Screw Worm drive filament extruder Bowden

Screw/ Worm drive filament extruder Bowden

Screw/ Worm drive filament extruder Bowden

Screw/ Worm drive filament extruder Bowden

byGadgetGuru314is licensed under theCreative Commons – Attribution – Non-Commerciallicense.

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

A 3D printer filament driver for a bowden setup utilizing a wood screw and geared stepper motor. This drive unit is capable of provding significantly higher forces than a traditional filament extruder because of the increased number of engaged teeth and the mechanical advantage of the worm gear/ wood screw. The wood screw is designed to be easily replaceable by unscrewing the tensioner bolt and poping out the screw, then putting the spring on the new screw, compressing the spring and inserting into the slot. The spring pressure on the screw holds the driver securely into the screws head to prevent slop.

The filament enters the top of the assembly through a filleted hole, passes in-between the spring loaded hinge and a wood screw, forcing the filament to travel the length of the wood screw and out the push to connect fitting. The wood screw is held in position by a compression spring and is supported by two 1/8 rods.

notesome images are slightlyoutdated.

This is an experimental filament driver based on Werner Berrys concept described here:

1x 5 long, 1/8 diameter steel or stainless steel rod for hinge

2x 1.12 long 1/8 diameter steel or stainless steel rod for screw support

1x compression spring. McMaser 1986K930 for hinge pressure

1x Compression spring McMaster 1986K710 for securing screw

1x 2in wood screw (sharper threads the better).

1x Matching 1/4 hex driver for chosen wood screw

shaft coupler for motor to hex screw driver

1x McMaster 91251A875 Tensioner Bolt for hinge

1x washer for tensioner bolt and spring

2x thin profile nylock 1/4-20 nuts for lever arm and tensioner bolt McMaster 90101A007

1x push to connect tube adapter (filament exit) McMaster 5779K117

1x shoulder screw for lever axle. McMaster 91273A301

4x. your choice thread tapered head mounting screws for assembly

4x M3 stepper motor tapered thread screws.

1x geared stepper motor. I chose the Kysan 1040212 Geared Stepper Motor. 26.4Kg-cm. NEma 17. for $45 at

Warning: Due to the relatively small bend radius the filament needs to follow, there is a chance of PLA filament breaking. If this happens, please let me know and Ill increase the bend radius in a bug fix.

180 degree filament curve added to help prevent filament twist.

Worm_Drive_Assy_v2.5_PAck_1.11.13.zip

Worm_Drive_Assy_v2.5_Solidworks2013__1.11.13.zip

Screw/ Worm drive filament extruder Bowden

byGadgetGuru314is licensed under theCreative Commons – Attribution – Non-Commerciallicense.

By downloading this thing, you agree to abide by the license:

Creative Commons – Attribution – Non-Commercial

Upgrade this Thing with Thingiverse Apps

Auto-magically prepare your 3D models for 3D printing. A cloud based 3D models Preparing and Healing solution for 3D Printing, MakePrintable provides features for model repairing, wall thickness…

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in…

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

Quickly Scale, Mirror or Cut your 3D Models

I had the same epiphany driving home from work today and quickly googled it when I got home, and here we are 🙂

There are three points I would like to elaborate on,

The larger the diameter of the screw, the less the angle of engagement. This will allow the screw to be supported and have a clean exit hole towards the nozzle. Plus a straight entry, no bend on the filament. A larger screw needs to have the thread ground, or precision turned so the pitch remains constant to prevent stripping the filament.

Use of a flexible drive cable to a remote servo. Heck, you already get substantial mechanical advantage with an auger, lets loose the weight while were at it. Since the screw is larger diameter a female square drive can be added to accept a cable.

Use a 120 degree V channel opposite the auger. It can be hinged to allow easy access to load/remove filament. In addition can be either spring loaded or screw tensioned. A steel plate with the V can be nickle boron nitrided for the best possible wear resistance and lowest coefficient of friction. Another benefit is that the two point contact will minimize twist imparted from the screw rotation, which should also be coated so the only advancing spiral drives the filament, as the forces will be mostly a belt/ladder effect.

I appreciate your design is cheap, I hope to improve on the concept and make it real.

EDIT: Oh yea, needs a solenoid for retract, with a variable stop for stroke. A screw will never be fast enough….

not so new idea its is used on the old Rapman extruder

Okay… What about some kind of razor blade? It could be used to prevent filament rotating, I think

Good idea. It might help a bit. My primary concern would be with the shear strength of the plastic not being able to take the force from the razor blade, resulting in shavings piling up in the drive system. But if the razor blade was distributed over a sufficiently long piece of plastic, it shouldnt be a problem!

what about using 2 screws – one left and one right, to prevent filament rotation?

Hi Renha, I did consider using 2 counter-rotating screws to prevent twisting, but with the configuration this device is in, I would be worried about any gear backlash associated with coupling the two screws. Any amount of backlash would likely result in the destruction of the screws or the gears as well as contribute to extruder inaccuracies. If two screws were to be used, I would further develop the counter-rotating screw adaptation under remixes. I thought that was very clever. However, the backlash issue needs to be addressed some more.

Socket head wood screw with flats grinding to the shaft of the motor?

The wood screw is coupled to the motor via a short driver bit found at hardware stores. There is also a shaft coupler which joins the bit to the motor shaft.

Interesting note, this has also been done by Stratasys, and I imagine others. See this patent:

Figures 11 and 12 show a similar assembly built into a quick swap extruder head.

Youre right! Big similarity. I know the concept has been around for years but didnt get much attention due to twisting problems. My design is special with the twist in the filament path. Im curious how stratasys addressed it.

Not sure, but I imagine a screw mounted at an angle equal to the pitch of the threads, which presses into the filament above the worm drive would work to prevent rotation.

On another note, I have been looking into designs for multi extruder assemblies and I saw this design:

I like the idea of a rotating turrent because the motion could also be used to pull the nozzles up out of the way by using a track which pulls each nozzle up as it rotates out of the home position. A wiping assembly as shown in this design could also be included to prevent any drooling, although it could consist simply of small pieces of metal at each of the park positions rather than a long track as shown in the linked design.

What I was thinking was that you could use a worm drive such as your design, with a spur gear on the end of each drive rod. These would then index, along with the nozzle, to a worm gear on the drive motor. The double reduction from the worm drives would drastically reduce the torque required to move the filament (although high speed extrusion might be a problem especially with stepper motors). For the indexing system, a geneva gear system could also be used to reduce the precision needed for the indexing motor (a simple DC geared motor could be used, with a tiny sensor to detect when one rotation has occured).

I do not have the time or the capabilities to properly develop this idea, so I am suggesting it to you in the hopes that maybe it interests you and you can have some fun designing and implementing it.

Interesting concept and I see where youre going with this. I can see how being able to change nozzles (or colors) on the fly would be useful in a print that requires varying accuracy at different locations (ie the fill versus the perimeter). Compounded by the use of only 2 motors, it would be light weight and would operate at higher speeds.

I do have some reservations with the design though. Im concerned the mechanical complexity is too high to be practical. I wouldnt recommend rotating the entire hot end assembly because maintaining its position in space relative to the print within a few thousandths of an inch is paramount. I also foresee repeatability being an issue with the nozzles lifting out of the way each time, it will be difficult to get their height exactly* right each time it is lowered again. A piece of dirt or crud could really mess the system up. I like your idea of using a worm gear, but its been expressed to me that using worm gears alone for an extruder isnt optimal because its difficult to retract fast enough to prevent drooling. The geneva mechanism for indexing the nozzles is also clever.

When you first described the concept, it reminded me of this:

I like your idea, -it has several clever elements to the design. But, I think the design above would be more reliable than a turret design. Keep those creative juices flowing!

I agree that the rotating turret design would require very high precision to ensure that as the each nozzle came into position it fell into precisely the same position each time. I wonder if a guide could be built into the design such that as the geneva gear rotated to the lock position, it pushed the hot end down into a holder which would ensure precise positioning. Unfortunately I dont think this design would work well with 3D printed parts, and would require high precision machined parts to work best, then again if you relied on a spring loaded mechanism to hold the nozzle into the guide, you might be able to have some slop in the system, but still maintain good precision. The hotend and worm drive would have to be as small as possible, as the entire assembly would have to be spring loaded, similar to the patent I linked earlier.

The major use I see for a system like this is for multi material and color parts as the ability to lift the parked nozzles would help prevent damage to the parts and drooling. Also it would be relatively simple to build in a sort of capping mechanism for the parked nozzles to prevent any drooling as well. This would circumvent the need for nozzle priming and wiping towers and ooze shields. Another advantage is that as you add more nozzles to a system, all of the offsets either reduce the printable area of your printer, or require a larger positioning system for the same size build plate, while with this design the active nozzle is always in the same position, and as long as the assembly fits on your system, you will still have the full build area. One final caveat, you would have to ensure that the turret never rotated more than one rotation though or you would a problem with the multiple filaments tangling, but this could be handled relatively easily in software.

Hi, Im borowing this idea to make a Delta printer at my university, what I would like to know is: why you made two 3mm channels on top of the 1.75mm channel for the filament? Also, do you believe there would be any problems modifying this to fit 3mm filament?

Hi Xaliax, the two 3mm channels on top of the filament are to reduce friction between the hinge piece and the filament so the moving back and forth doesnt build up heat and fuse the filament and hinge piece together. I see no reason the design cant be modified to handle 3mm filament. However, just a word of advice- 3mm filament is becoming outdated, 1.75mm filament is becoming dominant because of its ability to extrude smaller volumes per linear increment (due to cross sectional volume). As a result, its becoming increasingly difficult to find new or replacement parts for 3mm filament. If you still want to use 3mm filament, let me know and Ill post another version for 3mm filament. i dont think the filament channel in the hinge will comfortably fit 3mm filament.

Just a heads up, this was done a long time ago in the RepRap community, like…2008. You should check out this history of reprap: — it causes twisting in the filament, and generally gave people a bunch of problems.

sorry for sticking my nose where it does not belong, but I like the design, the only thing that im not sure is, where would you put that drive extruder on the printer, because the filament input angle is the same as the output. maybe the filament roll should be in an upper position? for example lake this image.

Na, I like noses- as long as theyre not too pointy.

Your point is valid. The photobucket image you linked is similar to what I had in mind. This extruder was designed for a delta printer with both the filament and filament extruder on the top of the 3D printer. The Filament was intended to be mounted roughly in front or above the extruder. As your drawing indicates, the tube is intended to leave the extruder and go up and then back down to the extruder head. Obviously not all 3d printers have the same architecture so I guess well just have to make it work with other printers. If it becomes a big problem, ill make a bug fix.

Thank you for the heads up. Thats a very valid concern and I can see how it would be problematic. Ill update the design with a bug fix shortly.

While this design isnt optimal or minimal in a few areas, its designed to be built with no minimal machined parts (short of 3-d printed parts). The shaft coupler between the hex driver and stepper shaft could even be replaced by a Jiffy Weld bond. My only concern there is if the shafts arent perfectly in line the bond might break from sheer stress.

Thank you all for your suggestions. Ill post a revised version within the next few days

along with what JAP said, i think that you may want to put a larger angle on the filament as it comes in, as it is it makes a pretty large bend that will jam a lot. looks fantastic and will definitely be making one once there is another one out there!

Thank you for your feeback Tom. Increasing the bend radius will reduce stress on the filament (although with this configuration I doubt jaming is possible. Id imagine the filament would break before the drive slips) My concern is the entrance angle. If the entrance angle is too severe, there will be stress there as well. This all depends on where the spool is located relative to the drive unit.

Bits From Bytes (BFB) printers has used worm drive extruders for years.

Further, the RepRap folks did it even before them

While im aware this is not a new concept, using a replacable wood screw is a unique solution. This design also has fewer components than the bitsfrombytes design and this one is meant for a Bowden Setup

Wow amazing concept, I think, probably, this model would be more stable and precise, but not sure about the part of the lid, as the movement of the filament would make the canal filament deeper even if it fits with the adjustment screw as the first extruder, maybe if the lid has 2 or 3 lined bearings for wear would be less.

Sorry for my bad grammar i hope the idea will be clear

Thanks for the Feeback JAP. youre right, it wouldnt be a bad idea to reduce friction between the filament and the hinge. Ill add two more steel rods for it to ride against in the next revision.

Hello again, i was checking, and thanks for taking my suggestion, but i have a doubt, my mechanical is not good, I hope it could be understandable, lets say that the extruder is aligned on the X axis, the filament is outing on this axis, but if im correct the rods wouldnt rotate because they are aligned too on the X axis, maybe if you use pellets or bearings aligned, the filament would have less friction

I dont think the rods will rotate when the screw is turning because the joint is so snug on the bottom and the filament wont be rotating so the rods arent even in contact on the top.

Screw/ Worm drive filament extruder Bowdenby GadgetGuru314 is licensed under theCreative Commons – Attribution – Non-Commerciallicense.

Say thanks by giving GadgetGuru314 a tip and help them continue to share amazing Things with the Thingiverse community.

Were sure GadgetGuru314 would love to see what youve printed. Please document your print and share a Make with the community.

To post a Make simply visit this Thing again and clickI Made Oneto start uploading your photo. Its even easier to post a Make via the Thingiverse Mobile app (available viaGoogle PlayandApple App Store).

Were sure GadgetGuru314 would love a Tip, but unfortunately they have not setup their accounts to receive them yet.

We can let them know you were interested. Simply click the button below and well pass along a note to them.