Machining an injection mold for acrylic lenses
HTML-код
- Опубликовано: 1 июл 2020
- Injection molding: a lot harder than you'd think! I picked up a benchtop injection molding machine to play with, and thought a "simple" double convex singlet lens would be a good first project.
📢 Twitter: / breakingtaps
This video is a mix of machining, injection molding theory chat, molding some acrylic lenses, and then analyzing their stress concentrations with polarized light and birefringence. Hold on to your butts! Oh, and we played with a new polycrystalline diamond (PCD) end mill too.
The machine I have is a Proto-Ject 150HP: it can deliver ~9k PSI injection pressure via an air cylinder plunger, 0.75-1.5oz shot size and has a 25T hydraulic clamp (I incorrectly stated it was 20T in the video). Max mold size is around 9" square, and 4-6" deep. The machine is still very new to me, but I love it so far!
Machine is from: www.manninginnovations.com/
Note: the clamp is a pre-production unit that the company was graciously willing sell to me, so I don't believe is available for general sale yet. But this is not a sponsored machine, I paid for it all with my own hard earned dubloons ;) 
Наука
dude... you are punching WAY above your subscriber count with the quality of this content. hopefully youtube recognizes this ASAP and sends more traffic your way!
This is great content and for a home job you did awesome! I am an auto engineer who specializes in injection molding lenses for cars (parabolic, fresnel, etc) and if you want optic clarity aluminum won't cut it. High carbon tool steels like SAE P20-HHMLQ are best because they can be polished after machining to a mirror finish (requires special diamond polishing for SPI A1 surface) Lower grades of steel may have impurities that prevent reaching A1 level. As for your injection troubleshooting you did very well but try to use the highest grade PMMA material, make sure to try it before use, and when setting up your injection parameters aim for the higher end of the melt temp/mold temp. Hot and fast works well for optics made with PMMA and high temp PC. Inject hot material as fast as you can which will require LOTS of good venting (its hard for me to tell in the video but I think you need more) Overall you did a great job and if you can crack optics at home its a good business!
What helpful advice! You seem very nice!
👍
underrated youtuber
Not for much longer!
Underatted RUclipsr
Yeah, totally blown away.
Overrated commenter
I know right, like wtf youtube?
you may have figured this out already but if you warm/heat your mold it will help significantly with the streaking and the internal stress. looks really good tho
I assist the injection molding process technicians in a Honda parts supplier facility.
You nailed it with your guess on why you were getting the silver/splay. try adding a small resin dryer to your setup, if you add a small dryer and a hopper that draws from it you will likely eliminate your splay problem. No amount of heat will help with non-dried plastic.
Item 7, mould temperature. Injection moulding is a thermal process, requiring accurate control of both melt and mould. Both need precise control,
around +/- 2C.
Item 8 mould packing. This controls the pressure applied during the cooling phase when material transitions from liquid to solid, the packing replaces the material lost to shrinkage, and defines the actual part dimensions and quality of the finished lens.
Good luck you are doing well.
In 20 minutes I learned a ton. Thanks for sharing your project.
Could you make a polishing-pad tool for your cnc?
Great stuff btw. I love when channels don't tell me to subscribe every 5 minutes. Just good content shown in a relaxed manner without wasting my time babbling around to get past the 10min mark for the sake of the yt algorithm. Reminds me of Applied Science. And I will subscribe anyway because that is exactly what I enjoy watching :)
polishing pads for CNCs are pretty common from my days as an ME intern hanging out with the shop guys. They would polish some of the prototype molds in the VM2 with little lollypop felt bits (don't remember exactly what they're called) and I remember them working surprisingly well. We weren't making any optical components, but it might be something to check out for sure.
@@zachbrown7272 felt bobs
Exactly what I was thinking, felt bobs mounted in the cnc. You already have a toolpath that worked. I would suggest trying around 10k grit gel and re-wet and add as you see fit.
Also, wanted to mention that you could try preheating the mold so less tempersture differential causing a thinner layer of rapidly cooling material which then gets sheared. Another thing that might help is making the lens thicker. And, if you really want to be a perfectionist you could preheat the mold, make the lens thicker, add angled vias at the top for distribution, and make the injection molded part a lens blank that would need a secondary 'trimming' operation. That's making it oversized so that the flows homogenize, and you could maybe make it square or some other shape that fits the flows (minimizes the shrear) or for ease of fixturing the trimming operation.
I think your beginning (pre-shot) mold temperature is critical and needs to start out much higher. Ideally the melting temperature of the plastic used. Letting it cool under force from the liquid state would keep a the long-chain 'mers (as in polymer) in random orientation all the way until solid instead of the shear causing the 'mers to be aligned more parallel, which changes the diffraction index.
I must say you chose a hell of a first project to try your new machine on. When I worked in a prototyping cnc shop I was told that the hardest thing to machine is a perfect sphere and that by making a sphere (or hemisphere) and measuring the deviation in the different axis you could 'super-tune' a machine tool (lathe, mill, or multi-function) to be even more precise than without.
Good luck!
Incredible. Instant subscribe. Thank you! I've been following the optics of acrylic resins since we first got clear resin at the consumer level, specifically for this. There's some great early tutorials on making molds for dove lenses w acrylics. Injection molding makes so much more sense. Again, thank you.
Thanks! I agree, there's just something reall tantalizing about being able to make stuff out of clear resins at the consumer level. Possibilites are endless (once the proess is dialed in)!
When I took injection molding in college, you have to look from the perspective of the plastic. The four variables are: 1. Time, 2. Temperature, 3. Pressure, 4. Shear. From the perspective of the plastic, that's the only things it cares about. All the care abouts can be lumped into those categories.
The material is plastics. This was drummed into us ophthalmic optics students!!
Reminds me of when I had to take my brakes rotors off soon after replacing them, for an unrelated repair. I was disheartened to see what looked like deep scratches, yet they were as smooth as glass to the touch. It was so strange, part of your brain and eyes say one thing, but your fingers and the rest of your brain say that is wrong. Not terribly relevant, but at least it makes _Breaking_ _Taps_ look better on the algorithm. :)
try annealing the lenses to see if the stress goes away
I wonder if the annealing could be done while it is still in the mold?
@@jcims You're generally supposed to pre-heat the molds anyway, so you could probably just keep the mold itself near the injection temperature and then slowly cool it. Should theoretically help, at least with certain plastics.
@@TheVideoGuardian I wouldn’t keep
the mould at injection temperature. Maybe around 40degC though.
@@welshsteve2009 Injection temperature is probably too much, but in some plastic if you hold above the glass transition temperature for several hours they become stronger/clearer/etc. (60C in the case of PLA, less time if above that.) I'm not sure how a mold changes things, but this effect works with 3d printed parts in an oven, albeit they suffer terrible warping. This only works with some plastics, however. (IE, not ABS or PET)
@@TheVideoGuardian Thanks for the reply. Yes, I agree. However, In the real world though, it wouldn’t be practical/economically viable to hold parts for any significant period at glass transition temperature.
This is amazing, the way you communicate information noting the limits of understanding, no hand waving generalized assumptions. Its so clear and informative, true educator.
It make me really appreciate the $2 shop reading glass they are selling. Thanks a lot 👍
You can add a cold slug catch feature to your runner if you had a little more room. One my early molds was to mold a simple flat disk, similar to the mold in your video. It wasn't a lens, so it did not have compound curvature, but instead, had flat faces. However, the plastic flow front profile during injection would have a similar profile to your mold. Back then, I didn't use cold slug catches. In my first version of this mold, I ended up with some inconsistent defects in the molded part. Adding venting to the opposite side of the cavity (from the gate) and adding a cold slug catch to the runner solved the issue. I later did molding simulations that showed me that cold slug catches almost always improved all numbers across the board - pressure drop, temperature gradients across the part, frozen layer fraction at end of fill, temperature variance, cooling time variance, and other numbers.
Well that brought back memories. Back in the dark ages I began my machining career, briefly as a mold setter (the guy who placed the set of molds into a mold base and set the whole works up in the molding machine), then talked my way into becoming an apprentice mold maker. It was a large company, but the only computer controlled machine I worked with was a NC Moore jig bore, as CNC machining centers were nothing like is available today. Thanks for the trip down memory lane!
This pretty sweet! I know a little about injection molding but not a whole heck of a lot. I just learned a ton, especially with optics. Thanks for this video!
This is absolute gold. Honestly your channel is something else entirely. Maybe my interests just happen to align with your content more, but every video of yours I watch seems to just be put on top stuff I've watched in a while. Easily up there with huygens optics, ToT, etc.. It's all the stuff I like in one channel with perfect quality and perfect explanations. God damn.
Holy shit, injection molding is so much more complicated than I could ever have guessed, damn the modern world around us is amazingly complex! I guess that's what you get when hundreds of millions of smart people have work hard and figure out stuff over the course of hundreds of years
Cool channel dude! Gonna blow up if you keep putting content like this out.
Really like the technique that use LCD screen and polarizer to show internal stress. Good job man👍
I like your talking, it's explanation, not babble. Seeing as I know little beyond A Level Physics (obtained more years ago than I'd like to count) I find your explanations very useful and I wouldn't understand what you do and why you do it without them. Keep doing what you're doing, please! 👍👍👍👍👍
Great video! Really well made and explained, not too long, enjoyable to watch and listen to. Subscribed :)
Just a random thing: everytime I feel blue, I watch one of your videos. So calm and educational. Thx!
Injection molding is a super deep rabbit hole that branches off into all kinds of non-injection derivatives. My dad retired from Olin 20 years ago and the guys he had been training for about 6 months on an Emu machine from Italy approached me at his retirement party and said "We are so f&@%ed. No one can run that machine but him. He has to be a savant". They were honestly afraid they would lose their jobs, tenured guys, machinists and professional adjusters! Imagine the most complex series of 3x 15' long joined machines you've ever seen, each section having 24" touchscreen loaded with page after page of controls and about 100 extra points of manual adjustment - and, all of the indicators have been translated (badly) by Italians who were NOT bilingual. Two weeks after he retired Olin had placed help wanted and industry calls for people who could translate Italian (because they had to bring the designers to their plant in Illinois, not the builders or machinists who made the parts but the guys who designed it because no one on earth other than him had EVER been able to run it lol). He's a true freak of nature, even now at going on 76 and suffering the onset of dementia which is a true shame because he knows things that no one else has ever known and no one else likely ever will again about how to work materials in constantly fluctuating conditions. Take a close look at a Winchester shotgun shell casing sometime and realize that millions and millions of those have to be made super consistently and within very precise tolerances for different and unique pressures relative to their gauge - or people get hurt, badly. Many seemingly simple everyday items out there have guys and gals behind them that KNOW super technical things, yet, most look at most of them and assume they're just some imbecile. Science's most overlooked aspect, to me, is how humanizing it can be.
Thank you so much! Helped me alot to understand basic tabletop injection molding
Another awesome video!
Love your videos!
Great job. Please more.
Just found your channel, needless to say I’m going to be binging a new RUclipsr now lol. Really enjoying this video and with this quality I think I’m gonna enjoy all of the others lol. Thank you for contributing these videos to RUclips they are truly great! Keep it up
Oh wow that is amazing. Some fine tuning and you got it!!! Subbed
Have you thought about heating up the mold to a good temp so that the transition is not as much in order to reduce the stress lines?
I don't know how I'm JUST finding your channel! This stuff is great!
1 year later and STILL producing top notch videos while remaining in stealth-youtube mode
I agree with everyone below. This channel is so underrated that it makes me wanna do something like actually share it. I've watched almost your entire catalog of videos and every one of them is better than the last one I watched. Hope to see you really dive into optical lenses. I've been collecting lenses out of everything I can get my hands on the last few years in the hopes of one day building a microscope for graving and general shenanigans. I know I could prolly find all the content I'd need to learn how to do this online, but some of those college level papers are wordy and long winded. At least some of em have pictures I guess. Lol!!
Anyway, thanks for the awesome content!! You're killing it!
There is a software company called moldflow that can work out gate location, runner location and temperature (cooling lines on the mold) to essentially control the shear of the material as you inject the plastic into the mold. They might assist you in that field. Also, the injection dwell time, velocity of the injection are all tricky variables to navigate.. I’m sure you’ve noticed by now. Also, there are different materials for optical purposes with different melting index.. I love this video. This topic did spur my interest in the 80’s..kind of brought back lots of memory on it as well. Thx for sharing.
Your on your way to a full home cinema system ! seriously great channel content...cheers.
Excellent video dude
I’m working on home routed polishing tools for SLA printed lenses. This is a great video!
Very cool, keep us updated! Feels like there is a ton of things that could be done with SLA optics if the grinding/polishing process could be automated a bit.
Just watched a few of your videos and I am subscribing right now, great work! really interesting :-)
Streaks are definitely from the moisture. The same happens for polymers like PC, ABS etc. In fact they are very little foamy resin solidified if one looks closer.
I know I'm 2 years late and maybe you already solved the surface finish issue, but if you haven't then look into burnishing tools. They're essentially a very tiny diamond tipped tool that gets pulled across the face of the surface with a set force and it will smooth out all the microscopic hills and valleys and essentially make the surface very close to a mirror finish. We use them at work all the time.
You'd just need to mount the mold in a lathe so you can spin it as speed to be able to drag the tool across the whole lens face.
So.. Obviously I'm binging your content like a big fan boy. This stuff is incredible. A while back I was making a cheap SLA machine and I messed up the lens in my projector. I was able to mold and epoxy resin a new lens which actually worked. I had no idea till then that you could possibly make a lens at home. Plastic world in my area even had particular epoxy designed for clear lens making. Under vacuum of course. Love your stuff.
❤ Thanks! Yeah, the whole area of molded optics is a deep, dark rabbit hole I've fallen down on several occasions :) Can either spend lots of time making an optical-quality mold, or use fancy single-point diamond lathes to fix the surface of molded parts (or just turn them directly... those machines are wild).
Optical epoxies are pretty nifty though, and definitely removes some of the major pains of injection molding. Currently noodling over an optical epoxy project, we'll see how it turns out :)
Would be interesting to anneal the lens and investigate if there is less stress present afterwards... Nice vid!
Oooh, I like that idea, thanks! Should be easy to test too
Anneal it in the mould to maintain its shape.
Impressive. Kudos.
Acrylic is abrasive and will wear your aluminum molds out. Also since it's abrasive, you can use it to remove burnt resin from the internals of your barrel and nozzle.
Robert Murray said it... you're going to need way better molds to get good optical quality. For small runs you might do just as well machining, sanding and vapor polishing the acrylic. Fun stuff, but as you say... many variables! Thanks for the video!
i just found this channel and no idea how i am only finding it now ....
Hey, I am a little late to the party on this one since I just found your channel. You do some awesome work and it's greatly informative! From what I understand the residual stress that is in these lenses is partialy caused by cooling at different rates so potentially another way to increase your optical quality would be to heat up the mold before each injection. Would be interesting to see how good of lenses you could make on your own keep up the good work!
Very important project !! I was trying to do the same but trying to finish the epoxy lens with 2000 sand paper
Nice video. For a smooth clear finish chrome the mold inter half.
I worked in an injection molding factory for a bit as an operator/material handler where they molded one small acrylic lense. What always struck me was that the runner/sprue was i'd guess at least 8x the volume of the actual 2 ~1 cm parts per shot. i was never able to find out the reason but i assumed it had to do with the nature of it being a lense . I do recall it didnt have a fanned gate, though that may have been due to its small size.
Cold shots on any plastic runner system should be 1 1/2 x the diameter of the runner in length beyond the next runner trunk for multiple cavities or the actual cavity inlet. For round cavities the actual gate should be a fan type but no more than 120° in arc width. Any wider and you begin to entrap air into the center of your parts. Leveling out internal shot pressures in parts can be handled by adding overflow pockets, parting line vents, and/or cooling lines for more even mold temperature control.
Really interesting that you get those rainbow effects. I wonder if you would be able to mill a diffraction grating, maybe using chatter to your advantage. In theory, it would show up on the cast too.
Very interesting idea. Turning a disadvantage into an advantage. I wonder if anyone has done research on producing chatter on purpose and controllably.
Super cool to see your videos since I have a recently built benchtop pro also with Clearpath servos. Keep em coming! 👍
Nice! How're you liking the servos so far? That's definitely an upgrade I've been strongly considering :)
@@BreakingTaps You have a way nicer spindle though ;) I love em! I used many of them in the past... 39 clearpaths on this crazy thing I built: ruclips.net/video/dzkMOAWC4Eo/видео.html . They are torquey as heck so they can maintain very high travel speeds even while contouring. But I think the main benefit is the built-in encoder. So you can run them fast and hard knowing they'll throw an error back to the controller if their actual position slips a certain amount from the requested position. That said, like most of this, its definitely not required. :) An initial test countour run here: instagram.com/p/CBtkbZynwS0/?
I'm just getting started in CAM though so it's a journey for sure....
Great to hear! Definitely keeping servos high on my "things I'd really like to upgrade" list. I don't often miss steps, but it does happen and basically scraps the part. Super annoying, especially because you don't find out until the end when you see it's shifted.
That sculpture looks great! Bet that was a nightmare trying to get all the servos and lights wired up :D
i make a bit of mold and you can never get a good finish with any bit ...
we use to polish the matrice with oil stone from 220 to 2000 and then finaly polish with daimond paste... and the "mold need to be heated and water cooled"when you take out the piece we use glicole... to prevent rust (sorry for my english)
I have a interesting idea for a new episode with your new PCD end mills, what about using then to machine a working RF cavity like the ones seen on particle accelerators?
Back-up lenses for semi-truck trailers (oh they don't have any)... and warm up the mold..
Nice cutters
Hey dude, been really enjoying the channel these past few days. I have a suggestion for you. Hit acrylic with a torch really quick and I'm willing to bet those machine marks will totally disappear. In the shop, when we cut acrylic we always torch the edges and they clean right up, clear as crystal. Might help a lot with your optical quality!
19:10 you could also try to bent it. Induced Stresses would show in realtime.
I wonder if you could use glass for the mold? Either full glass molds or mounts some pucks in metal blocks. That would allow very good surface finish and may allow good figuring.
may i suggest that for polishing aluminium you try some autosolve german polish with scrunched up tinfoil, it works very well and takes less time than a buffing mop on a drill, its something that motorbikers do when polishing the metal work, i also make my own Dremel bits for polishing and have found that multi-layeing a layer of green scouring pad with a layer of soft cotton materiel buff up metals real nice, you have to hand stitch them tight but they clean up metal work real well, i have used this to polish out scratches from glass and clear plastic.
I ❤ your videos.
So, firstly, this is super cool and I can't wait to see more injection molding
But secondly: When looking at the lenses under the polarized filter, my understanding is that the stresses are shown with a colour difference like in the center, but you seemed to be talking about the darker lines. I noticed that as you moved the lens around those darker lines moved all the way across the lens, not fixed in one spot. I might theorize that those darker lines are actually light coming into the sprue across the inside of the lens and refracting, and the reason those sections are smaller is because the bigger gate is pulling in and allowing more light. The coloured section that transitions from a dark purple out to yellow seems to be the stress you were talking about, as it remains stationary. This also lines up with what I've seen looking at cheap molds like clear disposable cutlery, being able to see that rainbow pattern, but only while I'm wearing my polarized glasses. You could likely test this by taking one and nipping off the sprue and seeing if the dark parts are still there, or if it's just the coloured pattern.
I'm absolutely not an expert here, I could be totally off base here, and it's possible I just misunderstood what you were describing, but I thought I'd share even if it's only for me to understand what you were saying better.
Heya Coatduck, thanks for stopping by! Loved your collab with Winston btw :) I think I poorly described myself... I was considering the dark regions and the colored regions all as "stress". But you have a really good point, I didn't think about the effect the sprue would have on it! It's a solid cylinder, it probably acts like a light guide and funnels in light from a different angle. I went ahead and took a before/after like you suggested, photos and some shakey cell phone video here: imgur.com/a/vPDiUSx The weird aliasing/pixelation is from sitting directly on the monitor, that disappears when you move it off the surface a little bit
It was hard to recreate exactly what was in the video... it is really sensitive to position of all the elements and their rotations relative to each other. The long dark gray lines are largely absent this time both with and without the spure. Soooo I think that must have been an artifact of positioning or focus or something? The only thing left are the green/yellow/purple regions that rotate through colors as you rotate the lens, so I agree that's definitely the stress regions
Not sure the takeaway, I suspect it was just something about the setup or where I was holding the lens? I was also using a camera with a longer focal length vs. a cell phone, so perhaps that had something to do with it too? Dunno, definitely not an expert either :)
Note: Some of the really dark, stationary lines are that streaking/splay I mentioned, it's hard to see on camera but very obvious in person.
@@BreakingTaps oh dang, thanks for the complement! I've been subscribed since your first Taig video
That was a really in depth analysis, thank you for doing all that work to sate my curiosity. If you were still wondering about the disappearance of those dark lines, I noticed a few things that might be a factor: It looks like in those shots they're being backlit by your monitor, whereas in the video they were lit from the front by a bright source. It might be the fact that the source is now larger than the lens, or perhaps just darker than your nice photobooth setup, or perhaps even to do with the colour index of the monitor, since it's putting out RGB "Whiteish" light, rather than a true white phosphor like in a white LED or something.
I have no idea if any of those would contribute, but they're leads if you wanted to track it down. But either way you got the data you were after, which is super cool. I can't wait to see more on the series. I've had a DIY injection molder in my head for nearly a year now. It'll surely be at least a year or two before I'd even start on it though. Gonna need a bigger apartment.
Instant sub, I’m amazed you don’t have a much higher sub count
Put a small buffing wheel in the CNC machine and step through polishing compounds until you have a mirror finish.
you could polish the molds in a simmilar fashion to telescope mirror polishing. Just make a small disc that has the exact same curve as the mold, put it on the mold and put a bit of water and very mild grinding dust in between them and grind away. there are many guides online on the exact technique of grinding telescope mirrors
Would be interesting to see stresses when lense or other sample bent or other loads.
Having done mold making, you getting close, you definitely need registration pins.
Certain types of plastics are super corrosive and abrasive to the mold at temp. PVC, and other plastics can dull the mold material even when the mold has a mirror like finish. Sinker EDM was used for stuff like this. My molds were also water cooled when injection molding for quality. You can also pre-heat the mold with the water.
Great video. I would loved to have had a table top injection molding machine….
Do you have vents in that mold? To let the air escape when injecting.
You could try putting the mould in an oven to pre-heat it to somewhere between 150°C an 200°C (a bit on the hotter side is probably better because I reckon it will lose quite a bit of heat while you're putting it into the machine before you're actually ready to take the shot). After pre-heating it would be better to mount the mould into your machine with two pieces of wood or cork (assuming those can stand up to the clamping forces, otherwise delrin or teflon might be good insulator candidates) between each side of the aluminium mould and the press face to prevent the huge thermal mass of the press' metal confinement plates/frame from sinking the heat away from the mould. I think that pre-heating the mould has a good chance of reducing the striations, stress patterns and/or any other injection difficulties you might (not even know that you) have! :D
BTW: the algorithm has picked up your channel and is randomly recommending it to people (like me) so be ready for a HUGE influx of new subscribers! I've seen channels from anywhere around 10k subs skyrocket to easily 100k-200k within a few days or at most a few weeks. Keep up the quality and keep doing whatever you're doing and the algorithm will continue to love your videos and send LOADS of traffic your way! :D
The hardness of the PVD tools allows you to cut hard materials without the tool "changing shape" ie wearing.
I've had experience hard machining hot forging die tools where we had to machine the part, measure it then adjust the part model in CAD to account for tool wear and then re-machine it.
This was with 1mm ball nose coated carbide endmills.
Woah that's wild. I feel like just breathing on a 1mm tool would break it :)
Use has drill bits next time you need a quick locating pin.
That is ingenious. I have plenty of broken or dull bits, and probably even more broken endmills now that I think about it.
You can get that aluminium mold to a mirror shine, use 1000/1500/2000/2500/3000 sandpaper grit in this order and then polishing paste wich is also in different grits. And i think if you want to get rid of the stress zone from the mold, you also have to heat up the mold as well when you do this.
@14:03 - that's why one of the mold shops I visited in China said that they only had women work in the polishing room; men were too impatient and would gouge the mold :D
seems like you may need to preheat the mold prior to injection. you may also need to cool the mold gradually. some sort of temperature controlled bath may work. the reason for the streaks may be that some of the median is being cooled faster than the rest during the injection, solidifying as it being injected. the initial median solidifies but in prosses heats the mold so the rest of the median is able to spread before solidifying. So preheating mold prior to injection may help. Also I did not see any vent holes, you may need to add those unless you use vacuum to evacuate any gasses prior to injecting median.
Dry the material in an oven for about 4 hrs @ 150 f and it will make big difference. Gate size and design is facinating, the change in shear can have counter intuitive effects. A flash gate might be how they do production optics.
Two thoughts: what effect would heating up the mold have on streaking, etc? And, how are you going to polish? Seems like you could take an existing lens and use it to polish the mold. Like coat a camera lens with a sandpaper and use the precise curve to get your polish done. Or maybe coat the lens in something protective and then use a lapping powder.
Should look into vapor polishing to finish the final parts
Are heated molds a thing? Seems like you could eliminate a lot of stress and shearing and such... you could even mill in cooling channels to cool it down rapidly after injection...
Nice mold. Injection molds can be way more complicated with extras such as moveable cores or ejectors on the moveable plate. They are also very expensive to build, some up to a million Dollar.
from what i know, when making eyeglasses, i think they just start with an acrylic block & machine it to near the lens shape, then clean it up until the measurement is right. it seems like youre doing a lot of extra work to me, but injection molding could skip the wait time of machining each new lens (if you plan to make a lot), so not a bad experiment.
i never worked with any of this stuff but since you said the stress in the lens is from the gate, my idea would be to remove the gate from the lens. imagine the acrylic is injected from all sides at the same time, the only way i could see to do this is by having a small moat around the lens, & use multiple gates on the moat which all channel to a primary gate. if the lens is horizontal & the primary gate is perpendicular, then the secondary gates (4-8? idk) should maybe have an even flow, but the actual edge of the lens would have a thin opening to the moat completely around & the acrylic would only start pushing into the lens as the moat has filled up. im not sure if it would remove the stresses from the lens, but it might at least be interesting to see how it changes, as most of the stresses should be concentrated in the moat i think, which is disposable.
actually, now that i think about it, my idea might make a lens with a hole in the middle, because i didnt think about releasing the air.. or does injection molding vacuum all the air first?
Dude you're kinda my hero -- take it you've got a background in optics/mat sci? Seems like you've had a pretty awesome little lab/workshop going too! P.S. Might try to get your hands on some polyolefin copolymers that are index matched to BK7.
Should go with polycarbonates for lens parts. Both halves of your mold need to be heated above 150 degrees F to minimize molded-in stresses and the gate shear. Also, clear plastic lenses require very slow injection fill rates.
The pcd cutter is actually better suited for use in hardened steel...aluminum can be a bit gummy some series more than others.....aluminum with a carbide ball end mill with a ZrCN Coat would be a great option
so many geology concepts!
Awesome video! Curious, do you plan on making a video about the ins and outs of the injection molding machine? I've been curious to add a small injection molding machine to my prototyping workshop. Thanks!
Thanks! I'd like to make a video like that, although I'm still very new myself... there is a deceptively large amount to learn about injection molding I've discovered. I'm noodling over a project that will need some IM parts, so I'll try to take down notes as I work on it and see about doing a dedicated "IM for the home shop" style video :) The machine itself is pretty simple, would be easy to cover. I've been pretty happy with it so far, especially once I understood the limitations of desktop IM a bit more :)
Pre heat the mould, use low pressures, optimal temp, use hard chrome plated steel mould
better quality acrylic lens can be made on a CNC lathe with monocrystalline diamond tool (cheap). actually you can make it on your mill by mounting workpiece in the spindle.
This is where the mold has to be turned rather than milled with a fairly large radius forming tool. Good news is that it's perfectly possible to simulate it on the mill by mounting the workpiece in the spindle and the tool on the table. Other features can be machined in a regular manner.
Hah, that's a clever idea. Might try it on a smaller piece and see how it goes. Or just finish re-assembling my lathe some day 😅
@@BreakingTaps, another technique to improve the surface finish is to machine it at an angle. Here's why. When milling any surface close to horizontal/flat with a ball end mill, it touches the surface with a tip portion of the cutting edges. There's a lot of friction going on there, chips get welded back to the workpiece (coolant should be shooting straight at that tiny spot), the tip has very low linear cutting speed, etc. All that is bad. So, it's a good idea to have some more acute angle between the end mill axis of rotation and a tangent plane to the surface being machined. To achieve that in the industrial environment, such operations are programmed with simultaneous 5-axis toolpaths, but the rough approximation of that would be just tilting the workpiece in a vise. The goal is to have a tangent plane at the lowest edge of the lens cavity at about 15 degrees to the horizontal. Although, It might be a kind of tricky to achieve without a tapered tool holder or something like a smaller ER collet extension.
Those aluminum molds are giant heat sinks that will rapidly pull the heat off of acrylic, which will make the stresses worse. You need to ensure that it cools slowly.
If your material is high viscosity, and the ratio of gate diameter to runner diameter is too large, and your gate is too long in length (all of which I saw in your video), you end up excessively shearing your material as it enters the cavity. I'm an Mech Engineer that works in the plastics industry. I design injection molded parts, injection molds and perform molding simulations (CFD).
wtf i was already subscribed.
Well subbed again.
Hey, nice little mold! I would try making the gate shaped like a hand fan - thin but wide. Maybe this will solve your "partial shots" problem. It should provide more flow for this thin and yet rather large part.
Thanks for the tip, I'll try that out on my next mold. Makes sense, since it would reduce the pressure at the gate and probably allow more volume to flow. Cheers!
Could you do a finish pass on the CNC with those silicone dental or jeweler polishing bits?
Those stresses look wavy, could you split the gate in such a way as to make theme destructifly interfere?
have you tried to preheat the mold, it may help some with of stresses. I only operate an injection mold so the only thing I can tune is machine based variables but we also don't worry about internal stresses because we run plumming parts.
why have I not found this channel sooner?
Awesome video! Did you have to dry out the pellets? I've tried a toster oven with limited success
Wonder if you could spin coat some clear liquid resin on the acrylic substrates to get better optical quality.