3 ms·

Though this is cool and I appreciate the ethos here the article doesn't really spell it out. All the cost, difficulty, time and money in injection molding is i

Though this is cool and I appreciate the ethos here the article doesn't really spell it out. All the cost, difficulty, time and money in injection molding is in the molds. They are expensive. When done well they can make 1000's of parts if not more. It's a great project, an amazing POC. There is the potential for small scale (garage sized) manufacturing to exist, the equipment is the cost of a (nice) car...

mildchalupa 3y ago
1000 pieces is a soft tool Millions are expected out of hard tooling.
throwup238 3y ago
I think the small scale, as in the footprint, is the point. They’d be much better off buying a used IMM if they want production parts but those are huge compared to TFA’s design. If they can make their own molds at a machine shop, this could hit the sweet spot for a product that’s too high volume for 3d printing but not enough to justify a professional mold. Set it up and run it once a week to make a few hundred parts, that type of schedule. Just practical enough for a hacker-run lifestyle business.
londons_explore 3y ago
I wonder if it might be possible to make moulds out of other materials for hobby-scale things? For example, a mould which is made of plaster/concrete with an inner layer of silicone? That would be pretty easy/cheap to manufacture. You'd use a 3d plastic printer ($150), dip the print in casting silicone, then float the result in concrete, with the concrete surface being the parting line. Then use a scalpel to cut the silicone along the parting line and add more concrete to make the other half of your mould.
- SOLAR_FIELDS 3y ago
  The design the buster beagle uses is an aluminum box you drop your molds into after printing them. Though this idea is interesting, I think, for people who are injecting with variable size molds that a single casing would not suffice for. A desktop CNC at that scale to make simple stuff like casing boxes is probably going to be more affordable and less hassle/mess than your suggested setup pretty quick though. Desktop is definitely not going to be good enough to make aluminum molds themselves though. I think you need a 5 axis CNC for that. Surprisingly my makerspace actually has one of those and I considered learning how to use it but decided against it. In practice however on these molding setups I’ve found you are already kind of constrained on your mold sizes a lot simply by virtue of your machine’s volume capacity. The one I have goes up to 3 cubic inches, but in practice you can’t use all of that capacity and you get weird/bad results for too small of capacity too so the actual ranges of shot volume I can produce in a mold are actually constrained to something like 1-2.5 cubic inches
jacquesm 3y ago
1000's would for most molds be considered a failure and very much excessive wear. You'd normally look for orders of magnitude more than that.
say_it_as_it_is 3y ago
It's only a Three-Dimensional inversion of an intricate model that requires very precise depths and must be reusable for hundreds of thousands of jobs. How hard can it be?
- jacquesm 3y ago
  And where both the material and the mold shrink at different rates and it would be nice if the product released from the mold after the injection phase. At most a weekend's work ;)
supermatt 3y ago
There are many videos on YouTube using 3d printed resin moulds.
SOLAR_FIELDS 3y ago
Total cost to do this is roughly (2021 USD when I went through this) $1000-1500 USD to build out a buster beagle with COTS components $700 for an anycubic photon plus a an anycubic treatment chamber $300 for all the ancillary materials like resin (Siraya tech is a great option), mold spray, pellets (purchased off eBay), $30 toaster oven from Walmart to keep the molds warm, etc etc Design your molds in fusion 360 and print them on the Anycubic So if you are patient and have more time than money it is theoretically possible to have a full fledged injection molding machine with up to 3 cubic inches (50 mL) in your garage for under $2000 USD It takes a lot of time and research to do this properly however. Designing the mold has lots of caveats and is very custom work. This process is not at all as simple as operating an FDM printer, there are a ton more variables and failure modes that go into injection molding so you have to be able to create an environment where you control all of them. One thing that could probably save a lot of time and heartache for people on this journey is finding someone on fiverr to design the mold for you for a few hundred bucks rather than trying to teach yourself some hardcore industrial engineering concepts. But it’s doable, I read an injection molding book and was able to produce something that worked after leveling up my Fusion 360 skills. I do want to underscore that yeah, as your comment implies, even though recently the money component has significantly improved with innovative designs like these that use COTS components the time and difficulty part is still there. Injection molding is not for the faint of heart, it’s definitely one of the more hardcore industrial processes out there. I have seen the endgame of this journey so I am happy to answer any other questions if people have them
- zer00eyz 3y ago
  How long did your molds last? PTFE and PTFE with glass? What kinds of plastic.
  - SOLAR_FIELDS 3y ago
    Molds last between 150-500 shots depending on all sorts of factors in my experience. Once you have the design dialed printing a copy of the mold is not terribly expensive, however, takes several hours and perhaps $5-10 of resin. Buster beagle sells an aluminum casing to surround the mold that greatly increases the life of it since the vast majority of the beating it takes is from pressure from the clamps. Not sure if you are talking about the PTFE molding process or not but this is just straight single shot injection molding, though it is possible to design multi shot processes with this setup. It’s just more complicated. I used polypropylene which imo everyone should start with first because it’s by far the easiest to work with and pretty easily sourced off of eBay. It’s suitable for a wide variety of components for indoor use. For outdoor use you would want something like polycarbonate. I think working with polycarbonate would not be as lucrative in a home machine and here’s why: temp and pressures for polycarbonate are often a bit harder to achieve and the temps you need for molding polycarbonate are often at the upper limits of what these resins used to make the resin molds will tolerate. I think it’s possible though, but requires careful selection of a particular polycarbonate with the correct thermoplastic properties (such as lower melt temp) to operate within the acceptable ranges of your setup.
bsder 3y ago
> All the cost, difficulty, time and money in injection molding is in the molds. They are expensive. You can 3D print high-temp resin, place it in an aluminum carrier and shoot roughly 20-50 parts. This kind of machine is a really good match for that.

mildchalupa 3y ago

1000 pieces is a soft tool Millions are expected out of hard tooling.

throwup238 3y ago

I think the small scale, as in the footprint, is the point. They’d be much better off buying a used IMM if they want production parts but those are huge compared to TFA’s design. If they can make their own molds at a machine shop, this could hit the sweet spot for a product that’s too high volume for 3d printing but not enough to justify a professional mold. Set it up and run it once a week to make a few hundred parts, that type of schedule. Just practical enough for a hacker-run lifestyle business.

londons_explore 3y ago

I wonder if it might be possible to make moulds out of other materials for hobby-scale things? For example, a mould which is made of plaster/concrete with an inner layer of silicone? That would be pretty easy/cheap to manufacture. You'd use a 3d plastic printer ($150), dip the print in casting silicone, then float the result in concrete, with the concrete surface being the parting line. Then use a scalpel to cut the silicone along the parting line and add more concrete to make the other half of your mould.

SOLAR_FIELDS 3y ago
The design the buster beagle uses is an aluminum box you drop your molds into after printing them. Though this idea is interesting, I think, for people who are injecting with variable size molds that a single casing would not suffice for. A desktop CNC at that scale to make simple stuff like casing boxes is probably going to be more affordable and less hassle/mess than your suggested setup pretty quick though. Desktop is definitely not going to be good enough to make aluminum molds themselves though. I think you need a 5 axis CNC for that. Surprisingly my makerspace actually has one of those and I considered learning how to use it but decided against it. In practice however on these molding setups I’ve found you are already kind of constrained on your mold sizes a lot simply by virtue of your machine’s volume capacity. The one I have goes up to 3 cubic inches, but in practice you can’t use all of that capacity and you get weird/bad results for too small of capacity too so the actual ranges of shot volume I can produce in a mold are actually constrained to something like 1-2.5 cubic inches

jacquesm 3y ago

1000's would for most molds be considered a failure and very much excessive wear. You'd normally look for orders of magnitude more than that.

say_it_as_it_is 3y ago

It's only a Three-Dimensional inversion of an intricate model that requires very precise depths and must be reusable for hundreds of thousands of jobs. How hard can it be?

jacquesm 3y ago
And where both the material and the mold shrink at different rates and it would be nice if the product released from the mold after the injection phase. At most a weekend's work ;)

supermatt 3y ago

There are many videos on YouTube using 3d printed resin moulds.

SOLAR_FIELDS 3y ago

Total cost to do this is roughly (2021 USD when I went through this) $1000-1500 USD to build out a buster beagle with COTS components $700 for an anycubic photon plus a an anycubic treatment chamber $300 for all the ancillary materials like resin (Siraya tech is a great option), mold spray, pellets (purchased off eBay), $30 toaster oven from Walmart to keep the molds warm, etc etc Design your molds in fusion 360 and print them on the Anycubic So if you are patient and have more time than money it is theoretically possible to have a full fledged injection molding machine with up to 3 cubic inches (50 mL) in your garage for under $2000 USD It takes a lot of time and research to do this properly however. Designing the mold has lots of caveats and is very custom work. This process is not at all as simple as operating an FDM printer, there are a ton more variables and failure modes that go into injection molding so you have to be able to create an environment where you control all of them. One thing that could probably save a lot of time and heartache for people on this journey is finding someone on fiverr to design the mold for you for a few hundred bucks rather than trying to teach yourself some hardcore industrial engineering concepts. But it’s doable, I read an injection molding book and was able to produce something that worked after leveling up my Fusion 360 skills. I do want to underscore that yeah, as your comment implies, even though recently the money component has significantly improved with innovative designs like these that use COTS components the time and difficulty part is still there. Injection molding is not for the faint of heart, it’s definitely one of the more hardcore industrial processes out there. I have seen the endgame of this journey so I am happy to answer any other questions if people have them

zer00eyz 3y ago
How long did your molds last? PTFE and PTFE with glass? What kinds of plastic.
- SOLAR_FIELDS 3y ago
  Molds last between 150-500 shots depending on all sorts of factors in my experience. Once you have the design dialed printing a copy of the mold is not terribly expensive, however, takes several hours and perhaps $5-10 of resin. Buster beagle sells an aluminum casing to surround the mold that greatly increases the life of it since the vast majority of the beating it takes is from pressure from the clamps. Not sure if you are talking about the PTFE molding process or not but this is just straight single shot injection molding, though it is possible to design multi shot processes with this setup. It’s just more complicated. I used polypropylene which imo everyone should start with first because it’s by far the easiest to work with and pretty easily sourced off of eBay. It’s suitable for a wide variety of components for indoor use. For outdoor use you would want something like polycarbonate. I think working with polycarbonate would not be as lucrative in a home machine and here’s why: temp and pressures for polycarbonate are often a bit harder to achieve and the temps you need for molding polycarbonate are often at the upper limits of what these resins used to make the resin molds will tolerate. I think it’s possible though, but requires careful selection of a particular polycarbonate with the correct thermoplastic properties (such as lower melt temp) to operate within the acceptable ranges of your setup.

bsder 3y ago

> All the cost, difficulty, time and money in injection molding is in the molds. They are expensive. You can 3D print high-temp resin, place it in an aluminum carrier and shoot roughly 20-50 parts. This kind of machine is a really good match for that.