I’d like a simple plastic injection machine or designs

[Russ BParticipant @russb]

I'm looking to reproduce some plastic pipe connectors, I can very easily source standard Tee's/Y's etc but I really want an exact reproduction of what the machine should have and the parts are incredibly basic in shape.

Does anyone have a simple DIY arbour press type plastic injection molding machine they'd consider selling?

I have been looking at making a simple aluminium cylinder with a pair of chinese ceramic heaters and a close fitting piston, a basic chinese PID temperature controller, fitted to a fabricated arbour press type stand, simple, cheap, effective – but it would take me a while to make, so I thought I'd ask if anyone has one they don't use anymore.

Regards,

Russ

[Russ BParticipant @russb]

[Russ BParticipant @russb]

Well I've drawn up some plans of my own as no one came up with anything!

The design I've put together looks like it should do the job, I have no injection molding experiance but some knowledge. A friend of mine (also an engineer) used to specialise in injection molding so I'll be running the design and numbers past him and see what he thinks, its obviously going to be a long and crude way away from what he's used to!

The hand lever arrangement I've designed offers around 3.5kpsi of injection pressure with a relatavily achievable 50kg pull on the handle. I just need to check the stress on several other components as the handle and lever had to be very substantial after looking at the stresses involved.

I've ordered a pair of temperature controllers, one for heating the cylinder, one for heating the die (as I expect 3.5kpsi of force won't be enough to inject my rather large part into a cold die.

Once it's up and running I'll make several "rods" of plastic for tensile strength testing and compare strength and cooled size vs die temperature to see if there is much effect on finished strength. I may also connect a loadcell to handle and measure strength vs injection pressure.

I think it'll be ok, it'll be interesting to see strength vs die temp and how much shrinkage is observed.

[Nick Clarke 3Participant @nickclarke3]

AFAIK there was a design in the model railway press years ago where the cylinder of the 'arbor press' was surrounded by the bit of a massive (125W?) Henley Solon soldering iron which had been machined to fit closely.

Could this be brought up to date?

[David JuppParticipant @davidjupp51506]

There's a book available written by Vincent Gingery that might provide a useful starting point.

ISBN 978-1878087195

[OldironParticipant @oldiron]

Would this be of any interest or help ? ARW injection moulder

regards

[Russ BParticipant @russb]

It might be worth checking out, do you recall when/where it was?

I have pretty muich done the basic design, it all used either M6 tapped holes or M6 clearance. I haven't made the die support yet as I'm not sure the square column will take the stresses involved as it was much much higher than I anticipated, so I will check the column isn't going to bend or stretch, then take it from there!

[Jeff DaymanParticipant @jeffdayman43397]

I would not suggest following ANY advice from ARW about injection moulding. He clearly has no clue about building a safe to operate moulding setup or how to operate it.

The Gingery machine has been built many times and does work for very small parts – but a common complaint from builders/users is lack of part to part dimension and strength control.

One key aspect of injection moulding machines that people seem to miss is that the mould needs consistent and even clamp force across the whole mould, and the mould and injection barrel need to be kept aligned. Setting a mould loose on the beam of a hydraulic jack press and having the injector cocked at a strange angle under huge force is asking for an accident.

To calculate the clamp force needed, to need to find the projected (plan view) area of the part then multiply by your max possible injection force. Multiply that number by two or three and that is how much clamp force you need as a minimum. If your planned parts are large you will be surprised at the forces involved.

For any consistency of strength and density in your finished parts, barrel temp, mould temp, injection pressure, heating time, injection time, cooling time (at a minimum) will need to be controlled. Most resins do not do well sitting in a hot barrel between shots either (called "holdup time". Feed resin, heat and shoot, repeat, with no wasted time, with carefully controlled parameters, is the name of the game for consistent parts.

If I were you I would look into making a mould or moulds to standard industrial practice and handing it to an injection moulder to run in a commercial injection moulding press. Find and talk to a moulder and he will steer you to what mould dimensions and construction will run in their presses. The moulds are where the art is. Think of it as if you wanted to make turned parts – would you build a lathe or use a ready made one? Good presses are already out there at injection moulding firms. Fewer such firms now in UK and North America than there used to be, but still plenty around.

Just my opinion from many years in industry involved with plastics design and moulding.

Edited By Jeff Dayman on 02/11/2020 15:07:47

[Russ BParticipant @russb]

Jeff,

Thanks for the input, regarding clamp force, I had planned on justing bolting the two halves together, as I don't mind making one, then waiting for it to cool, taking it apart and making another, I'm not really trying to brake any speed records, maybe half a dozen in an hours, and I'm happy.

The barrel/nozzle are fixed, I'm relying on a brass piston in an aluminium bore to maintain it's own alignment, it's a sqaure piston diameter to length, 25mm diamter, 25mm skirt, I could easily extend that skirt to 30-40mm if I have any problems with rev1.

It sounds like I won't be getting the finest quality parts, but I will be testing and gather quite a lot of data and see what works best. I will keep in mind "hold up" time, obviosuly it's a 1 shot wonder, but I don't know how long I need to leave the material in the got barrel before I consider it to be thoroughtly melted and ready to go so given your comments, I'll test parts for strength having given them maybe 2, 4, 6, 8, 10 minutes heating up, see if a sweet spot appears??

Edited By Russ B on 02/11/2020 16:08:38

[Jeff DaymanParticipant @jeffdayman43397]

Hi Russ, You don't mention what type of resin you want to mould. Some resins will tolerate several minutes of holdup time before starting to degrade, and others less than 30 seconds before they start to degrade. (polycarbonate is one example that degrades very quickly).

Knowing your resin's melt temperature and using a thermocouple / readout on the barrel would go a long way to knowing when to shoot the resin. If you mount the mould such that it can be lowered away from the barrel between shots, you could fill the barrel with some extra resin than needed for filling the mould, and do a test push on the lever to see if your shot melt is flowable, onto a scrap piece of card or wood or metal sheet. If you see that it is ready, you would raise the mould and shoot it asap. Commercial presses have the ability to pull the barrel and nozzle back from the mould to purge out any degraded material and ensure fresh, ready to inject melt is in the barrel prior to production run starts.

Again if you went the route to make only the moulds you need, designed for use in a commercial press, you would not need to worry about many of these things and the mould would have ejector pins to eject your finished parts and be ready for next shot in seconds, rather than having to undo bolts in your moulds. In addition you would have full control of your process temps and timing to control your parts' mechanical properties.

BTW mould shrinkage varies with resin type as well – you can google resin brand names through Matweb.com to find the general shrinkage factor for mouldmaking.

[Jeff DaymanParticipant @jeffdayman43397]

Hi Russ, one further note – getting your piston out of the barrel to reload resin, if the piston is loose and not joined to the connecting rod to the lever, is likely to be a real pain and a sticky mess depending on the resin used. Getting everything clean enough between fills will also be difficult. In any case I would suggest having the piston joined to the lever so you can use the lever to get it out while hot.

Again using an injection moulding press eliminates all the faffle above and will result in much better part properties. The moulder will also likely share a huge amount of plastics expertise with you in your first discussions about the parts you plan to make.

I suggest getting someone you know to make 3D printed models in full size of the parts you plan to make, and if you visit a moulder or moulders take them with you as well as info on what resin you would like to use.

[Chris Evans 6Participant @chrisevans6]

As another who spent his working life in the injection mould game a agree with all Jeff Dayman says. Have a look at some of the standard mould base kits around from the likes of DMS (Die Mould Services) Hasco and DME (Die Mould Europe. A small 150mm square mould base is cheap and for all you need a mild steel (Din1730) one will do. Find and talk to a commercial outfit before you commit. You will be looking for someone with small say 25 ton press.

[Russ BParticipant @russb]

Chris/Jeff,

What if I only want 10-20 of a particular part, what are the costs involved?

Jeff, the piston is connected via a gudgeon pin, you can't see it in the screenshot as it's the same colour as the conrod, I just modelled the one in for stress anaylsis, the opposing pin will obviously feature equal and opposite stress. The head of Cylinder is flat and removable to aid cleaning when the jobs done.

I have one of the finest, if not the finest FFF 3d printer on the market so I can print the model myself. I'm looking at a small 6mm pipe connector with a 1-1.5mm wall thickness. We're looking at about 55mm total length, the most difficult part I want to make is a 4 way cross.

Chris, it's always good to have some companies to call up, thank you, putting injection moulding terminology and quieries into google is a waste of time, I was trying to find out how NBR parts are made the other day to see if I could make my own and it was like I was asking the meaning of life, every answer was totally different and none hit the mark!

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Russ B on 03/11/2020 08:04:10:

,

I was trying to find out how NBR parts are made the other day to see if I could make my own and it was like I was asking the meaning of life, every answer was totally different and none hit the mark!

.

Is this any help, Russ … it seems a good basic overview :

**LINK**

https://www.rubbermolding.org/nitrile-rubber-molding/#read

MichaelG.

[Old SchoolParticipant @oldschool]

I spent a long time in the plastics in my opinion you are wasting your time. In one of my laboratories we had a small simple injection moulder, made by a well known test equipment manufacturer for moulding tensile test samples what it produced was very inconsistent. You would get short shots voids etc in mouldings lucky to get one in ten that was any good. No good for test work. We tried to use it for polypropylene and high density polyethylene. The machine ended up in the scrap metal skip. It was replaced by a proper small injection moulder that made consistent test pieces, filled the hopper with plastic pressed the start button and left it to it.

Good luck with you your project it will be interesting and probably frustrating, injection moulding is a interesting process but not certainly not simple.

[Mark BarronParticipant @markbarron78673]

Plastic injection moulding is straight forward if you have the correct equipment. If you're making a 'home brew' moulding machine the chances you'll achieve a good quality part without spending a large amount of time and money are minimal.

If you only need a handful of parts, have you considered rapid prototyping using the SLA / SLS process?

If you definately want to use injection moulded parts, there are some Chinese toolmakers who will do this for you via Alibaba.

[Jeff DaymanParticipant @jeffdayman43397]

Hi Russ, I don't know what mouldbases cost in UK but in Canada a basic mild steel 150 x 150 mm mouldbase from Hasco would be the equivalent of 500 UK pounds, before you cut any metal. To do a thin walled 4 way cross 55 mm over ends you will likely need 4 side actions or 4 hand loaded low draft angle cores in the tool to "hollow out" the part. This is not beginner level toolmaking.

The way many moulding shops operate is they have a setup charge to hang and prep a mould in a press, of about 500 dollars in Canada or 300 UK pounds per mould. It will take 20 shots or so just to get the mould parameters set and parts running smoothly. The moulder will likely have a 250 part minimum order qty of parts per run. For such a low quantity of parts I don;t think a moulder would be interested in handling multiple hand loaded cores, they will likely want an automatic running slide core mould.

One way or the other, I think to do several moulds and parts you are looking at a multi thousand pound expense. Most people setting out to do injection moulded parts do so with making and selling many thousands of parts in mind, to offset the expense. Of course these parts need to have consistent quality if you are selling them. Are the fittings you want to make marketable to others for a particular machine?

An alternative to injection moulding these parts may be to make a cast silicone rubber mould off a 3D printed master and cast urethane parts in the silicone mould. Excellent results with good detail and finish can be done with this method. Many types hardnesses and colours of urethane resins are available, some look and function very like inj moulded parts. Since this is a no heat no pressure method it is easy to do in a home shop. Google Smooth-On and Hapco for silicone mouldmaking and resin info.

Re your NBR part enquiry – many rubber parts are made by compression moulding. A two part steel mould is made, usually hinged to open like a waffle press. Raw rubber strips are laid across the open mould by hand. A press with heated platens clamps the mould shut and exerts tons of pressure and high heat to cause the rubber to flow and fill the mould. After curing, mould opens and parts are picked out of the mould and the excess rubber flash removed. For large volume production often many moulds are arranged on a turntable and rotate from station to station to get a shorter cycle time. For small parts the moulds may be multi cavity tools and make lots of parts per cycle.

If you want to make just a few fittings for your own use at min cost I suggest machining them from plastic or metal.

[Russ BParticipant @russb]

Jeff,

As I suspected, very expensive to do properly! I already have polyurethane and silicone casting equipment as I made a valve head and cast a PU sealing face onto it for a small (by my standards) injection machine (typically pulverised or pellet fuel injection)

I might contact the local supplier to see what other options might be available as the PU i have would be too soft although probably chemically compatible.

I’m still going to gave a look at this DIY machine, i’m too curious…..

[Mark BarronParticipant @markbarron78673]

Russ,

I was thinking that we used this company about 25 years ago for very small batch prototype parts. May be something similar to your requirements: http://www.manumolding.co.uk/manumolding

[Chris Evans 6Participant @chrisevans6]

Jeff, where in Canada are you ? Many years ago I visited a few toolmakers and moulders in the Toronto and Windsor areas.

Chris

[Jeff DaymanParticipant @jeffdayman43397]

Hi Chris, now in Waterloo Ontario. Started work in a mould shop in Windsor early 1980's and later worked many years in Toronto. Most of the smaller inj moulders and toolmakers are gone now, most out of business due to low cost Chinese made tools and parts. Still a few big specialist plastics firms and a few very good general moulders in Ontario. Lots of moulders doing high volume low profit ultra QC sensitive car parts. Also a few very small very specialist moulders for "not the usual thing" plastics. I still work with several types of plastics firms except the car part makers.

[Chris Evans 6Participant @chrisevans6]

Hi Jeff, it's been about 25 years so I can't remember any company names. One was a "Hot Runner" specialist and they where running multi cavity thin walled margarine containers cycling at about 5 seconds. I managed a toolmaking company for over 20 years, mainly intricate stuff (2 shot/3Shot) stuff no one else wanted to do. We also did mobile phone moulds but as you say the cheap Chinese stuff killed the trade.

Chris

[Jeff DaymanParticipant @jeffdayman43397]

Hi Chris, one big name in the thinwall packaging container business was Stackteck, they are still in business, along with a lot of other package / container and closure moulders. Husky and Mould Masters were the big guns in hot runner systems for many years. Still operating but I think smaller scale these days.

I have also done some work with multishot moulding, especially for TPE or TPV overmoulding. Had several large structural / cosmetic mouldings projects with gas assist and also coinjection two polymer moulds (nightmares). One interesting project was a family of moulds for two shot rigid plastic keycaps for industrial keyboards. The letters were moulded first in cav 1, usually in white or yellow, the shot lifted up and set back into cav 2 with a helical cam, then the cap shot around the light coloured letters. The second shots were all colours. Sealoffs at the colour boundaries was required to be zero flash. OK once we had it dialed in. The idea was that the letters and numbers could never wear away, in critical reliability systems ie powerplants and military stuff. That was OK, keycaps lasted forever, but people still pushed the wrong buttons once in a while!

[Chris Evans 6Participant @chrisevans6]

Jeff, I am sure it was Husky that I went to see. Keyboard keycaps, we made around 30 16 impression mould tools for IBM keyboards. That's 30 for the main button and 30 for the printable cap. CNC spark macines came in for the multi cavity stuff. Long retired and out of the game now.

[Author]

Posts