DIY Extruder Screw
OK. After the discussions on this thread about adding compression to a wood auger by welding material into the root of the auger to create the compression zone, I came up with an alternate idea to use the wood auger as the basis for a machine to make an extruder screw from scratch. I thought I would start a new thread rather than taking the existing thread any further off-topic.
To recap, the reasoning behind my idea was that the pitch of the wood auger is already acceptable as it has already been proven to work, so instead of trying to use a wood auger itself as the extruder screw, it could be used as the basis for a machine to make an extruder screw instead..
Having already looked into making an extruder screw in my machine shop, I realised that that manufacture of an extruder screw is not easily possible without some form of NC machine, which is not something that is accessible to a majority of the wider PP community. Even with the machinery I have at my disposal I could not make a screw without first making some form of machine or machine aid.
So if it was possible to make a simple machine, to make a better extruder screw, from widely available materials with hand tools, then it would help to improve the efficiency and quality of extruder output for the wider PP community.
The basic proposal is to use the wood auger as the transport mechanism (lead screw) along with a grinder following a profile template to make an extruder screw that has a compression and metering zone. I outlined my ideas on how this might be achieved here. Of course, there are also many other ways to achieve the same, and the purpose of this thread is to explore how this might be made to work and to provide a place to collaborate and document development.
I have started to gather materials to make a prototype and should have a basic first-blush together over the next week or so.
Please feel free to continue the conversation already started on the previous thread, and also contribute with ideas, critiques and your own creations.
I had some time to mount the grinder today (after I fixed my welder 🙁 ). I ended up having to raise the auger up from in-between the rails of the base as the cutting wheel did not engage deep enough into the workpiece. So I used this opportunity to simplify the ‘nut’ design. It is now a simple round bar sitting across the top of the auger. It works very well. No bearings, no complicated design. Much easier to build. The mechanism also still runs pretty smoothly, not as smoothly as the previous design, but I think that it is not important.
So for anyone who was having trouble visualising how the grinder was to be mounted, now it should be clearer to see 😀
Next step is to finish off the cam profiles and workholding and then I will be able to make some tests.
You can read more info and see some more photos here – https://deeemm.com/index.php/entry/general/diy-extruder-screw-making-machine-part-2-the-grinder
There’s also a sketchy video here – https://www.youtube.com/watch?v=kHPv1I7LjN0
I wonder if you could adapt something like this https://www.harborfreight.com/electric-chain-saw-sharpener-63803.html . It seems to have many of the functions. The wheels are probably too thin and expensive but maybe change out the wheel.
OK. Some progress.
I’ve made the basic machine with moving table / carriage. This has proven the concept of using the auger bit as a lead screw. The results are quite good and should easily be good enough for stage 2 which is to develop the tool and workholding.
I have (for the most part) followed the philosophy of readily available parts and construction with hand tools.
If you are interested in a step-by-step explanation take a look at the write up at the following link – https://deeemm.com/index.php/entry/general/diy-extruder-auger-machine-part-1 There’s also some learnings and caveats that are worth considering if you want to make your own version of this.
Here’s a short video showing the operation of the carriage.
I wonder if you could form the nut for the wood auger by self tapping it (perhaps heated) into a block of plastic (hdpe?). The block could then straddle and ride on top of a box beam or upside down channel….just some brainstorming
Off on a tangent, I was trying to understand whether incorporating a vent in the extrusion process would improve quality (reduce voids?). I found this https://www.ptonline.com/articles/extrusion-know-how-the-whys-hows-ifs-of-vented-extruders which does not appear to recommend a vented approach.
So some further thoughts on the design for the machine.
I like the suggestion of having the pattern and work piece in line as it provides for a simplified machine, however it does limit the workpiece length to the same overall length as the auger.
I also managed to do some quick tests with creating a ‘nut’ to work with the auger so that it could be used as a lead screw. The solution is rather simple. By using two rods of a suitable size, one above and one below the screw, mounted perpendicular to the direction of travel, the screw can be tracked fairly accurately and without play. The rods ideally need to be mounted in bearings as the rotation of the auger bit causes the rods to rotate. On the test I carried out, I found that for a 12mm auger, two 13mm rods were required. (see sketch).
Workpiece holding could be one issue if we were to employ a chuck and centres as it adds complexity to the design, however it should be possible to allow the workpiece to rest on pairs of bearings (one pair at each end). The workpiece is mounted to the end of the wood auger using a simple coupling – the same style of coupling that will be used in the final machine.
The possible inclusion of additional bearings on top of the workpiece should also be considered as the forces from grinding would have a tendency to make the workpiece want to move.
The entire assembly could be mounted within a simple RHS frame where the table / carriage is mounted to two drawer runners (These are basically cheap and accessible linear bearings).
Profile guides are mounted to the outside of the RHS.
Tool deflection is a concern. I think some form of guide rails to the inside of the profile guides would be a good idea to minimise this. This does then pass these forces into the workpiece, which will then have a tendency to want to rotate, which is also another design constraint.
The entire construction is possible with the use of hand tools, without even the need for a welder. Of course this is not the easiest of best way to make this machine, but the fact it is possible is good.
thanks a lot; i was about to do that 🙂
that’s what i came along so far. I am putting this all into the library :
1. Lathe only : target screw=22mm pitch,motor=800W-1500RPM|3Phase|1:30. duration=2 days, skillset=advanced, result=excellent, output=5×4 beams with good surface and consistency, status=documentation missing, tested: yes, there are 15 happy extruder owners out there
2. Additive Wood Auger Compression: Hack: 500W-1500RPM|3Phase|1:30, skillset=novice, output=waiting for tests from client
3. Bronze cast: skillset=advanced, no details so far
4. New & in progress: simple DIY mini lathe like device; stepper & leadscrew to move a slide hosting a modded bench grinder
I don’t know what the v4 product designers have on mind in terms of beam quality nor the v4 extruder specs or how to source those parts needed but I guess I keep bugging them 🙂
Materials I have gathered so far:
>Drawer guides (for table movement)
>Metal Hinge (for angle grinder attachment)
>Auger screw (for table transport)
>Angle grinder (I already have this)
Materials I (may) need to get
A 1000 – 2000 grid polish is doing alright too though – you can do this with a finger sander or grinding paper – manually – that’s good enough to send the screw to hard chrome plating / nitrating (recommended, not cheap) 🙂 read more about :
– the ultimate intro : https://www.youtube.com/watch?v=GridDLZ56yQ
For garage projects you can get a away with an SDS drill hack 🙂
Interestingly, I came recently along a nice topic – thanks to the incompleteness and exclusivity of v4 😉 : making ‘screw tips’. I am opening another post about this. Looks like that you can compensate screw issues by making a custom tip – or even stack them up. I need 2-3 more lifes just for testing it 😉
Nice design. I think if you were to reduce it to basic functionality the cost would be pretty low. All the 3D prints and bearings are nice but could be done in other ways. Printing the timing pulleys and the grinder mount is convenient. Here the grinders can be bought for $15. You can skip the chuck if you are willing to size the hole in the timing pulley to your part.
The Robodigg parts are nice, but if you need something local.
Do you really need super polish? Isn’t there always going to be shear within the plastic? Of course you could use polishing wheels on the grinder and use a router speed control (<$20) to slow down the grinder.
Interesting, I have SDS concrete bit that is waiting in the project cue.
Happy put a cap on a year long topic, somebody did it 🙂
The actual article has been posted in full length and a nice list of research references here (or check hackaday). Unfortunately the files are still incomplete but I keep watching out for the rest and add it to the library.
The build involves several hundreds of dollars, 2-3 times the price of an extrusion screw on Robotdigg.
Be aware that a good screw needs a certain surface polish to prevent plastic from sticking … as mentioned, be prepared to polish it up with 3000 grid paper, all day long 🙂
Yes the RHS frame that the grinder is mounted to slides within the larger RHS mounted to the slide allowing for the grinder position to be adjusted.
The grinder can also pivot a little on its mounting bolt to compensate for any angular deviation between the wheel and workpiece as the grinder frame pivots.
That looks great. Do you have some axial adjustment for the grinder to set the channel width on the screw (presumably wider than the grinder wheel thickness)?
Yes, I think that any grinder could be made to work. It’s just a case of presenting the workpiece to the wheel in the correct orientation.
I’ll try and make some time this weekend to mount the grinder
@deeemm, Possibly you addressed this in your design but are your sure about the angle grinder? the discs wear out in no time and you have to continuously correct depth of ‘grind’ which is ok with the xy vise but yet the procedure invites more mistakes… I found today a number of smaller bench grinders and the appropriate grinding discs are also cheap on the net.
My plan is to use an angle grinder – either 4 or 5 inch size – just a small hand held one. These are available everywhere. They also mostly all have screw mounts on the sides for the handle (also usually on both sides for left and right handed people), so it should be possible to make a mount that holds it. I plan to hold it in a vertical orientation.
I have also seen 3D printed mounts for similar tools. Worst case scenario is simply to use one or two large worm-drive universal clamps. I have done this before when I mounted a router to my plasma CNC. It works pretty good. Lots of options depending on what people may have available. For me I want to try to avoid using my 3D printer on this build as it is not tech that everyone has access to. Just basic tools.
nah, i just did hold the thing in place to see how much shatter and cleanup work it creates; there aren’t smaller grinders here in the shelf here; i am still puzzled how this can go; more search on amazon may give me some options … vacation is in the door step; best luck to you 🙂
one thing is for sure: the grinder is best on a cross-slide vise (30$) with a way to align the angle to the flights, easy with some set-screws and the arrangement you have already , yep.
Cool. One thing that I was planning when I get the time to make the rest of the device is to angle the grinder wheel so that it aligns with the angle of the flights, and also to choose a wheel that matches the width of the channel (if possible). If the wheel is narrower than the desired channel width then the wheel or workpiece will need to be reset.
Do you have a link to your video?
Is delrin a thermoplastic? could be a good way to create the nut if it can be molded.
great stuff indeed. just some hints, folks use ‘delrin‘ also for mill & lathe lead-screws. it’s self lubricating and has basically no play; that can last for decades; I am considering it for making a screw based injection; anyways, in your case that looks alright; machine build around the job 🙂
s2019 Sorry I missed your comment on making a nut from plastic.
Yes, I think that this could also work. I don’t think ‘drilling’ through the block would work, but I bet you could make the nut in two halves and press them together around the auger – much like the split nut on a screw cutting lathe. These two halves could then be mounted into a casing.
3D printing is another alternative, although some consideration of the design would need to be made.
An issue with trying to make a single design to suit all augers is that I think that no two augers are probably the same. Not only would the size differ (obviously) but I would imagine augers from different manufacturers would all be slightly different too.
Is there a way to embed videos?
Yes, I agree. I think that the finish will require several stages with final hand finishing / polishing. This method will take a long time and will never replace other faster manufacturing methods, but it does offer an opportunity for people without access to a machine shop to be able to produce a better extruder screw.
Adding a radius to the root of the spiral is easy, it’s just a case of shaping the grinding wheel. In fact if you look at the ideal profile for the spiral, it is non-symmetrical. The pressure side of the flute is squarer.
I think that another factor will be changes in diameter to the abrasive wheel. it will wear as the job progresses which will affect the finished dimensions. So as with any dimensional grinding, the grinding wheel will need to be dressed / replaced / reset as the job progresses.
@s2019 I think that the air is simply pushed back up towards the hopper / load end as this is effectively a lower pressure area within the extruder.
hey, I still have trouble understanding your machine but here a few things which I think apply to whatever way you carved out the material and assuming you want a good finish and be fast.
– it’s easier to polish if the inner angles of the channels are roundish; that also reduced the required motor power- there is a lot of grinding involved (see ‘long video’ in the auger hack thread):
1. angle grinder with grinding paper disc; that takes off the most of the shatter marks; it’s also ideal to widen channels in the feet zone quickly; runs best at 70 RPM
2. hand belt sander; you need around 10 belts for 65 cm; this gives it pretty fast a good surface; runs best at 300 – 500 RPM
4. a round file to blend in messed up areas; that can save a lot of work
5. lots of grinding paper – good one; you cut them in strips and fold them and grind the channels with 2 thumbs pushing pretty hard into the channel; 2-3 hours; at 800 RPM; up to a 1000 grid
– the more easy to have lots of feeding volume is to widen the feed zone’s channel width; I start grinding at around 6-7 mm flight width; in the feed zone it’s just 2-4 mm left; and in the compression around 5. that way you don’t need to cut too much root material which becomes pretty difficult to reach at some point
– it’s easier and gives better results to focus on each section individually instead of making cuts/grinds along the entire screw (from right to left):
1. feed zone (left)
2. meter zone but keep blending into the feed zone
3. compression zone, same as the meter zone; you blend in also in cuts into the meter zone.
In total 2-4 hours just for this part. I am uploading a timelaps tomorrow;
I would imagine it is hard to get an exact spec as from what I understand commercial screws differ according to material and application. I’m aiming for the typical load / compression / metering zone design for my application.
The ideal compression ratio will also depend on the size and quality of the granules. If they are large and irregular there will be more air to expel from the mix which will require a higher compression ratio.
I’m trying to 3D print directly from granules and so the quality of the output from the shredder / granulator along with the characteristics of the extruder screw are very important in being able to get a repeatable and accurate output from the extruder.
For example the load zone screw root sizing is critical in being able to actually get granules into the extruder, with the regular wood auger the channel size is fairly small, in fact not much bigger than the granules produced by the standard PP shredder, so being able to make this larger is a real benefit in my project, but possibly not so critical for larger auger sizes where the channel size is considerably larger than the granule size.
I’m actually having to design a new shredder / granulator for my project to more accurately control the granule size. It is an interesting journey.
Strange, I cannot see my reply, but when I try to repost it tells me that I have already posted.
You must be logged in to reply to this topic.