Home injection molding part 3: Making stuff! (kind of..)
Вставка
- Опубліковано 15 тра 2024
- Check out my 2nd channel, TT Racing: / @ttracingyt
It’s time for some home based injection molding. I’ve had this machine since 2014, but it has been sitting idle for a while now, due to an error on my part. Previously, I attempted to mold some bespoke silicone washers to seal the machine, and did manage to mill a two part aluminium mold ready for recycled plastic. In this video, I make some further repairs and alterations and finally put through some shredded PLA with frustrating results. Is it worth more modifications and testing? At the moment, I think not.
Please note that this machine is sold to suit 110V mains voltage. In Australia we use 220V, and I’m pretty sure I removed one heater and converted the other. If you also use 220V, you would need to do the same or use a step down converter. It’s also worth noting that this machine was never intended to be used with this type of plastic. If I were to stick to the original Plastisol material, it would likely have been flawless.
0:00 Introduction
0:42 Recap
Part 1: • Home injection molding...
Part 2: • Home injection molding...
2:09 Silicone o-rings (again)
2:50 New PID controller (with autotune!)
My new controller from eBay: www.ebay.com.au/itm/304598395639
3:47 Additional chamber heater
4:13 Hose repair
4:28 The compressor gives up
4:42 Machine modification to use a plunger mechanism
6:54 PLA tests and (poor) results
PLA injection molding guidelines: www.totalenergies-corbion.com...
10:02 Virgin ABS (worse) results
11:10 Hot glue test
CNC Kitchenm hot melt glue filament video: • Can you 3D Print with ...
12:10 Evaluation
Recycled plastic sheets video: • Melting plastic waste ...
Industrial injection molding video: • Technology of Injectio...
Buster Beagle 3D: www.busterbeagle3d.com/
13:25 Conclusion
Buy quality and affordable filament from X3D. Buy 3, get 1 free and a free sample pack with every order: www.x3d.com.au
Get Quality Resins from 3D Printers Online. 5% off storewide for Teaching Tech subscribers [Code: tech5]
3dprintersonline.com.au/
Take a look around and if you like what you see, please subscribe.
Support me on Patreon: / teachingtech
I used to work in injection molding for nearly a decade.
somewhere north of 70% of the energy to heat plastic comes from shear friction in the compression zone of the screw.
That machine you had was relying on the thermal conductivity of plastic to "soak" the heat until. You probably degraded a bit of the plastic near the barrell sleeve, which probably made the part brittle and also chemically "stickier" (degradation via hydroxyl boding or crosslinking) to the mold surface.
Drafting and surface texture are very plastic specific -- it is possible you made a mechanical "puller" with your dremmel on opposite mold surfaces (think, negative draft angle around a feature) and that could be what caused the part to stick. I would want that texture of the surface to be at least sandblasted, if not smoother -- preferably semi polished to a 400-800 grit sandpaper.
Usually you want two molds if you're using a desktop machine to reduce barrel residence time to reduce degraded plastic. I would say maximum 50%/50% prime/regrind ratios for strength, but honestly probably 70% prime is the easy sweet spot for strength.
A Problem though, is you're overcooking the plastic.
With the setup you could probably turn your regrind into a smaller diameter extrusion and using the extruded rod directly, like you did with your hot glue gun is a lower tech alternative to direct injection. But there is nuance to how you'd want the plastic to flow in a reducing barrel diameter, like a longer landing zone to make sure the shot has enough thermal mass to fill the cavity.
I would kind of like to see some of this industry knowledge propagate to the 3d printing community at a hobbyist level; as I know that it circulates in industry-relevant circles, but is also extremely useful when considering 3d printing constraints.
Duuuuude, I think this is the first time you’ve used this kind of humour in a video. So nice to see your personality shine a bit mate ! Loved it !
That felt very much like a Top Gear cold open. I like it.
I wasn't sure if people would make the connection, but figured doing a Clarkson voice would be too weird. Glad you got the reference.
@@TeachingTech eyyyyy! Keep hiding the top gear/grand tour reference Easter eggs. I'll be looking for them!
Thank you for being one of the public pioneers in this subject. It takes people like you to inspire others. With enough interest, people can really make a change in the way we waste/recycle plastic for the future.
PLA is really difficult to inject. We have an injection molding machine at our university, and somethimes I make it run on ground PLA parts. With the PLA's really low glas transition you have crazy sink marks, then the part is too soft to eject, or if you cool it sufficiently, it shrinks and this render the ejection difficult too. It is great because is emphasize the effect of parameters such as temperature, pressure, cooling etc so it helps student to understans what's happening during the process. But if you want a good part, it is not easy at alls, and this with a small but indutrial machine. So I'm not surprised that you are encountering huge difficulties with this hobby machine that may already be difficult to use with "adapted" materials. Sill it is a really interesting project.
Look at John SL-Random Products channel for good info on injection molding and yes, Buster Beagle. One of your issues is that you are using a machine that is not deisgned to do what you are trying to do with it. You need a smaller heating chamber and load. All that is happening is you are heating the outside of that giant slug and not getting a good enough internal seal to push it out. Injection molding happens at higher pressures than you can generate with your set up. There is a reason your machine only uses Plastisol! It would be just as frustrating to try and use your blender as a microwave. 😀
That's a good summary (unfortunately).
You may have suffered through this, but I consider this one of the best videos you've made! Especially the subtle humor when something didn't go as planned.
Enjoyed your comment 'character building' . You have a good attitude and the patience of Job
If you're not aware of the channel "Brothers Make", you should probably look into them if you change your mind on stopping this project.
They're not the most educational channel, theyre more about making safe youtube for the kids than the technical aspects.
@@tenabrae thanks! ...i was about to use much less appropriate words! :)
@@simonschneider5913 haha honestly it's a shame because they cover a lot of topics you don't see a lot of info about, and they obviously know what they're doing.
@@tenabrae hmm, if thats the case, I will take a look!
This video is really timely for me! I've long fancied the idea of casting stuff in aluminum. I've melted down many cans and cast ingots in preparation for it... but like you 1) I don't actually have a real project to cast in aluminum, 2) I don't want to spend the money to get a "legitimate" setup, and 3) I really don't have the space to store all the tools and equipment I'd need to be successful at it. I loaded it all into the bed of my truck and took it to the dump this morning... and this afternoon, your video came out. Like I said... timely.
Must have been so frustrating, well done on sticking with it as far as you did!
Never laughed so much. Really enjoyed your journey!
Michael, your perseverance is inspiring even if the results are underwhelming. It's painful to watch someone I admire struggle so much but this was ultimately an informative and inspiring video.
Spent 7 years building plastic injection molds. Ya, these results are to be expected when trying to DIY the process at home
something ive been meaning to ask for a while but keep forgetting, what do you think of the way that brothers make do it? obviously HPDE/LDPE and PP is different than ABS and Nylon, but I think their stuff looks mighty decent for DIY
@Mobile_Dom do they ever mention what kind of injection molding machine they use because it looks like it works really well.
For complex stuff, a DIY approach will be a tough hill to climb. But for simple parts it can we work. We started with an lns manual injector, and I made molds on the Tormach. We made about 1000 parts that way. I then engineered a conversion of that system to use pneumatic cylinder for injection, and had it all controlled by an arduino mega that tracked cycle time and operated all the valves. That was perfectly reliable and we made another 6000 parts (with one mold update) that way.
The problem was injection volume. Cycle time was about 90 seconds because enough plastic had to melt to inject, but then it injected over 50% of the barrel capacity.
So we engineered a system from scratch with double the barrel volume and then the cycle times dropped to about 35 seconds because the plastic from the top of the barrel was as in there for more than 2 cycles.
Anyway, we made a ton of parts before stopping that business. For us, approximately 7000 parts with a net proceeds of about $8 a part. Not bad for under $2509 of home brew plastic injection.
Another great video Michael. Love the work and inspiration you give the community regardless of one's status, ie beginner, intermediate or advanced
Plastic used in injection molding have a high mfi value. That’s where this HF filaments are coming from. They use a higher mid value so it will be more fluid when melted.
I had been following this with interest as I do have a project for injection moulding. Thanks for your efforts and honesty about the process. I guess I stick to 3d printing the parts.
That's good to document the full journey with all the frustrations. It's how actual science & engineering usually goes : lots of frustration, problems within problems within problems.
Witchcraft! I was fortunate enough to be shown around an injection molding factory about fifty years ago. The factory had a dividing wall. On one side was the molding machines, operated by unqualified staff (probably replaced by robots now.)
On the other side of the wall were the Tool and Die makers... multi-talented, qualified, articulate and helpful. The machinery was mouth-watering, even including things like EDM machines. While there were CNC machines, they were not like today's perception. I had a good look at an auger (material feed screw), it had a conical core, and the 'threads" had a varying pitch. As the charge of plastic became more liquid the screw moved it at an increasingly higher rate so it would flow all the way through the mold before it became too solid. As I said, Witchcraft! 😜
Thanks for another fascinating video. 👍
*** I just remembered: On the wall in the foyer of the factory was their logo printed around a thirty-six inch micrometer. The micrometer was a gift from an engineering supply firm who placed an order for "Thirty 6-inch micrometers"... they received the 36-inch micrometer instead! OOPS!! 😧 😵💫 🥴
Yeah, pretty difficult stuff, Michael. I hope a solution appears someday.
Anyway, stay safe there with your family! 🖖😊
I'm sorry that things aren't working out for you as well as you had hoped but I love your sarcasm when describing how "happy" you were with the failures. I had a good chuckle. Thanks and I hope you don't give up too soon.
11:00 forbidden rice
Jeez Michael, 11/10 for perseverance.
well, you tried.
thanks for sharing your experience with all of us. The good ones and the bad ones 👍😀
To get silicone O-rings quickly - get a silicone hose and cut it to slices.
I did this once, and it was so satisfying I spent half an hour slicing the whole hose into rings. Then I had a sudden flashback to when I first got braces and was handed that baggie of tiny rubber bands....
The injection molded hot glue looks very promising, unironically
"character building" XD That's one way to put it! :D
Thanks for the Video Image, with the Key Message right in the Image. Thanks
Nice top gear reference
Another great one. Thanks for sharing the pain :)
A thought I had that might make opening the mould faster is milling the shape of the nuts on one side. At least a recessed bit that can hold that side for one handed screw removal
Well, I'm glad I decided against trying more ways of recycling plastic after watching this.
I don't have anywhere near the waste quantities as you do and melting it into sheets works quite well for me.
I even add back the waste from the cut sheets too.
Just for reference, the injection molds I've seen are polished to an almost mirror finish to avoid sticking. But it looks like that machine is your major issue.
Make a mold to duplicate the glue sticks make them from your PLA and use these in the hot glue gun to inject your molds. The idea of the Hot Glue Gun is pretty cool. Thanks for all your videos, cheers!
You have a funny idea a fun Michael. Very entertaining video, thanks.
You should build the buster beagle one just to see the right way to do it on budget, I did the cheapest and simplest MK1 and it worked right away.
Even if you have nothing that you need injecting, you should do it for sake of relief, experience and knowledge.
silicone: I had better results using ABS (which naturally doesn't stick to silicone) that was soaked in soapy water. You also have to wait extra long for the high temp silicone to really dry which is longer than the 1 day. I also clamp the mold shut with bolts with serious pressure so there is almost no trimming. 1mm bubble holes are needed; like those many cylinders you see on a new tire. also soapy blades slice it better but i use trim scissors.
This feels like a case of "i don't know what I'm doing so i'm gonna do it my way"
Not quite. The trouble is this started ten years when there were less options available for DIYers. The machine was never meant for this, hence the troubles. If I was starting again with something suited to the job, PLA properties would likely be the only significant hurdle.
Wow, you do have determination - one of the companies I worked at for 14 years had an injection molding facility - from what I saw the whole process is not for the faint of heart even for some of the mid-size companies like that one.
You could try removing the part at around 50-60c so the PLA isn't already brittle.
3 world records within one project is *impressive* to say the least!
What helped me the most in casting silicone sealings, suction cups and airtight parts is using sharp (~30°) mold edges around the sides which help to cut the thick paste flashings clean where the excess material is squeezed out. If you need really sharp edges on your silicone part you can use multi part molds with flat contact durfaces which aren't perfectly airtight preventing air pockets to appear.
I worked as a tool and die maker many years ago. We made plastic injection molds for the toy and aerospace industry. Because they were built to make thousands of parts they were made of steal but aluminum should work for small runs. On all the molds we made the mold cavity was mirror smooth aiding in the release which seems to be one of you biggest problems. If you ever feel the need to try this again you may be able to use some of the electrical parts but throw away most of your current machine. I'm looking into making a mold for a project and have found some great recent UA-cam videos. You have great content, thanks.
awesome, I waited so long for this
let's gooo new part of the series 😄😄
You should add Jack bolts to separate the mold halves
Interesting. Your persistence is inspirational. You might want to mill pry slots into one half of the mold. This is done even with professional molds.
There are mini desktop injection mold machines that act more like a production machine (screw material drive, actuators for opening and closing a mold). Of course they still arent nearly as cheap as your setup so it would be a hefty investment
I think I've compromised the whole process by relying on a machine purchased ten years ago that was never intended for this job.
Not sure why you were still short shotting even after adding a vent hole. Maybe material was cooling faster than you could inject it? This guy seems to be having success with a similar setup: ua-cam.com/video/wMRSPXt48CI/v-deo.htmlsi=NctHQqJkZ0PkROJo
I have not read thru the comments but after watching this video I would say a bunch of things. Just like with your 3D Printers there should be a "melt zone" and a "not melted zone..." (clever, I know) and the same thing needs to be seen in the injection molding process. There are some folks here on YT that have indeed built successful injection molds so I would recommend to take a peek at what they have done.
That said, I would say that having such a large heated chamber would not be the best way to get the pellets ready to inject. The large bore means you will have temperature differences throughout the material. That plus a mixer in a molten plastic would probably not work (perhaps as you found out).
Injection molds often "melt" the plastic suddenly via a shearing action via a turning screw... they do have heaters but that is not to fully melt but rather "get the plastic ready" to be processed. The injection pressures can be kind of high but I have not seen the calculations for this in over 2 decades...
I am not sure how to make this in miniature at home but as I noted there are several folks that I do believe have accomplished it so I hope you look into it some more. Some suggestions would be to make the chamber much more narrow, to hold just a bit more than 1 shots worth of material at a time.
The time it takes to inject and fill the mold is very important, it is usually a fairly rapid process to fill the mold, and then to transition to a pack and hold cycle. 'pack' and 'hold' are what they sound like - how do you "pack" the mold while the plastic is all still fully molten and say 95% full and then freeze the plastic in place and while this "freezing" is taking place hold some amount of pressure on the mold.
The idea is that the mold can heat sink the temperature out of the plastic very quickly and is often done with water cooling... Could also be done with a larger mold or a mold inside a mold "shoe" and perhaps a fan???
I don't know if you indeed already know this stuff and considered it for your project, I don't know everything and I truly like all of your content and applaud the efforts and ideas put forth. It would be great to see you solve the problems and present the next revision of your work and I hope it continues as home injection molding is very interesting as injection molded parts "should" have the best properties as compared to a 3D printed twin when it comes to break strength, appearance, durability, etc... Good luck and Cheers!
At my job we have been using sla 3d printed molds and 2 part polyurethane mix. The results are quite good. We use a large syringe to inject the poly. The polyurethane is thin enough that you don't need a ton of pressure, and the parts can be demolded after about an hour. Full cure takes a while longer.
I definitely want to see more... an industrial name for this kind of thing is "LPMS" (low pressure molding systems) and they do everything from glue-stick format to bags of pellets and air pressure molding.
My pet project is some fittings for my homebrew setup -- 3d print doesn't work because you can't clean them right (mold between layers or toxic UV cure).
Getting a commercial mold cut is 5-20k, so I 3d printed one and... have been having about the same amount of luck as you've been having.
It's not always the destination that matters but the whole journey along the way to paraphrase a quote from the series finale of Star Trek Voyager. "To the journey".
Haha character building, well said!
I had a similar experience with a fairly expensive 3d printer I bought in 2014, didn't have much success with it for a few years and I just now started to heavily modify it (it is well built mechanically so it is worth it imho), new hotend, new board with klipper, closing it up etc etc. lot of work but it is fun, not expensive and will hopefully be worth it 😊
BusterBeagle3D here on youtube has a great injection molding kit!
The mold release issue is due to surface texture. The more smooth you can get the mold surface, the higher quality the parts will be. If you are satisfied with the size/shape, consider making a more expensive mold for them out of aluminum.
I agree the mold stair stepping is not ideal at all. There's draft in the CAD model but when I have steps each vertical wall doesn't have it. Not sure if you saw part two but this mold is aluminium, just with a tool diameter mismatch which caused this problem.
Thanks for another awesome video,
It did seem that you might have given up on the PLA just as you were close to making some progress, I'd be fascinated to see what sort of difference dehydrating the PLA would have made.
Another way forward might be to 3d print a glue stick substitute out of ABS or PLA filament and then inject that,. the glue stick gun looks like it has a lot of potential for low cost injection moulding, though it might be limited to simple moulds that are more forgiving of shrinkage.
If the 3D printed glue stick replacement works a silicon mould of a glue stick filled with used PLA or ABS might be a relatively low cost (but low volume) way of recycling printer waste.
Also it might be interesting to experiment with silicone moulds they might be more forgiving when it comes to part extraction, though viscosity might be a challenge, There are some videos by TheCrafsMan SteadyCraftin that might be a useful reference if you haven't seen them already,
Thanks again and good luck with your future endeavours :)
That ABS clog looks like cheese and makes me so hungry.
I can send it to you, although I will be liable for any illness caused.
@@TeachingTech Hahaha very generous, but I think I'll hit the local market instead. ;)
Amazing work!
Maybe you can try using bicomponent poliurathane just like you used the glue.
Been meaning to give this a try for some time now after seeing some great results online.
Och my god so much work. Wow....
I think you should contact Brothers Make in the UK. They seem to have no problems with a homemade injection moulder for PLA. They do bake their moulds in an oven first though
Thank you for this video sharing your journey of reusing your 3D printing leftovers! I think you had some nice results and that your approach should not be compared with the results of industrial machines in the outcome. I myself am working on converting plastic waste into products and think the textile press method is a great example of where the direction can go. Recently I have been working on a "pellet" extruder for a 3D printer that uses shredded plastic. On the topic of reusing PLA prints, I think the approach taken by the recycling factory company is a good one. Not one person should try to make it happen, but rather a group of people / community. Maybe you can find someone in your area who is active in this field and who can/will take your print waste. The process of converting print waste was very nicely shown by another youtuber here called dr dflow, maybe check him out if you want to do the process yourself (not really worth it unless you convert tonnes of plastic).
Thanks again for the nice videos! Have a great day!
I was thinking about injection molding to create crystal clear object, that cannot be printed.
Maybe try a heated form and directly extrude from printers nozzle into it. (or direct extruder mounted on the top of the form and send M-commands to extrude the amount calculated + bit more)
I would be very interested to see what can be done with 3d printed molds and hot glue. espcially since notrmal 3d printing would not stand to the hot glue temps. curious as to your solutions. thanks for sharingthis, seeing something fail is always a good learing experience.
McLube 1850
Its what i use in the factory i work in.
Edit:
Also you'll want to heat the mold to about 80% of the melt/glass temp.
This would be a good method to make new filament perhaps, less grinding involved etc.
if it doesn't come out perfect, filament like shape is easier to chop up in even pieces for regular filament extruder ...
The key difference here is the BIG chamber
Geeez man! That was SO messy! Absolutely not worthy at all! Thanks for the video, Michael. I hope the views at least gives you some consoling.
As someone that operates a big injection molding machine I see 2 big Problems.
1.) As others have sait, you are overheating the PLA. It yould at most flow like hot melt glue. Overheated PLA is a brittle mess.
2.) The injection is extremely slow and unsteady. I think The plastic is flowin in multible "bursts" that don't fuse with each other.
To deal with those Problems you need to increase the injection pressure, probably beyound the 8 bar a normal air compressor can provide.
I really liked, that you also show what doesn't work.
Since 3d.Printers are good nowadays, xyou need a Part that you want tripple digit or high doulbe digit coppies, for injection molding to make sense.
Also dont be discouraged by the problems with your mold. It is common for professional mold to be send back to the mold maker once or twice because something did't work like in the mold maker expected and simulated.
Can you try using the 3D printer hot end to inject filament into the mold. Consider to use a 3D printer heater to preheat the mold.
This looks like a nightmare... I've watched a few channels that do hobby level injection molding, but never seen anyone use a machine like this one before. I've seen the ones with the big lever arm you pull down to get the shot into the mold. 🤔
With the recycled PLA sheets, can they be used in vacuum forming?
What a saga
Randomly skipped to 11:00 and thought you had the weirdest rice cooker ever
I want to make a video on this working so badly. Been doing a lot of research on injection molding recently.
Yeah, we know you enjoyed the secret satisfaction of getting three world records... :DDD
I saw that somebody use a glue stick model for the last of his rolls of filament and use those to weld parts together. A mold of a glue stick, toaster oven, time and your glue gun might have a good result too
Add a couple jacking bolt holes on your mold.
Did you try putting your mold under the hot end, of one of your 3D printers and just use that to get a molded part? I've been curious if that could work at all.
Not enough pressure and it extrudes way too slow.
Usually with these things the plunger is actuated by a pneumatic piston and the pressures are pretty high as well as injection times just some seconds, could be one of the problems why this screw thing didnt work
Have you looked into the Inari P35?
Don’t let Slant3D see this! LOL!
Joking aside, I don’t know if you have access to an SLA 3D printer that can print high temp materials but producing the molds from SLA prints might allow you to make more custom molds that perform to your needs.
Automotive gasket sheets + circular leather punches
Look up buster beagle 3D he has a injection rig that is in it 3rd or 4th iteration and has no issues like you’ve described.
That mould should pop open a lot easier than that. Do you have drafts etc?
It would be interesting to have a multi material part, part hot glue and part 3d printed.
This was an entertaining clusterf---. XD
Why not laser cut the silicone orings?
Would a tapered chamber - like a heated funnel - maybe heat things down near the exit?
I was just looking into what it'd cost to try this, and discovering how expensive it'd be to get the equipment to melbourne. Still, cheaper than a pellet grinder and a filament extruder :D Really don't likethe idea of just tossing PLA though.
I have no experience, but seems like people are using way less volume and crucially way smaller diameter for their diy melt chamber/extruder. Seems to me that such a big pool of plastic would need a long time to melt, and the material near the walls would just overcook. And besides you need a hydraulic press to move the material with enough speed with how thick molten plastic is.
I'd like to see you try to injection mold that recycled material after letting it sit in a dryer for a few weeks. It's easy for me to say typing it into a screen, but I'd love to see if it works..
I might be wrong here, but I feel like you would have a much easier time removing the part from the mold if the parting line followed the curve of the part rather than bisecting it the way you have done.
Also, why are you casting the silicone o rings? It would be far easier to just buy a sheet of silicone in the thickness you need and cut the o rings you need from that (You can use a machined die to get very round cuts).
Make PLA sticks and see if you can injection mold by running PLA through a glue gun?
The screws must really be moving on industrial machines given how quick they shoot a part, reopen and then are ready for another set of parts.
I'm sure you are probably aware of Action Box's Injecto 2.0. Maybe they could offer some help with your setup.
If you do end up spending any more time on this project I run a small injection molding company and would be happy to answer any questions.
Stinky clog... I know that one😅
I saw the "not worth it" in the thumbnail and thought I'd share some numbers from IMM processes... 500 degrees F.. 10,000 PSI injection pressure.. 5, 10, maybe 50 *tons* of clamping force.. 500 PPM H2O. And if you have a degree in plastic engineering or a few years of experience operating an IMM, you can probably get it making good parts in 3-4 hours of setup work, not counting hanging the mold. Assuming the mold is properly designed, of course.
This is a very dangerous process to mess with at work, let alone at home. Most burn injuries are caused by operators coming into contact with cooling water, not the material itself. And the amount of energy needed to dry and process the material *before* molding will make molding less economical than CNC machining or even ordering a finished part in many cases.
You can buy a 90s-vintage hydraulic powered IMM for a couple thousand dollars, and at least have a somewhat capable machine with reasonable safety features that rapidly makes quality parts. If you want to "make stuff" that's what I'd recommend.
14m:14s The drill is spinning the wrong way??
I'd like to see how injecting molding glue with a 3d printed mold goes.
As per usual, hotglue saves the day
What if you Print some hot glue sticks out of pla then use the the hot glue gun
Injection speed seems very slow. Fast and forceful seems to be the approach with successful takes on this.
Do you watched precious plastic and action box plastic injection molding machine videos..
How about asking for suggestions for injection mold projects, to see if one sparks your interest?
Can you use ABS to 3D print a glue stick, then use the glue gun to injection mold? i would feel very bad for you if it worked. But it kind of sounds like it would lol
not enough draft
if you really want to stick to this, theres a method to build two really strong plates with linear guides and clamping, then you put an insert into each plate that contains your actual part . you can finish the parting area by hand, with enough clamping force you can get seamless parts. [EDIT: without symmetrical linear guides and clamping, including towards the nozzle, none of this makes any sense to me. without pressure its not really a reliable flow anyway...]
this is used when different small parts in small batches are being produced. to me, that seems to be the only way to make something actually useful at smallest scale..