The Complete Homemade Juggling Beanbag Guide

Search posts
Forum index


Joshua Clifton -

The Complete Homemade Juggling Beanbag Guide
A comprehensive instructional guide and technical treatise on designing and making spherical, paneled beanbags for juggling and footbagging

My website with free download links:

Reddit posts:
2nd Edition:
3rd Edition (latest):

Montage of my beanbag designs:

Hello, jugglers! I am posting here because I have written a document that might interest some of you. It is an extensive guide and technical treatise on how to design and make spherical beanbags for juggling and footbagging. It has been my pet project off and on for many years (it began in the 1990s). The guide documents and other files are free (donation requested) and you can read more about the project and download the files at my website: I would greatly appreciate any input and advice on this project.

I believe my patterns to be superior to all others I have encountered, including Marylis Ramos' patterns, except for her orange peel and 8-panel octahedron designs, which are almost identical to mine. There are comparison illustrations and discussions in the respective chapter documents.

For those who only want the patterns without all the instructions and other information, I added a compilation of just the patterns. That document is in the "Other Files" category of the Downloads section, and is called "Juggling Beanbag Patterns by Joshua Clifton.pdf".

The following information is excerpted from my web page.

I have tried to make this work accessible to readers lacking technical knowledge, making it easy for them to simply print the patterns and sew the beanbags without having to wade through technical information, yet also include all the information that will enable those with a mathematical background and interest in the geometry and design theories to delve deeper and satisfy their curiosity and understand how these designs are created. The boldfacing I use throughout the documents are an attempt to enable readers to scan the documents quickly and glean the most important information.

The original motivation behind this guide is that nobody (that I know of) provides definitions of the pattern shapes of spherical beanbags so they can be drawn in any size or improved upon. In the case of the typical 32-panel design used for footbags, which is composed of pentagons and semi-regular hexagons, nobody seems to have a good answer to the question of how to size the patterns to produce a desired finished size.

My guide answers that. Each beanbag design document not only includes ready-to-print patterns in six sizes, and instructions for scaling them for other ball diameters, but also formulas to calculate the pattern dimensions for any ball size, and illustrated instructions for drawing the patterns (by hand and with a CAD program). Each design also includes mathematical definitions and structural analyses of the pattern shapes (including four variations of the 32-panel structure), and explanations of how I developed the designs. In Chapter 2 there is a section on figuring out how much you need to adjust the pattern sizes to account for things like gather applied to the seams, or your material choices.

With the exception of the regular polygons, I designed all of the panel shapes myself using math and extensive experimentation. All designs up to the 14-panel use curved edges to produce better spheres, and most of the polyhedral designs have modified face shapes that produce better spheres. I discuss the mathematics and techniques I used to create the designs so that someone with the aptitude for it could follow my process to create new designs, or improve mine.

Examples of my color arrangement diagrams:

My guide documents include the following:

  • Information and advice on fabrics, thread, template material, filler, beanbag weight and size, fabric markers, stitching and knotting techniques, and finishing techniques. (Chapter 2.)

  • Ready-to-print patterns in six sizes for each design with instructions for scaling the printout for other sizes, formulas for calculating pattern dimensions and step-by-step directions for drawing them (by hand and with SketchUp), illustrated instructions for assembling the beanbags, and 249 illustrated color arrangement ideas including the balls and the assembly layouts (examples above), with 84 arrangements for the 32-panel structures. (In their respective design chapter documents.)

  • A list of other people's online tutorials for making juggling beanbags, footbags, and other fabric balls. (In the Introduction.)

  • Fabric ball project ideas with photos (Christmas ornaments, decorative centerpieces, baby toys, etc.). (Appendix I.)

  • Full, illustrated explanations of how I developed each design and the mathematics behind them, and comparisons to alternate patterns in some cases, including Marylis Ramos' patterns. (In their respective design chapter documents.)

  • Examples of other designs and variations. (Chapter 4.)

  • A chapter on the theories and mathematics I use to modify polyhedral face angles and to design curves for polygonal panels to produce optimal spheres. It includes tutorials on how to calculate "Isovertex" face angles and on how use the Tangent Chord Angle Theorem to calculate arc radii that produce specified tangent angles at their intersections. Accompanying the latter are explanations and examples of why circular curves do not necessarily work best, and how to design non-circular/Bézier curves that work better (particularly for the orange peel ball). (Chapter 5)

  • Step-by-step instructions for drawing spherical polyhedra in SketchUp. (Appendix II.)

  • An appendix illustrating how I create the HDR photos of my beanbags. (Appendix III.)

  • A list of the juggling beanbag manufacturers whose websites I used as resources for this work. (Appendix IV.)

7b_wizard - - Parent

Hi Joshua,
the first four shapes in the first picture, I have sewn them too. I now prefer sewing 2-panel balls - they've the least longest seam, thus less & faster work to do; but also no meeting points of seams ('T' or 'X').
Those patterns with sizing instructions of yours for the standard 4-panel beanbag might become really useful to me, as I have to go by trial and error for a differently sized ball.
I am, though, sceptical about the efficiency of using ever smaller and smaller upto 32 panels - so much precious lifetime goes into sewing panel by panel bay panel ... a.s.o. (x32) - while of course these then look absolutely greedily ornamental ;o)
Thanks for sharing.

Joshua Clifton - - Parent

Thanks for replying!

That's true. For pure stitching efficiency, the 2-panel ball is the best. I find it difficult to assemble, though, because I have to continually align concave curves with convex curves. So for optimal efficiency, I prefer the 4-panel orange peel ball, for which I can just lay the panels flat together and sew along the lines (it requires only 13% more stitching). My overall favorite, though, is the octahedron (8 panels, 59% more stitching), because it is still pretty easy and quick to make, and is very round and elegant. The 2-panel and orange peel designs, and even the cube, are a bit squarish with the fabrics I use, and I like a perfect sphere.

I'm guessing the lack of seam intersections is why the 2-panel structure is used for baseballs. It has no weak points and also has the least amount of stitching that could burst.

The designs with more panels are great for their beauty, and some people like to make them because they enjoy the process. But they certainly aren't necessary for juggling, and for me they are very tedious. I might make them if I really wanted an elegant set to juggle with or to impress others, but otherwise I'd stick with the octahedron.

The 32-panel structures (hexagon or triangle versions) are the ones typically used for footbags (Hacky Sacks). My understanding is that people use them because the dense seam structure forms a skeleton that keeps the ball round when half-filled and makes it springy and more responsive to kicks. It's not important for juggling except for those with a passion for beautiful and creative designs. I love the latest one I made, which is the one in the photo, but I sure wouldn't want to make a full set of those!

If you use my patterns, let me know if they produce the sizes I specify. I haven't had the motivation to test the sizing with other fabrics, so I don't know how accurate they will be for whatever you're using. I size them for corduroy, and I've been wondering if other fabrics will size the same way.

7b_wizard - - Parent

Oh yeah, right, sewn folded flat together makes it much easier!.

Have you ever made any out of genuine leather - there different thicknesses available - I had used between 0.7 and 1.4 mm thin\thick - they feel great to touch, to juggle, and just to know they will really last!

In matters of the outcome of the sizing for different fabrics, I think with the same sewing line, the outcome should pretty much be the same, unless maybe for really thick fabric or leather that doesn't easily bend when newly sewn, but should adapt with use.

Joshua Clifton - - Parent

I've never used real leather, but I made a couple balls with a material called marine vinyl that is 1mm thick and has the surface texture and general properties of moderately stiff leather. I made them for games of catch. We don't have baseball gloves, so I wanted a ball that had the feel and roughly the size of a baseball, but would be softer and a little lighter so we could catch it with bare hands (they're tightly packed with plastic pellets). They work great and are fun to throw, are very durable, and do have a great feel in the hand as you said!

Since the material is thick and stiff, I did not use a seam allowance on the panels, but assembled them edge-to-edge, right side out (I used a single-needle baseball stitch and very strong upholstery thread). Marine vinyl is not as tough as leather, and the stitches need a good bite into the edge so as not to tear through (I made that mistake on a few stitches).

The first three photos are of my 2-panel, baseball-style ball, and the fourth is of my dodecahedron. That was my first one and I don't care for it as much.

For future visitors to this thread, if the above photos are no longer available, you can see the first three in my 2-Panel Baseball Chapter document near the end of the "How I Developed This Design" section (currently page 25), and the fourth, the dodecahedron, in the 1 - Homemade Juggling Beanbag Guide... document, Chapter 2 - General Information and Techniques, "Fabric" section and "Stitching Techniques" section.

7b_wizard - - Parent

• for artificial leather - (of a good kind that doesn't easily rip) - I also found it pretty good and well apt for beanbags; just one property needs being careful about, which is that it melts on contact with fire or embers; which also goes for polyamid sewing thread.
• for sewing concave to convexe &v.v., I mark the four middles of the panels soas to keep adjusting these together while sewing. Or even in case do angled stitches.
• look great, your baseball style balls above. I think, with the middle somewhat thinner, they would become even rounder - the concave part then needs adjusted, though, for the circumference to remain same.

Joshua Clifton - - Parent

Oh, also, what patterns have you been using? I like to research alternate patterns and see how they compare to mine. Are they available on the web?

7b_wizard - - Parent

No, just private cardboard (lol).
For my patterns, I mainly went by cm of circumference needed, then trial & error for rounding the edges more or less well.
The oldest, classical beanbag's 1 of 4 spindle panels, I got by copying it from a worn off one cut open.
For the 2-panel patterns, there's no strict dimensions, but some variation possible (e.g. for how thin the middle). For the corners, I even optimized with a little 'S'-curve in each round corner of both panels; or call it a little mould. [ alas, I've long not made any and currently have no access to where they might be ] Roughly from memory: two-panel panel with moulds [80 KB]

Joshua Clifton - - Parent

LOL, I like the caption above the pattern! The S-curves are an interesting design idea. Does that make a rounder ball, or make the "corners" a little less pronounced?

7b_wizard - - Parent

hehe^^ yes. corners less pronounced (thus rounder ball). but it's experimental and maybe not necessary with well shaped or even mathematically calculated roundings of the pattern.

Joshua Clifton - - Parent

Try my patterns if you feel up to it and tell me how the resulting ball compares to yours. I'm not fully satisfied with the nature of the ball my pattern produces, particularly where you put those notches, and I want to improve the shape, using Bezier curves instead of circular. But I don't know how to design the curves. Maybe your little notches will be a clue to what I need to improve.

I've done some research on the panel shape and even unwrapped a baseball to see what I could learn from that, but I have not yet figured out what an optimal shape would be. I've found very little information on this shape, and there is a lot of variation between different patterns and between those and the real baseball panels. I think different baseballs also use different shapes.

barnesy - - Parent

That’s a lot of words! (He says, while probably launching into a lot of words). I’ve only dipped into bits for the moment but it looks like great work. I’ve never understood or explored the mathematical side of beanbag patterns, so it’s good to see an approach coming from the other end of that scale. Though I think a lot of the corner cutting decisions I’ve made make it a little incompatible with my approach.

I used to be known as a beanbag manufacturer. That was 20-odd years ago when the juggling shops didn’t sell underfilled juggling balls. That market has changed a lot! I was using a 4 panel beachball style pattern. It was never particularly round but that was mostly OK because of the fill level. I was probably too reluctant to share my patterns but it never felt like that was the important information for me to share.

These days, I wouldn’t consider anything other than 8 panels. My patterns have been through various size manipulations but IIRC they started out as Marylis’ pattern. I just have a load of pieces of cardboard with cryptic squiggles on them. I sew that pattern by sewing pairs of panels and then continuing with the 4 panel beachball style following the method I put on the web (more on that below). I once had an interesting chat with Gregory from GBallz: his 8 panel approach is quicker than mine. He sews pairs of panels, then sews two twos together to get half balls. He then turns one inside out inside the other (or was that the other way round?) and finishes the inside out ball by sewing those together as one seam. Sorcery! One day I’ll try that, but it doesn’t fit so well with my preferred sewing machine: a hand cranked Singer. The other thing that interested me from that chat is that he said he can make a good six panel ball quicker than an eight panel. Good for him: that’s not how my attempts worked out. I tried six panel rounded squares for one set. I like the result but I had to deal with eight vertices per ball. I found that a lot more difficult than (what felt like) two vertices for my 4 or 8 panel approach. The shape can be good with 6 but 8 works better for less effort. More than eight panels looks great but I’ve never tried it. Partly because of the increased time and difficulty but mainly because of the increased number of reasons for two balls to not be identical.

I see you offer sewing and (optional?) cutting patterns. When I started, I was using a cutting template but I felt that it took too much time for the extra marking, and that the precise positioning of the sewing template within that shape introduced possible error. So I switched to marking one shape and using the sewing machine foot edge as a guide for the seam position. I’m sure that there are issues with that, both mathematically and from an information sharing point of view. But it’s quicker, so for me it wins. Any spherical imperfection seems very slight at this size.

I see that you linked to my old guide on the 2diabolo site. That site is long dead and is now being squatted by people who think it’s fine to throw porn videos at whoever wanders in, so you may want to change that link! I found that out on my phone while sitting between my two small children who fortunately were engrossed in the TV. Thank heavens for Bluey! The best reproduction I can quickly find is here:

Joshua Clifton - - Parent

Wow, Dave Barnes of Barnesy Bags! What an honor to meet you here! Your tutorial has been part of the backdrop of this project since near the beginning around 2012! Nice work!

I updated the Barnesy Bags link to the archived version, and updated a few others in the list that were offline. The new version of the document is now on my website. I never saw porn at the old URL, though. Right now I just get a few brief error page redirects and then a blank page. I guess somebody evicted the squatters. Thanks for alerting me to that, though!

Yes, LOL, I have always been very verbose in everything I write. I enjoy meeting others who are the same. I know that it will repel many potential readers, but that's just who I am. However, it might serve a purpose:

I have been a bit concerned that my documents will to a small degree hurt the sales of footbags and juggling bags by the many professional makers, and possibly create new competitors. This may not be a bad thing, but I don't like to hurt small businesses making excellent products, or upset anyone. So perhaps the intense and lengthy nature of my post and my documents will at least filter out the casuals, who will instead continue buying their beanbags. Only the truely dedicated with time on their hands, or the empty of pocket, will make use of it, and those types ought to have a good resource to help them pursue and enjoy this craft, I think.

It's funny that you got started in your hobby/business the same way I did, by being short on cash as a teenager in the 1990s (my timeframe, anyway) and deciding to figure out how to make the 4-panel balls myself instead of buying them! (My origin story is in the 4 & 6-Panel Orange Peel Ball PDF, "How I Developed This Design" section.)

One advantage of the 8-panel over the cube, or any structure with 4-way seam intersections rather than 3-way, is that you can continue the same thread (if long enough) through the entire ball without re-stitching any seams. My assembly methods try to take advantage of that when possible. Tying off threads and starting new ones slows down the process. I don't know how it is with a machine, though.

That's cool that you got to interact with the (I assume) founder of Gballz! That's another long-standing side character in this project. His balls look beautiful ;-) I like his assembly methodology. That would work pretty well. It's very similar to how I assemble some of my panel structures.

As for stitching and cutting templates, I know what you mean. The additional alignment and tracing is tedious and time-consuming. But, at least for hand stitching, and especially for high panel count structures, the combo template idea I developed eliminates the alignment issue and even makes the tracing quicker. It's a stencil on the inside and an exterior template on the outside, all one piece, with tabs around the outer edge to hold it down with (the #1 root PDF, chapter 2, "Making Templates" section has a discussion of the three template types).

The interior tracing for the stitching pattern is easier and faster than tracing around the outside of a template, and the exterior/cutting pattern just needs a few sketched lines around the corners between the tabs. The template takes patience and care to cut out (requires an X-Acto knife, and I haven't even tried to make one for a curved-edge pattern), but with a durable plastic it lasts pretty much forever, and it's a great way to draw lots of accurate panels quickly.

Giocoleria da diporto - - Parent

Hello, first of all thank you for this monumental work on making homemade beanbags.

I haven't read it yet, but seeing the size of the PDF and the date you started, as well as all the constant updates you've made, you can tell how much time and inventiveness you've invested in all of this.

I therefore wanted to ask you:
- if you think that even now, with all the numerous models on the market, even quite similar to the homemade ones in terms of shapes, partial fillings and type of fabrics, it makes sense to produce them yourself
- and if you have ever compared your creations with those of the most used brands (Flying clipper, GBallz and all the others).

Thanks again

barnesy - - Parent

It may be pride but I like the ones I make more than any others I've tried, and I'm free to experiment. People seem to rave about ultraleather but I don't like it so much. Back when I started making beanbags I think I had a bit of a niche in the market. These days it's completely different: I imagine I could be happy with quite a few of the available options. They'd all be good at gathering dust!

Little Paul - - Parent

Learning to sew and making your own juggling balls can be as engrossing and enjoyable as learning to juggle in the first place.

It’s not for everyone, but there is absolutely lots of value in making your own props, way beyond the end product.

7b_wizard - - Parent

shipping is expensive.
numerous, but different and very distinct models with other properties than what you need or like.
you want the fill like you want it and the weight that you prefer - and that's where the weight per volume per ball size come in play.

7b_wizard - - Parent

...and also the sizes are standardized. near to no room for own preferences of endless possible combinations of size + material +weight + density + surface + squishyness. always a pain to have to buy balls.

Giocoleria da diporto - - Parent

How many balls do you make yourself? Because to know which size, material, weight, density, surface and squishyness are perfect for you too many attempts you have to make…

7b_wizard - - Parent

I'm not sure, I understood this right.
My best selfmade are of leather, just the cherry pips inside broke, so they are now way too flabby - I'd have to refill them.

"knowing what is perfect for me" - I need 6.0 cm, 80 g, but standard is 6.2 cm, 110 g (Play in Italy online offers to fill e.g. mmx according to your wish).

I saw a Drop Props ball, pretty much my preferences (strong fabric, slightly underfilled, plastic beads filling) and not expensive, but that's US with if at all, expensive shipping to europe.

I bought cheap 6.0 cm beanbags, but they are actually really 6.2 :(

"many attempts" - .. ahm ... yes, I do.

Giocoleria da diporto - - Parent

So you could try the Infinities, size XS (61mm), filled with plastic if you want balls at least 75% filled up or plastic and iron if you want them less than 75% filled. And you decide the exact filling percentage.

Look at this graph:

Obviously this level of customization means that they cost significantly more than other beanbags on the market.

Giocoleria da diporto - - Parent

Let it be clear that they are not the only ones which allow you to choose the exact weight of the balls. I have indicated these only because they are the ones with the highest degree of customization, but I have never used them: I currently juggle some GBallz I bought years ago.

Giocoleria da diporto - - Parent

you want the fill like you want it and the weight that you prefer
This is currently possible with several manufacturers.

7b_wizard - - Parent

Yes, it's pretty new service. But it works mainly for hybrids only and with notably less filling makes them 'russian'.

Giocoleria da diporto - - Parent

Take a look for example at these Infinity beanbags:

You can choose:
- color of the panels
- thread color for stitching
- thread thickness
- size, between 7 diameters
- weight
- filling material (they use two different types of plastic and iron: they avoided millet because it can give allergies), with filling tables that give indications on the weight based on the percentages of fillers chosen.

7b_wizard - - Parent

Thanks for taking the time.
New information for me - I didn't know about Infinities yet.
So, what I see is 18.90 € for one custom ball; seen handmade and guarantee and all that goes into it, service, know-how, --> a fair price, but still luxury; advanced jugglers might need more than one set of upto many balls which makes it a little fortune or call it investment.

For the first selfsewn ball ( after a longer time not making any ), I need 2h or more. A few balls later I have sped the process up to like 75 minutes. Sacrificing a few evenings, I'll have a new super set within less than a week. Assuming, my time were worth 20-30 €, the value of my balls would be roughly the same as bought customs.

Joshua Clifton - - Parent

Thank you very much for your kind words :-) Your appreciation means a lot.

1) There are three reasons I can see for making one's own juggling bags (the others have mentioned some aspects of these): A) You're poor and don't want to spend the money on a good set, but are adept at crafting (that's why Barnesy and I started). B) You want to customize them in some way that the manufacturers aren't doing. C) You love to make things yourself, and, as Little Paul said, you enjoy the process, creativity, or zen of it, and the satisfaction, pride, and bragging rights of having made a beautiful set of beanbags yourself (especially the high panel count designs!).

2) I have never owned any pro beanbags apart from the original, cheap, Jugglebug cube bags I got when I was very young. They were being sold at a shop where I attended a few juggling lessons. So I do not know how the pro bags compare to mine except for what I can judge from photos and descriptions.

The one exception was many years ago when I attended a juggling function and tried a set of pro bags somebody had. I don't know who made them, but I remember them being single-color corduroy, 14-panel. They were superior to mine in weight, which is how I learned the importance of weight, and started adding metal BBs to the plastic pellets I use. It really improves them. I also preferred the corduroy to the denim I was using and switched.

I have no doubt that many of those companies make products that are at least as good as mine. The only reason mine might be a little better is that I can take the time, if I care to, to perfect the stitching, knotting, and other assembly so that the bags are very durable and elegant. Most mass producers probably cannot take that kind of time - it would make the bags too expensive. (I guess that's a minor 4th reason to make them one's self.)

Giocoleria da diporto - - Parent

Thanks for the reply. As far as I'm concerned, I believe that reasons A and C are clearly the main ones, given the fair level of customization achieved by some manufacturers.

7b_wizard - - Parent

The only customization I ever met is weight of the millet filling by italian manufacturer Play for hybrids only. It's still millet thus a different volume then. The ball size also stays the same.

You can customize e v e r y t h i n g when making them yourself!

Giocoleria da diporto - - Parent

No, currently there are several manufacturers that allow you not only to choose the weight (and not only filling with millet, but with plastic granules or metal microspheres or a combination of these, in order to ensure every possible weight and every desired subfill), but also the fabric used.
Ball sizes are usually only 3 or 4, rarely more.

7b_wizard - - Parent

...yet another example: I bought seven of such smaller 5.sth cm 4-panel beanbags in a toy shop; I even tried them out. Now, what happened: of the green, blue, red, white and yellow panels, the yellow ones soon lost their coating leaving the bare fabric. Now the balls are eggshaped :(

Giocoleria da diporto - - Parent

Of course, as in any market there are bad products, medium quality products and good products.

RegularJugular - - Parent

I am glad such comprehensive documentation on making beanbags exists. The next time I try to make beanbags I will refer to this.

Thank you


Subscribe to this forum via RSS
1 article per branch
1 article per post

Forum stats