View Full Version : Injection Molder - finished

05-11-2003, 06:20 PM
Well, I finished the injection molder last weekend and made the first mold. A simple slug mold to turn scrap plastic into 1/2" slugs to feed back into the machine. I have images of the finished machine and the first mold and some pieces injected with it.

I do have some changes to make now that it's working. I'd like to drop the heater cartridge a little deeper towards the nozzle and I need to fix an oversized bore in the upper bushing block (poorly sharped drill bit). I'd like to change the jack screw to a toggle clamp for faster, easier clamping of the molds.

The thing gets pretty hot so I have it on some firebricks. The bottom plates only gets warm but the upper plate get a couple hundred degrees easy. I need to wrap parts in some thin insulator material. The top plate makes a pretty good place to set a mold to preheat it. I've read some people use a toaster oven to pre-heat molds.

overall it's a pretty easy machine to make and operate. You have to know where your hands are when you move around it when it's up to heat. I got some metal donated but i bought the controller off of Ebay and the heater cart from Gingery. Over all I have about $70 raw cash in it, plus tooling that I just had to have to finish it. What a shame and no, I'm not married.

05-11-2003, 07:04 PM
Those finished parts look really nice. It makes me wanna go to my garage right now and start making one.

Several questions.

1. Does the mold have any "risers" where the gas and excess material is allowed to squeeze out?

2. Did you use any parting agent?

3. Is the heater designed essentially to heat the entire apparatus along with the plastic to the working temperature, or is there intended temperature gradience between the plastic in the cylinder to the mold?

4. What's the working temperature?

5. Did you find that the amount of pressure makes a difference to the quality of the finished part?

6. How long is cooling period, so that the molded part can be removed?

Needless to say, I'm very impressed.


05-11-2003, 08:54 PM
Very Nice! The toggle clamp will make things go quicker. Try using it around some kids. They will be amazed. Wheel molds are fun to make little toy cars, and initial (letter) molds are fun with the kids too. My complements to the builder. A1 job.
regards, Matt

05-11-2003, 11:51 PM
Spiffy! I look forward to seeing some finished parts made with it! Good job!

05-12-2003, 01:38 AM
Thanks guys, next will a couple serious molds.

1, nope but I do know they are used to let gas/air escape in the big boys because the trapped air can become so pressurised that it can combust in the mold and burn the plastic. They work at much higher pressures than I am.

2, no agents, just polished it.

3, yep brute force with the heat. No gradients on this one.

4, temp runs between 350-450 deg F depending on plastic. I probably have to run higher than an industrial unit would as there is no plasticating screw to aid the flow or a heated nozzle. So it runs purely on residual heat. Heatig the mold is a must. Think of lava tubes. The plastic forms a skin along the walls and the hotter plastic runs through the middle. As it cools the passage gets narrower until it shuts off and if the mold isn't full yet then it's never going to be.

5, the finish depends on the finish of the mold. The pressure is used to get the plastic into the mold as fast as possible. Also you hold pressure to prevent sinks until the plastic sets up.

6, cooling was pretty fast under the garden hose otherwise the plastic would stay flexable for probably several minutes in the mold just cooling it in air. I couldn't wait that long.

Needless to say this isn't a production machine, though I do believe a HSM could make a decent screw type injector. Probably won't be as efficient or tuned but no reason it wouldn't work.
But hey, how many people can injectino mold plastic at home? It is pretty cool to pull the part out.

05-12-2003, 02:40 AM

05-12-2003, 12:59 PM
Brunneng, thanks for the reply. If you get a chance, can you post more pictures. It certainly gets my creative juice running. Thanks for inspiring us all.


05-12-2003, 04:31 PM
Yes, it looks good. Makes me want to get the hydraulic press out and fabricate one too. Perhaps on the english wheel frame I use for everything else, ha ha.. Yeah. good ideal. Good implementation into reality.

Most people can find thousands of reasons why something WON'T work. Few can make it happen.

How are you going to make your dies? DO you dremel them? or?
Are you using a relay or relying on the controller to handle the 750+ watts of the heater? Most won't do that for long.

[This message has been edited by ibewgypsie (edited 05-12-2003).]

05-12-2003, 05:26 PM
I'll put up some more close-ups.

The first mold was made on the mini-mill as will the next. I have a rotary table coming in for making arc's for the sprue's corners. just old fashioned machining. Slow and tedious.
I'm building a cnc mini-mill to make molds (4"x4" max) and small parts. I'd like to make it around sub 0.25" tooling and make it high speed (6-40k rpm) tho I know there's a lot to doing that with heat and balancing plus harmonics and vibration. Ready made spindles are out of my budget.

The heating element is 250 watt and is plenty for this size. The main block is 1"x2"x4" cold rolled. The controller switches the element via a 360 watt relay which I can beef up.

I'm working up the design for my own controller using a chipset designed just for thermocouplers. It is $19.95 for the 2 chips which is a self compensating link and temp stable amp to feed to an AD converter and micro controller. It supports 4 types of probes. This way I can create a unit with adjustable curves and multi-zone control for much less than the thousands industrial units go for.

I have so much more respect for mold makers now. I can't believe what they could make before cnc.

05-13-2003, 04:51 AM
Brunn.. did you see the post I sent alistair?

It looks to be like inline roller skate wheels with a ball bearing drawer slide for the cross. The router and the probe follow each other. BUT, I bet with a lil ingenuity a person could make a reverse model with the tool on the other side. This would enable a die mold to be carved from existing items. Ie: reverse engineering.
I have revamped a cnc, also I plan on making a digitizer for it to read shapes into 3d files. As you probably are aware, the software development time kills you.

I was and am, really watching this post since it answered a couple of questions I had in a previous one.

I plan on making a similar machine, later..

You should think of putting a relay in for the heater. I would. The controllers I have had to work on, have a reed type relay. very small, the surge will kill them.

If you find a process to linearize and amplify the thermocouple, holler. I was told there is one ic chip that does all that, cold junction compensation and all. I have 4 or 5 Parallax BS2's laying around here. With a analog-digital converted input it would be easy to make the controller you speak of.
I have had experience with the binary type wheel input also with them. That is the easiest way to input a number.
Using a RTD type tempature sensor you could use the Pot command directly to read tempature.

I got my last PID controller from Grainger, they were as cheap as the ones I saw on ebay.
I just wish it had a pc interface made into it.

05-13-2003, 02:09 PM
For die shaping in that manner I believe what you want is a pantograph style machine. Usually of heavy double bar construction. That way you can work off of a larger master, say 2x or 4x size out of plaster or wax and reduce it at the cutting end. Like this one:

The heater is controlled via a relay. If it gets to hot in use, I'll replace it with a bigger unit.

I started playing with a Parallax stamp but now I program the pic chips directly. More speed and control. The chipset I'm talking about is exactly that - thermocoupler interface and cold junction compensation in one chip, the second chip is a precision op amp with very high temp stabilization to boost the coupler output to feed to an ADC. It made by Linear Technology #LTK001 Thermocouple Cold Junction Compensator and Matched Amplifier. The tech sheets are at:

05-13-2003, 05:20 PM

Geeze did you see the price on the pantograph.. choke choke choke.. I could get a baker 6 speed.. Thanks for posting it tho, it was nice to look at, like a young pretty girl. whoo hoo..

I plan on putting a touchy feely probe on my cnc bridgeport to scan items. I hope it don't cost me that much.
I have drawed it from simple slide potitemeters to slide encoders to rotary encoders with pendulum. Not sure yet, I am kinda busy anyways. I can do the point Visual basic program, also the inversion program needed to make a negative from a positive impression.

I'll look into the chipset, I just sold a bunch of opto-22 stuff on ebay tho. (ifn I was ready I would have used what I had)

I am up to my neck in this casting outfit, under a lot of pressure for results.