PDA

View Full Version : 3D printing small housings



ebt
27-04-2016, 10:31 AM
Im pondering 3D printing some small housings for elecs, so before I charge into it like the proverbial bull.... has anyone else done this, and if so, care to share any experiences?

WFO
27-04-2016, 03:01 PM
I think you'll be lucky to get them watertight

Have you got a pot to pressure test them in?

Paul Evans
27-04-2016, 03:13 PM
Im pondering 3D printing some small housings for elecs, so before I charge into it like the proverbial bull.... has anyone else done this, and if so, care to share any experiences?

What sort of budget are you looking at, I know a guy who gets lots of stuff printed for Design models, the stuffs fantastic and not unreasonable, I,ll try to get the company details latter :-) , given the cost of machining would milling be a possibility? its got an awful lot cheaper?




Have you got a pot to pressure test them in?

English Channel :rofl:

ebt
27-04-2016, 05:02 PM
What sort of budget are you looking at, I know a guy who gets lots of stuff printed for Design models, the stuffs fantastic and not unreasonable, I,ll try to get the company details latter :-) , given the cost of machining would milling be a possibility? its got an awful lot cheaper?



English Channel :rofl:

Its prototyping to be honest, Im thinking the option would be to fling it at one of the guys printing as a service.... machining might well still be cheaper, thats what I was exploring...

I've got access to some pretty hefty testing facilities....but yeah, the channel works a treat ;)

LearnerDiver
27-04-2016, 05:17 PM
Its prototyping to be honest, Im thinking the option would be to fling it at one of the guys printing as a service.... machining might well still be cheaper, thats what I was exploring...

I've got access to some pretty hefty testing facilities....but yeah, the channel works a treat ;)

Give me a shout if you want some turning or milling doing.

WFO
27-04-2016, 08:22 PM
Getting it printed is for next to nothing if you just want to check fits and form factors, couple of quid depending what detail you want.

It would be really exciting if you could get prints done that would be pressure tight.

ebt
27-04-2016, 08:25 PM
I take it you've tried then? My housing plan was pretty simple, planning on double barrel o rings. From the 'sound' of your post this is ambitious for 3d prints?

Hot Totty
27-04-2016, 08:56 PM
PM reallymadrob he plays about 3D printing camera stuff and may have some insights

nigel hewitt
27-04-2016, 09:07 PM
We use a 3D printer to prototype shapes but the surface finish isn't something I'd trust for a seal.
Can be brittle and you can't make them solid (the program matrixes the inside) or they warp.
Perhaps you can get a better printer than ours.
Great for deciding what something should look like though.

Paul Evans
27-04-2016, 09:51 PM
I take it you've tried then? My housing plan was pretty simple, planning on double barrel o rings. From the 'sound' of your post this is ambitious for 3d prints?

Very at a sensible price point. Sorry been gossiping in town with mates, are you prepared to share the model so I can assess best method. Almost certainly needs to be machined for the rings :-) On trian now will pm tommorow. I have some gurus on Manor royal who can probably help :-D

Sent from my D5803 using Tapatalk

ReallyMadRob
28-04-2016, 09:48 AM
In my experience... typical 3D printing is good for structural, planar parts but as it's built up in layers is poor at staying non porous or dealing with twisting loads. Although I can print solid I reckon you'd want to do further finishing before you trusted even a small part to stay dry inside - that might include vapour smoothing with something evil :-) That's just to get the body of the print less porous, surface finish would need work to mate faces and o-rings too.

It would be great for prototyping but you'd have to overengineer and process prints... whereas machining solid material should get you waterproof for free :-)

Rob

WFO
28-04-2016, 10:37 AM
I take it you've tried then? My housing plan was pretty simple, planning on double barrel o rings. From the 'sound' of your post this is ambitious for 3d prints?

Yes in my opinion it is ambitious.
However the thing I had printed was only intended to be a check on the form factor of the computer... it was done with melted filament printing.

Something done on a better printer like one of the resin layering ones or the fancy laser sintering method may work.

On the other hand... buy a block of acetal from one of a good range of suppliers, get the decent stuff that's been ultrasounded to make sure there is no coking porosity in the middle of the bar and then mill it out... dead easy and consistent.

ebt
28-04-2016, 12:48 PM
Thanks fellas, confirmed my suspicions there.... I'll go back to the machining options.

cprobertson1
20-09-2018, 07:25 AM
--EDIT--
Whoops, I just realised I necro'd that post - sorry folks - I thought it was far newer O_o - should have double-checked the dates! Lesson learned!

Hey, thought I'd share some of my experiences 3D printing small pressure vessels for hydrophones that have survived down to 50m. I eventually moved over to plumbing parts (PVC pipe with a solvent-welded cap at one end, silicone penetrators for cabling, and a screwed connector covering the other side for servicing (the transducer is a separate module, hence the penetrators!) - I ended up potted my electronics in resin because the mica capacitors were doing weird things under pressure! :rofl:- which doubled as water ingress protection - so you might benefit from that as well.

Anyway - some tips that may help.

Get the thickest layer height you can: fewer layers = fewer potential ingress points
Over-extrude: I over-extrude by 150-200% just to fill in the gaps
Sealing: if printing ABS, either friction weld or melt some plastic onto any cracks you see, and then acetone smooth it. For PLA (which I use), I find coatings work best. If you preheat your print in some water to somewhere below its glass transition temp, you can then dunk it in molten wax (I use paraffin heated in a double-boiler (a pot inside another pot with water in it)) - the preheat stops the wax clumping together on the surface. Polyurethane varnishes may work, but the adhesion isn't great on them.
Annealing: annealing (https://rigid.ink/blogs/news/how-to-anneal-your-3d-prints-for-strength) works wonders for pressure vessels. You heat your oven up to about 5 degrees above the glass transition point of your plastic and leave it there for an hour (with nothing in your oven by a baking tray - want to get everything an even temperature) - turn the oven off, put your part on the tray, and leave it to cool over the course of several hours) - works WONDERS for most prints - BUT be aware, that your print will expand in some axis, and contract in others as the material relaxes.
Engineering: avoid sharp corners (which act as stress concentration points, and don't always seal correctly in the first place!), circles are your friend, and go for solid infill. If printing a cylinder, try to make the ends hemispherical if possible, and you may find it worthwhile to put reinforcing rings on the inner or outer surface depending on your design.
Pre-stressing: you may find it worthwhile to pre-stress your cylinder by printing two cylinders, with an internal overlap (so, the OD of the internal cylinder overlaps the ID of the of the outer cylinder by, say, half a millimeter) - you anneal the inner cylinder as above, and once cool, you then heat the outer cylinder to its glass transition temp, slide it on, and then rapidly cool it causing it to contract and stress the inner cylinder, in a hopefully even fashion!


The bodies for my hydrophones are just using PVC tubing these days though! Depending on your application, you may want to look at them too - the section-40 stuff can take something like 40 bar without problems (actually, when I hydrostatically tested it with an hydraulic pump, it failed at 80 bar (internal pressure, mind) - but I err on the side of caution. You can get cable penetrators for them easily enough (just go through the end-caps - be aware that coax can act as a water ingress point if you get a leak near the surface, it will happily flood into your sealed unit... I found that out the hard way!)).

You may also find it beneficial to "float" a bit of pipe inside another bit of pipe - effectively double-sealing it (as you need two penetrators to get through both sides)


Anyway - have fun! Do let us know how you get on :D

You may also want to check out the EEVBlog forums (https://www.eevblog.com/forum/index.php) for tips - they're always happy to help with electrical projects, including enclosures!

----

EDIT: copper/iron/lead tubing is also a way to go - I've had some success with the 38.1 (1.5") tubes - though they look a bit dodgy if you have wires coming out (like, the police will not like you playing with them kind a dodgy - paint them bright colours and clearly mark what they are!) - they were hydrostatically tested (to death) and survived to around ~100 bar (properly hydrostatically tested, too! We make oilfield parts at work and sometimes we need to pressure test things - we have a concrete bunker with blowout valves for it!).

Anyway, under testing, the failures were all in the penetrators and the transducer seals, rather than the electronics vessel, with that design at least. Resin is your friend for sealing penetrators, by the way! You can also just buy better penetrators (make sure you get the right size for your wires or you will NOT have a good day with them!)

WFO
21-09-2018, 11:36 AM
Since this thread we bought a HP fusionjet printer, which prints solid.

I've printed a battery box for the caving lamp which has done fine down to 2m, but am short of time to play with much else. Happy to print some bits on this machine if anyone else wants to play though

cprobertson1
21-09-2018, 01:30 PM
Since this thread we bought a HP fusionjet printer, which prints solid.

I've printed a battery box for the caving lamp which has done fine down to 2m, but am short of time to play with much else. Happy to print some bits on this machine if anyone else wants to play though

VERY nice! I'm totally jealous!

Just out of curiosity -Is that an FDM or a stereolithography unit? I see you can get direct metal laser sintering processes for it too depending on your application!

As I said, totally jealous ;)

WFO
27-09-2018, 03:47 PM
None of the above, it is thermal fusion of powder (so no support material) but 100x faster than SLS type powder bed machine