I wont really get into molds much as this post is probably already boring enough. But I will say that molds are generally very expensive. The cheapest mold in our plant (besides the inserts we use in our babyplast) cost us roughly $150,000. Most are probably upwards of a half a mil. Now thats not to say you cant have an cheap mold made say out of aluminum or some other cheap metal. but it probably wont hold up very long and most likely wont make you very many good parts.

To answer your questions. No, I also seriously doubt that most toy companies spend $100,000+ on each one of thier molds (although I do know for sure that LEGO does, they have 50+ machines in thier germany plant, all shiny new custom built arburg machines which are switched out for new every few years). What most toy companies probably do is called insert molding. Which is simply instead of changing the whole injection mold when its time to make a new part. They exchange the Core and Cavity inserts in the mold itself. In doing so they avoid the high costs of having a whole new mold built for each different piece. Also some toy companies probably use molds made out of aluminum or cheap steel contracted to be built in china (where injection molds can be machined very cheaply). But like I said before these molds probably dont hold up for any extended period of time. Some companys also use multicavity molds to mold different parts in the same shot (think all those different army men that come in a bucket for a simple example). Toy companys also dont need to worry about stringent dimensional tolerances either. Which if you were making your own 360 laptop case or something similar I'm sure would be a concern for you (since you would want the case halves to fit together nicely).

So, I've seen a few youtube videos of people converting the Harbor Freight mini-mill to be CNC and I've seen the maker bot plastic extrusion 3D printer, now let's connect the dots and create a plastic injection molder.  With the HF CNC mill, one could, in theory, mill metal molds for an injection molding machine.  The maker bot guys figured out how to control the melting of plastic for their extruder.  If you put those things together, you get the basic parts of an injection molding machine.  The only thing left is a hopper fed piston type arrangement that can soften/melt the plastic and push it into the mold.

To me, the injection molding machine is the last unconquered territory of the homemade fabrication machines.  CNC has been done.  Circuit board reflow has been done.  Picking and placing of electronic parts has been done.  3D printing is a work in progress.  However, injection molding plastic parts is still beyond the capabilities of the average DIY'er.  To have a plastic part injection molded by a company costs a couple thousand dollars minimum.  If I had a benchtop molding machine and a CNC metal mill that could crank out molds for me, I'd be in plastic molding heaving.

element14 is the first online community specifically for engineers. Connect with your peers and get expert answers to your questions.

Another thing to consider is that heated plastic degrades very quickly in some cases. Even in a normal injection molding machine some materials (nylon, acetal, delrin) cannot sit inside the barrel for any longer than 30 seconds or so before they start to "burn" or degrade. Also some plastics materials retain moisture (abs, acrilyc) and without being properly dried prior to processing can lead to undesireable molding defects.

To put all of this in perspective, to mold say the case for a gameboy (two halves, very simple design. molded out of probably polystyrene or abs). Nintendo or whoever they contracted to mold their gameboy cases probably used at the very least a 100-150 ton molding machine with a 15-20oz barrel (about the size of a large SUV) to run a single cavity mold. To injection mold something like Ben's 360 laptop case would most likely require a 200-300 ton machine with a 20-30oz barrel.

I agree that the idea is awesome (who doesnt want to make thier own mod stuff). But i work in the plastics (injection molding) industry and i'd say that if you seriously want to mold your own cases you are better off finding a used cheap 25-75 ton injection machine you can buy and stick in your shop. I'm pretty new here and dont want to discourage anyone from trying to DIY anything they want to, but injection molding is a very precise process. Even if you did manage to build something that could squirt melted plastic into a mold. Unless you can do it very accurately, controlling the injection speeds and pressures along with keeping consistant clamping pressure on your mold you would not make anything but garbage anyway. You can get a decent (read probably used and abused but still functional) used 25-75ton press for 2500-5000$ maybe cheaper if u swindle em. Or you could send me your mold ;P

Just to elaborate on my first reply a bit. This is the only commercial "tabletop" injection molding machine I am aware of that exists: babyplast We have one in the engineering department of the plant I work in. It makes parts about the size of a thumbscrew or smaller. To try and mold anything larger than that in said machine would just result in disaster.

There are two books by vince gingery, one on how to make an injection moulding machine, and one to make an addition to a pillar drill to do injection moulding.  They are small though and can only do very small things.  Also you'd stand a good chance of burning the plastic or not heating it enough in the barrel before injection. I think it'd be more for hobby, or perhaps at most making small things for moddlers.  Every now and again a small table top moulder appears on ebay.

Premier Farnell Ltd, registered in England and Wales (no 00876412), registered office: Farnell House, Forge Lane, Leeds LS12 2NE.

Like I said in my first post I'm not trying to discourage anyone from trying to DIY. Only trying to shed some light on the subject of injection molding.

So I get your point that traditional injection molding requires massive machines and a lot of power. But some of this doesn't make sense.  If that's the only way it can be done, does that mean when a toy company makes a new toy with 50 different plastic parts that they spend $5-10 million creating the molds and molding the plastic?

While i agree that this could a worthwhile project, I would like to add some thoughts from my own work on my makerbot. On the makerbot the plastic is run through a very small heater section, and it has gone through a number of changes due to inconsistancies and difficulty of long term operation The current design is now quite stable and i've been running it for many hours without any problems to speak of. To scale the heater design up from one of those may prove quite difficult to get consistant quality. But mechanically according to the baisc picture of an injection molder on wikipedia, it seems simple to create, all you need is a motor attached to a wormscrew that moves the plastic pieces into the heater and the pressure from the incoming material forces the molten plastic out an oriface into the mold.

Another good example I can give you is tooth brushes. You would think that something as simple as a tooth brush would be easy and cheap to manufacture but the fact is that its just not so. The fancy tooth brushes you see today with the nice "grippy stuff" (TPE or Thermoplastic Elastomer) on the handle are made in multicavity "co injection" molds with state of the art robotics to insert the brushes and to removed the finished parts. These (usually massive) machines have multiple injection units and rotary "turntables" that swing the mold into a different position for different phases of the molding cycle. All just to make a simple toothbrush. The initial investment in these machines and molds goes into the millions. For something that to the end user only costs a couple bucks. Here is a simple example of this on You-Tube: http://www.youtube.com/watch?v=TJm_R2G-K_k

Imagine if you could mold the shells of your game machine laptops instead of gluing/stacking/screwing CNC'd plastic parts.

I think what makes this project unfeasable for most people is that you basically have to make every object you'd like to produce 2 times, which takes that much longer. So unless you are going to be doing a medium or large scale production of items its probably not worth the extra time to make a mold and then cast the part over just maching the part. Also you have to have the design right the first time as its much harder to make changes to a metal mold than to adjust a machined part through an additive or subtractive process. Again i think its a great idea and it be great for someone looking to make and sell their project on a larger scale, but for a hobbyist making a prototype or custom project, not as useful. But hey that's just my opinion.

Injection Molding is not the only solution to making a plastic part either. There is extrusion molding (pvc piping), blow molding (plastic bottles), compression molding (mostly used for rubber or similiar materials but also plastics in some cases) and combinations of all of these processes with one another and probably even some other processes that I'm unaware of. Vacuum forming is also another widely used method around the modretro and benheck forums which turns out some nice end products.

I can understand that you need a huge machine if you're going to use physical pressure to melt the plastic and ram it into the mold.  But what about using heat instead of brute force to melt it.  Then wouldn't you need just a simple piston type of setup to push the plastic out into the mold?  It seems like this can be done at a smaller scale without massive SUV sized industrial machines.

I should probably explain a bit how injection molding works. Injection molding machines are sold based on clamping pressure (tons) and barrel capacity (how much melted plastic can fit in the barrel in oz). Now the basic jist of injection molding is that material (plastic) is fed into the barrel of the machine where it is conveyed by a rotating screw through the barrel to the front of the screw whilst being heated to melt the plastic (90% of the heat generated to melt the plastic is shear heat and not heat generated by the electric heaters) esssentially building your "shot size" at the front of the screw. After the shot size is built in the front of the screw the screw then acts as a ram of sorts to inject the plastic into the mold. Now the best thing I can think of to compare this part of the process to is play dough. Remember when u were little and u had those play dough sets where u stuffed the play dough into a hole and then pushed it through with a ram to make spaghetti or w/e. This is about the same consistancy as melted plastic (or at least as close of a comparison as I can make). It requires tremendous amounts of pressure to push this molten crap into a mold and effectively "pack out" the part. The typical machine in my plant is 150-300 tons with barrel capacitys from 5-35 oz and the hydraulic injection pressures set on these machines can reach 10,000 - 20,000psi (this is at the ass end of the barrel, at the nozzle tip you can generally say this pressure is amplified by 10x+) This is why the clamping ends of these machine need such force to be applied to the mold halves to keep them closed during injection. without proper clamping forces you would just blow the mold open and squirt hot plastic everywhere.