Universal 3D printing filament spool standard 2014

I have had some very interesting discussions regarding actual spool manufacture, from spool suppliers and material providers. It sounds all very viable to make a low-cost, strong and lightweight spool in a variety of ways.

We should see some ideas from 'them directly' in the next few weeks, so we need to focus on the discussion of dimensions so they and anyone else interested have something to work with / agree on.

It's starting to look like a outer dimension of around 200mm and inner core of 100mm or more with a 50mm mounting hole would cover a lot of machines already out there, and also many existing and new materials too.

General consensus from talking to various people seems to be that a narrow spool is better than a wide one, for better feeding and general handling both during manufacture and use bu the customer. - Anyone got any views on that?

If you go for a narrow width (~50mm) you will just about fit ~750g of material on a spool. (Again the 'spec' could just state max outer dimension of 220mm or whatever)

What amount of filament do people like? - for me 750g to 1Kg is about ideal, below that it's a pain and above to about 1.5Kg is okay, but not much more.



And as a side note - I mentioned on my blog that Makerbot seems to have a new size spool and interestingly they also seem to have gone narrower (now 40mm wide instead of 76mm) and bigger diameter (250mm rather than 200mm), I can't tell what the internal filament diameter is (does anyone have a new Makerbot spool?)

---

[richrap.blogspot.com]

Quote
nothinman
Don't get me wrong, I'm not criticising your design, it's actually pretty interesting. It's just that it's actually hard to make tooling for it and it requires a lot of assembly.
If you're making one spool then it's fine. If you cut it with a laser -- even better.

But again: why would YOU want/need to make spools?

Marcin

I don't want to make spools (beyond proof-of-concept, anyway). I just want to provide a reference design. I think having a design for a decent-quality spool that can be made by anyone with access to a lasercutter and packing tape is worth spending a little time on.

---

[whosawhatsis.com]

Quote
richrap
What amount of filament do people like? - for me 750g to 1Kg is about ideal, below that it's a pain and above to about 1.5Kg is okay, but not much more.

I do the majority of my printing in white and black (boring, I know). I like 5lb (2.25kg) spools of filament because it results in less empty spools floating around my garage (why don't I just throw them away?) and less mounting and un-mounting of spools for my printer. However, Ultimachine seems to be the only US-based company to carry spools that large (anyone know of others?), and they always seem to be out-of-stock of the 5lb spools when I need to order some. So, I usually end up using 1kg or 2lb (0.91kg) spools since that is what is most commonly available.

Edited 1 time(s). Last edit at 03/19/2014 12:24AM by NewPerfection.

---

Help improve the RepRap wiki! Just click "Edit" in the top-right corner of the page and start typing. Anyone can edit the wiki!

Quote
richrap
What amount of filament do people like? - for me 750g to 1Kg is about ideal, below that it's a pain and above to about 1.5Kg is okay, but not much more.

Outliers like the above comment notwithstanding, I think 1kg is the happy medium.

Actually, if we're making a wishlist, I'd like to have filament sold by volume (not length, though length is proportional to volume for a given diameter, regardless of density) instead of weight (mass). 1000cc is a nice, round number. That would be 1.25kg for PLA and 1.04kg for ABS, and it would mean that (assuming the packing density on the spool is calculated correctly), you would know exactly how much space you need to hold that filament regardless of the density of the material. In practical terms, volume would probably end up being determined by dividing the mass by the density, though also measuring length would be a good way to verify the average diameter (by calculating cross-sectional area, though this is no substitute for actually measuring diameter to check it for consistency).

---

[whosawhatsis.com]

Instead of making an additional standard , why not to try to use existing spool designs and define one/two (1kg/2.2kg) as standard ?

Chri

Edited 1 time(s). Last edit at 03/19/2014 01:20PM by Chri.

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools

Quote
Chri
Instead of making an additional standard , why not to try to use existing spool designs and define one/two (1kg/2.2kg) as standard ?

Chri

Existing spool designs do not solve most of the problems and are usually not very recyclable.

Even many of the filament suppliers I have talked too, say spools are too heavy, take up too much room and they don't really want to ship spools that can't be recycled or disposed of in a eco-friendly way.

So we may as well select what works best for most users and 3D printer designs because the filament manufacturers and spool suppliers are already receptive to many of the problems being highlighted, and want to help.

---

[richrap.blogspot.com]

i rather meant the measurements, so the dimensions not the material, this would be another different point

Finding one single design for all Printers won`t be very easy, people with small printers prefer small rolls (mobility), people with large printers large rolls.
So if there only should be one size, what would be the optimal amount of filament per roll?
Also for example if i want to try out a new supplier i would prefer rather less material than more , so if the roll is crap there is not so much money wasted.

Maybe it sould be splited in different tasks ?

* Payload ? what would be ideal ammount of filament for most people ?
* Dimensions? what would be the minimum optimal inner diameter for example ?
* Material ? rolls are all basicly the same, so how it is made is a thing of its own
* Features ? holes for filament to put through while not using the roll ect.

Chri

Edited 1 time(s). Last edit at 03/19/2014 07:32PM by Chri.

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools

Quote
Chri
Finding one single design for all Printers won`t be very easy, people with small printers prefer small rolls (mobility), people with large printers large rolls.
So if there only should be one size, what would be the optimal amount of filament per roll?

Well, there are diameter constraints that affect the usability of the filament. Spool it too tightly and it won't straighten back out easily without breaking, at least for the more common materials, so the diameter should be a constant. Changing the width is safer, and like the idea Richrap posted, you could make a cartridge that would hold one full-capacity spool or three 1/3-capacity spools (some space is lost to the extra spool walls). Perhaps it we should define one "full-width" design for 1kg, then say that other sizes can be used. For instance, a 1/3W version for a 1/4kg spool that you can fit 3 of in the space of one 1W spool and a 2W that should hold about 5lb (2.27kg). Then a machine built for a 2W spool could hold 1x2W, 2x1W, 6x1/3W, 1x1W+3x1/3W, etc.

---

[whosawhatsis.com]

BTW, in this case, you probably wouldn't want to have empty space on your spool holder, but it would be easy enough to print a dummy spool hub in one of a couple of standard lengths (or cut a length of an existing tube of the right diameter) to fill it up. Allowing the width to be completely variable would be more problematic.

---

[whosawhatsis.com]

Quote
richrap
General consensus from talking to various people seems to be that a narrow spool is better than a wide one, for better feeding and general handling both during manufacture and use bu the customer. - Anyone got any views on that?

And as a side note - I mentioned on my blog that Makerbot seems to have a new size spool and interestingly they also seem to have gone narrower (now 40mm wide instead of 76mm) and bigger diameter (250mm rather than 200mm), I can't tell what the internal filament diameter is (does anyone have a new Makerbot spool?)

These are important questions, I think. I'm building a small maker supply company. We buy filament from manufacturers in bulk spools, then manually transfer to smaller spools. I've tried lots of different spool types, and I tend to prefer larger widths (aka a larger "Traverse"). The reason is it reduces the number of times the filament needs to cross back over itself. When handled poorly, these cross overs backs can create tangles that impact usability. This is a very small point, however, and (IMO) should not be considered above something like feeding.

Makerbot's recent move is important, too. As the largest manufacturer of desktop machines, most manufacturers and suppliers will need to have a product that is compatible. Amazon describes the dims in this way:

Spool Diameter: 9 in (~230mm)
Spool Width: 2 in (~50mm)
Spool Hub Hole: 2 in (~50mm)

I can order one if there is interest, and document the dimensions.

I thought it might be helpful to at least share the terms used by spool manufacturers. These could help make sure we're speaking about the same thing. The diagram below (from TW Metals) is pretty good.

---

makerstash.com
@makerstash

If we are talking about a wish list for filament spools, I'd very much like to second NewPerfection's input for 2+kg spools, for the very same reasons he mentions.

---

--------------------------------------------------------
3D extruder performance tests
dzach's ORDbot Hadron build

I agree with the standard, but as mentioned there may be problems identifying the original source of the actual filament. In the electrical field, materials are marked with an OEM origination labels which identify the maker of the actual part and that of the packager or reseller. This would be an idea to investigate as it would help in identifying the source of materials no matter whose spool it is on.
Another idea would be to use "Split Center" spools, which separate with a quarter turn, such as are used for wire. This enables supplies to be purchased in coils (lower cost) and then transferred to feed spools for working and eleviates the need to constantly pay for a spool which will just be disposed of once empty or pile up in the corner. There are alot of techniques used in the electrical field that with a little tweaking can easily be adapted to our arena so as to make our lives easier and suppliers more accountable.

Edited 1 time(s). Last edit at 03/23/2014 09:40AM by eades850.

eades850 has got a point here. I, too agree that the less spools made the better - however, tangling of loose coils might get problematic as pointed out before.

I am unsure about the environmental impact of cardboard spools, though. IIRC, it also takes a lot of energy, water and nasty chemicals to recycle cardboard or paper in general. Not to bash or badmouth your idea, Rich, but I would like to see whether using cardboard really is the best approach or if there is a better solution.

An alternative could look a bit like the PET bottle recycling system: you return the empty spools and get a discount on your next purchase (or return the spools to makerspaces/FabLabs that handle sending them back in bulk?). For this solution, the material of the spool doesn't matter that much. A spool that you can disassemble to occupy as little space as possible would probably be favorable.
Also, the obvious penalty here would be carbon emissions from shipping the spools around. The advantage of less manufactured spools vs consumed fuel has to reach a toppling point - the question is: where exactly is this point?

A followon to uGen's comment would be to use a biodegradable/compostable core tube of set diameter and span without sideplates, the coil being shrinkwrapped or tiewrapped to it. This core/filament package could slide onto a split-apart spool prior to unwrapping. This would combine the best of both worlds concerning production costs while minimizing disposable/returnable materials.
You can go to Lowes or Home Depot and get an idea of this technique by looking at coils of shrinkwrapped Romex residential wire.
We take these coils and slide them onto split spools for coil perimeter feeding, or lay them on a "Spinner Table" which is basically a big Lazy Susan when we want to feed from the coil center. Both techniques work very well, but my preference is the spinner table as it keeps the coil packaging intact. Using this technique with shrinkwrapped filament would help minimize dust accumulation from static buildup during use and would eliminate coil free-wheeling which can occur when perimeter feeding. And, if you have any old 16mm theater film cans you have a ready-made dust/moisture proof storage container. (Yes, I'm that old).

Hey, this sounds like a very good compromise!

The problem with a return system is logistics and hassle which might drive people to not return spools, so minimizing the spool (or the part(s) of the spool that has/have to be shipped) sound really reasonable to me. I wonder how the shrinkwrapping might be implemented best? In any case I can imagine, handling has to be extra careful when unwrapping...

Shipping on the core only also seems to address several problems mentioned in this thread:
- large spools are not always ideal. When you can customize your spool and only ship the core with a user-specified amount of filament, flange size becomes a non-issue?
- global "standard": I guess you can find cardboard tubes everywhere - or get them manufactured to spec fairly easily. To compensate the bore/hub difference in diameter, each user can print an adapter that fits their own needs best (or this might even be integral to the flange parts).
- less wasted material, less shipping weight
- spool traverse only depends on the core. The flanges are user-supplied and should fit to cores of arbitrary length.

I would support this kind of standard!

Quote
whosawhatsis
Perhaps it we should define one "full-width" design for 1kg, then say that other sizes can be used. For instance, a 1/3W version for a 1/4kg spool that you can fit 3 of in the space of one 1W spool and a 2W that should hold about 5lb (2.27kg). Then a machine built for a 2W spool could hold 1x2W, 2x1W, 6x1/3W, 1x1W+3x1/3W, etc.

Exactly, this provides quite a lot of flexibility for machine designers and also volume of filament.

@Makerstash -

Quote
Makerstash
"larger widths (aka a larger "Traverse"). The reason is it reduces the number of times the filament needs to cross back over itself."

- I was told by a few filament suppliers the exact opposite of this, a smaller width causes less issues with tangles.

@Dzach & NewPerfection - Bigger spools seem to be less popular, and most of the filament suppliers are looking at 1Kg as a 'standard' but allowing a double width spool could get to 2Kg as whosawhatsis mentioned

@eades850 -

Quote
eades850
"This enables supplies to be purchased in coils (lower cost) and then transferred to feed spools"

- I do like the idea and that's exactly where I started from - can we make a re-loadable spool? but quickly this looks less and less likely as you start talking to suppliers and thinking about usability along the chain.

@uGen - Yes there is a better solution. Cardboard has issues, and already one supplier has come up with a very good 'almost perfect' solution to this, so I am much less concerned about the manufacturing of the spool, and still want to focus on the dimensions.
The supplier will want to present the idea themselves to the community, so hang on a little and see what you think. I believe it solves almost all the issues, of being lightweight, eco-friendly and can be manufactured anywhere in the world.
Everyone I have talked to agrees that spools are not going to be shipped all over, they need to be made locally, and this needs to be an open-source and open-standard design

@eades850 - see above.

I do believe we may have now already solved the how to manufacture it question, and also what material to use.

The critical things we need to agree on are -

'Bore' - That's looking like 52mm is a good option
'Hub' - 120mm is what I would like to see as a minimum diameter
'Flange' - Maybe make this a maximum dimension of 200mm?

'One' standard width could be 50mm.

Volume of filament? ~ 1Kg Max on a 'standard' 1-wide spool?

Agree / disagree ?

Does anyone have any other requirements or reasons why they want/need something different.

How we can accommodate different volumes and types of materials is a good focus point.

Does anyone know of the minimum bend radius for filled filament materials - wood / chalk / mineral / carbon / metal / etc? or can anyone find out. The ones I have experimented with have a 160mm+ internal diameter

---

[richrap.blogspot.com]

RIchrap, are you sure about the 120mm core thing? I looked at some of mine, and they had cores closer to 100mm. This is a standard size for cardboard mailing tubes that would make a good source of hubs for these spools.

---

[whosawhatsis.com]

Quote
whosawhatsis
RIchrap, are you sure about the 120mm core thing? I looked at some of mine, and they had cores closer to 100mm. This is a standard size for cardboard mailing tubes that would make a good source of hubs for these spools.

I also have lots of spools with around 100mm cores, and they still make me feel a little uncomfortable with hard 3mm materials. I would like them a little bigger, but that's only my personal opinion, it's more about what's appropriate for most materials being used and thinking about the future. And of course all the other things like taking up space etc. so I expect we will end up with something closer to 100mm in the end.

I'm not so concerned about being a standard size of cardboard tube now as other production methods are looking much more likely to make this a reality.

---

[richrap.blogspot.com]

One solution would be to contact every filament manufacturer involved and see if we can obtain a 5 meter section of 3mm filament of each type..ie Taulman618, laywood, pla, asa, abs etc. as 3mm will have the largest minium radius when compared to 1.75mm filament. Then either ourselves or by a contracted lab determine the minium radius before unrecoverable diametric/linear deformation. then go with the greatest diameter +10% for the standard hub diameter. As for the other parameters, I agree with RichRap...


'Bore' - That's looking like 52mm is a good option

'Flange' - Maybe make this a maximum dimension of 200mm?

'One' standard width could be 50mm.

Volume of filament? ~ 1Kg Max on a 'standard' 1-wide spool?

as an aside, when we get core wound wire, it is wrapped with a long strip of clingfilm, such as is used on shipping pallets to contain its contents. This works very well to contain and protect the wire while making it easy to unwrap once spooled by just finding the film end and peeling it off in the same manner and direction as you would unroll the wire, or in this case filament.

My thoughts for what a standardized spool should look like

It needs to have an inside diameter of over 3 inches, any smaller, and from my experience the filament has a higher probability of getting tangled. I really dislike cardboard spools because they are flimsy and do not work with all methods of holding filament. I think that it should be about 8 inches wides, and 8 inches in diameter. This allows it to carry a touch over half a kilo of filament. I like smaller filament quantities better because it allows you to procure many different colors at around a comparable low cost(about half the cost). The the split-up nature of 3d printing creates vast number of supplier for filament and 3d printer is it is hard to put in place standard. I do believe that is someone was to make a standardized role(meaning they would have to front the money). Then, sold it to the filament manufactures at just a tiny bit above the cost to them. That they could institute a standard because corporations usually pick the cheapest option. Once, a fair number of companies start using the standardized filament spool, 3d printer makers, could start making it where only the standard filament spool fits inside of their printer.

Why would you need a standard hub?

Only dimensions I see worth standardizing are the bore and possibly transverse.

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

This is the design I'm going for:


Bore: 52mm
Hub: 100mm OD, 5mm wall
Flange: 200mm OD, 3mm wall
Width: 50mm ID

This has sufficient capacity for up to 1kg of ABS, which is the lightest.

Marcin

I bought one of the new M@kerb0t spools of filament, just to make sure we had the data. Here are the measurements:

Bore - 55 mm
Hub - 88 mm
Flange - 250mm
Traverse - 30 mm



Pictures are attached to the post.

There are a couple of things about the spool that are worth noting:

1. It has a recessed bore on one side. The recess is about 12 mm deep. Based on the pictures from the M@kerb0t website, I'd assume this is used for mounting.
2. The side of the spool (opposite the recessed bore) has open channels that allow you to see the filament. This could be stylistic, or it could allow you to monitor how much filament you have while the filament is in the machine.
3. The label says the spool holds 0.9kg (2lbs).
4. The spool is marked. with the statement - "Patent Pending." Not sure what this means exactly, but we might want to understand how this impacts our work.

It works for me. My only question is the material. Are you looking at a cardboard/cardstock composition or plastic? Hence, is it reusable, returnable, recyclable or what?

Quote
eades850
It works for me. My only question is the material. Are you looking at a cardboard/cardstock composition or plastic? Hence, is it reusable, returnable, recyclable or what?

If you're talking to me -- it's stiff cardboard. Can be reusable, you can return it if you wish, or recycle.
I'm also thinking of making them out of plastic too -- PLA, or something biodegradable; but will start with cardboard...

Marcin

If you use cardboard, how do you ensure that the roll is running smooth at the outside ?
Alot of people are running the rolls sitting on four bearings (Prusai I2`s , Mendelmax, ... )

Chri

Edited 1 time(s). Last edit at 04/04/2014 12:34AM by Chri.

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools

Quote
Chri
If you use cardboard, how do you ensure that the roll is running smooth at the outside ?
Alot of people are running the rolls sitting on four bearings (Prusai I2`s , Mendelmax, ... )
Chri

What do you mean? It's not a car wheel, it's turning at a rate of maybe one revolution per minute... if that...
You can make the core out of concrete and it wouldn't affect anything

Unless you meant something else?

m.

Quote
Chri
If you use cardboard, how do you ensure that the roll is running smooth at the outside ?
Alot of people are running the rolls sitting on four bearings (Prusai I2`s , Mendelmax, ... )

Chri

I think he is referring to the flange plates and not the core. Cardboard flange plates have a horrible tendency to warp and wrinkle as they age giving the people who use the 4 bearing roll bases fits if they do not transfer to a plastic spool.
An idea for this would be two removable flange plates that snap into the 52mm bore and are slightly oversize say 240mm. These could be transferred from spool to spool as needed and overcome the warp problems by applying outside pressure to the spool flange walls, thus making them run true.

Yes eades850, this is it what i meant.

Chri

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools