Cost effective way for manufacturing a small plastic part

[Bob n AboutParticipant @bobnabout]

Or put another way, about 18 seconds a part, for printing.

Edited By Joules Beech on 29/03/2017 12:23:17

[Bob n AboutParticipant @bobnabout]

Having just drooled over the spec for a Form 2 SLA printer, it might be worth finder an owner of one and see if you can strike a deal. 20 x 20 x 5 array of parts should print in about 6hrs using highest quality setting. Thats 10.8 seconds per part, and this is how 3D printing can compete with injection moulding by bulk printing.

[Nick HulmeParticipant @nickhulme30114]

Posted by Joules Beech on 29/03/2017 13:23:30:

Having just drooled over the spec for a Form 2 SLA printer, it might be worth finder an owner of one and see if you can strike a deal. 20 x 20 x 5 array of parts should print in about 6hrs using highest quality setting. Thats 10.8 seconds per part, and this is how 3D printing can compete with injection moulding by bulk printing.

With a resolution an order of magnitude off the OP's tolerance aspiration? 😀

[Bob n AboutParticipant @bobnabout]

LOL, out of interest I just priced up having them done at a bureau printing service and the estimated quote came back at £870+vat So my offer for an SLA printer was far from wild as that would include my CAD work… Oh well…

x7 3m x 6mm nylon rods for £8.54 and cut the parts yourself accurately…

The level of accuracy asked I would be looking at Acetal or Delrin to make these parts.

Edited By Joules Beech on 29/03/2017 14:04:39

[ZebethyalParticipant @zebethyal]

Might be worth obtaining a quote from someone like Proto Labs, they can do 3D printing, CNC machining or injection molding.

They specialise in low volume injection molding, where it is often more cost effective than either of the other two methods they offer.

[Jeff DaymanParticipant @jeffdayman43397]

In my opinion, the part design as presented is far from optimal for injection moulding. Wall thickness varies greatly, and one hole will need either a slide or a hand loaded insert to make. Holes would need minimum 1/4 degree per side draft angle. It is not apparent where gating and ejection may be on the part, these details would need to be discussed. I would improve the part design before starting discussions with Protomold or other RP moulding firms.

It may be possible with an improved part design to hold +/-0.05 mm tolerance on hole diameters in nylon, but hole to hole and overall dimensions will probably have a larger tolerance, more like +/-0.2 on length.

There are post moulding techniques that can be used to reduce the hygroscopic tendencies of nylon over time, but the engineering grades of it, like DuPont Zytel, will be more dimensionally stable than cheaper nylons.

If we knew more about what this part needs to do, perhaps other more stable resins could be suggested which would be more stable dimensionally. JD

[Nick_GParticipant @nick_g]

Posted by Joules Beech on 29/03/2017 13:44:07:

The level of accuracy asked I would be looking at Acetal or Delrin to make these parts.

.

There is a chap who has recently made some parts very accurately and very nicely finished out of Derlin for me.

PM me if you want his contact details.

Nick

[Neil WyattModerator @neilwyatt]

Posted by Joules Beech on 29/03/2017 13:23:30:

Having just drooled over the spec for a Form 2 SLA printer, it might be worth finder an owner of one and see if you can strike a deal. 20 x 20 x 5 array of parts should print in about 6hrs using highest quality setting. Thats 10.8 seconds per part, and this is how 3D printing can compete with injection moulding by bulk printing.

10.8 seconds! I find that difficult to believe, even with an array of five printers! Don't just extrapolate from a volume as there's a lot of time wasted jumping between the 400 parts, and even with that the chance of a fail amongst 400 parts that then ruins a large proportion of the others is very high.

On the other hand, I suspect that, depending on the final use, you may be able to print to a suitable tolerance, it's certainly possible to print PLA so that a steel rod can be a sliding, push or force fit. I printed (around) 6mm holes top take a 6mm plain shank bolts that needed tapping in with a nylon faced hammer. Holes down to 1mm were feasible.

You will need to experiment with settings.

Neil

[HOWARDTParticipant @howardt]

CNC lathe with c- axis and bar feeder.

[Bob n AboutParticipant @bobnabout]

Hi Neil, SLA is not the same as FDM (filament) printing. Some SLA printers use DLP or LCD screens to expose a complete layer at a time and these things can really motor through layers. Carbon Clip is another form of SLA, but its prices are way out there. However what SLA machines do in hours Clip can do in minutes. Worth looking it up just to have your mind blown.

That means CLIP can print each part in 1.8 seconds, 400 per layer, is your head spinning now…

Edited By Joules Beech on 29/03/2017 19:10:49

Edited By Joules Beech on 29/03/2017 19:12:38

[Neil WyattModerator @neilwyatt]

Forgive me, my eyes read SLA but my brain thought PLA!

Neil

[Mark BarronParticipant @markbarron78673]

For 200 pcs, I would suggest a low-volume injection mould tool. Maybe give this company a call:-

http://www.manumolding.co.uk/

An aluminium tool would suffice and should be cheaper than steel.

Thanks,

Mark.

[Mark BarronParticipant @markbarron78673]

Sorry, for 2000pcs, not 200 pcs.

[Author]

Posts