a terrible idea but it would be amazing if it wouldn't wear out super quickly, and slip all the time, and and and. Terrible, but fun!)

[@buun1456]

💜

[@super_jo_nathan]

Loved hearing you talk about your process! Have you maybe considered 3d printed flexure joints for the gates?

[@super_jo_nathan]

I'm pretty sure that there are some bistable flexure mechanisms out there that could fit your needs very well as they don't require sourcing extra parts

[@SparksCuriosity]

Thank you, and yes! Since filming I've been looking into a few different compliant mechanism ideas that I think could potentially work but I haven't had time to test anything yet. Being able to replace the springs would solve so many issues if I can figure out something printable that's still robust enough at that teeny size. 🤞

[@super_jo_nathan]

@@SparksCuriosity ah yeah, the size constraint does add a lot of challenges. Stuff like that doesn't translate well to screens 😅. I do hope you can figure something out! It would be a very elegant solution if you can make it work

[@DarrousHadiLEGO_ROBOTS]

Hi, I am a mechanical designer, and I would be more than happy to help you to develop this machine if you want help with design, CAD, or part sourcing and making it more user-friendly.

[@regineb.4756]

I don’t know if this information is use to you: East Germany specialized in wool production and production of knitted goods. Unfortunately it wasn’t appreciated after the reunion and I watched a documentary of how many techniques, especially regarding knitting machines, are nearly lost.

[@LegoDan2002]

Microcenter has torsion springs in their 100 pc misc spring bags

[@claireoconnor6692]

It would be AMAZING to see a machine that could do garter stitch in the same way as a hanamaka or Pingouin machine.

[@claireoconnor6692]

The two flat beds (with a double ended needle) also make lovely ribbing.

[@rileyrustad5810]

Hey! I love your work, and it's exciting to find others with aligned interests! I've been working on a 3d printed knitting machine off and on for the last few years, and I think we're like 2 mirrored parallel lines here. I found this video and your motivations one, and it's spooky how close our projects are. I wonder if you'd be up for any amount of collaboration? I haven't posted any of mine publicly, but would if it's helpful?

[@SparksCuriosity]

Ooooh it must be something in the zeitgeist, that's very cool. I'd love to hear more if and when you post something!

[@MarjoleinVeenendaal]

What are some good resources on designing 3D objects with code? I’m a Python developer and I generally make models in F360, but I find designing in OpenSCAD really intimidating. As a side note, do you have a ko-fi or something? Would love to support this work!

[@enjoying28]

search for 3dpinted spring shapes for your cam and springs

[@wyw876]

From a programmer that's worked with mechanical engineers, one takeaway lesson is "if you can't make it as precise as you need, make it adjustable". Another lesson, from James Bruton here on YT, is to implement that adjustability via minimal-volume 3d-printed sleeves to abstract away the bridge between your mechanical need and your available materials.

[@sleepydrJ]

Your yellow and white components make me want one….. deceiving me into thinking it will all be friendly and easy. Please make your future versions puke-grey-green, or yell much louder with your “it’s not a product!!” section. Thank you.