Saturday 16 November 2013

The last year and 3D printed pressure vessel.

I haven't done any rocketry at all this year because i've been terminally busy working on a startup for inkjet printing PCB's in silver nano-particles. We've released the printer now so hopefully I should have some time for rocketry though it won't be much. We released the printer on kickstarter, you can find it here. Everything considered it was a pretty interesting year. I got to learn about business stuff and other interesting things I hadn't really cared much for until now. Also I learnt a bunch more about electronics! Readers who are that way inclined may also notice an improvement in my spelling too! Though I am not sure if I can keep that up here...

Buren joined the airforce at the beginning of this year and spent most of the year away doing training but he is back in Brisbane now and keen to get back into rocketry. We had lunch yesterday and talked about a new project we could work on. I have basically no money at the mount so a big project is out of the question but we came up with a few ideas for smaller projects which haven't been done before, mainly neighbourhood of 3D printing.

Call me a mechanical engineer but in the past I have been pretty sceptical about the usefulness of plastic 3D printing. I never really saw any application for printed parts as structural components. In the company I was exposed 3D printing fanatics and I think I caught the bug. They got me thinking about printing high load bearing parts from plastic. Something I thong would be really cool is to print an entire hybrid motor (tank and all) in one piece.

This isn't as crazy as it might initially seem. Porosity is probably the biggest issue but airtightness can be achieved by sealing with epoxy or partially dissolving it with a solvent depending on what sort of plastic you're using. Strength was my main issue but even if you assume the strength between layers is 5% (papers say 5-20%) you can just make parts thicker. They will be heavy but there isn't any reason why they shouldn't hold up. Heat is obviously a major issue but for a hybrid I think it would be possible with a clever injector design. So I think printing an entire  hybrid from plastic would be a really interesting challenge and not too costly either as I have pretty much everything I would need.

I figure the hardest bit will probably be the pressure vessel. The other day I made a simple test vessel. Its pretty basic, just a 20mm ID sphere. I don't have my own 3D printer but we have one I can use at work. Unfortunately it only can print in PLA. The wall thickness is 10mm. Assuming a inter-layer strength 10% of low strength PLA (difficult to tell what I have) it should be good to 850PSI (leaks will be the main issue) I think there will be much better materials than PLA (have ben doing a bunch of research) but for now I will have to put up with it.





The first attempt was printed at a density of 20% and it had a bunch of leaks. It skipped a layer half way which probably didn't help so for the second attempt I printed at 100% density. I made the hole for the fitting slightly smaller so I could drill and tap it. In the end I just epoxied it in because I didn't want it flying out. I would like to do some tests on tapped threads in the future. I am not sure I will ever be able to print a thread using a hobbyist printer but thats something to explore too! I tested it with compressed air @120 PSI and there was a really small leak. I was pretty happy with this as the printer I am using wasn't set up well and there are pretty big gaps between layers. I should be able to improve this with tweaking. I ended up pouring a bit of super glue in which plugged the hole. Now it doesn't leak at 120PSI. Unfortunately I don't have any source of higher pressure to test it currently. The startup is currently in a large industrial complex owned by a university and there are a few gas cylinders....around. One night I went out to try a few but unfortunately they are all empty....I've been thinking about buying a high pressure dive compressor for a while now, this would be a good excuse!

I think I will also buy a 3D printer just for experimentation. I also want to get a bunch of different types of plastic to test with. There are some nylon ones with a fairly high UTS which look promising. More to come!










7 comments:

  1. I printed some sanitary fitting caps with ABS on a commercial FDM printer and attempted to burst them. Unfortunately, they were too leaky, it just made a weird spray head.

    Later I printed a camera bay that had 15psi pressure differential across it, and got it to seal well enough by dipping it in MEK/acetone. I never got around to trying to seal the caps and re-hydro them.

    If you're bursting stuff at hundreds of PSI you definitely want to be hydrotesting, not gas bursting. I prefer an actual hydrotester, but i've heard people have found success with a grease gun and a viscous liquid like honey.

    I was going to say "good luck with the Kickstarter", but I see it succeeded already. Congrats!

    ReplyDelete
  2. Thanks ben! Yep I usually fill the vessel and as much of the line as I can with water. Its not the best method because if you leave it too long air goes into the water but as long as you're quick and its in a protected area its safe... I've used a hydraulic pump before but its really messy and difficult to get fine control over pressure. I've been meaning to get a tester...but theres always something better to buy.

    ReplyDelete
  3. Hi, John,

    Firstly, huge congrats on the super-successful Kickstarter campaign. I've been following it since the start, and just kicked in a couple of bucks, just to get the updates. I'm very much looking forward to hearing how you progress. I think this project is just at the right time - I could use one right now.

    Secondly, my team and I have been working on a very cool new 3D printer for over a year now, and will be going to KS shortly (working on video this week). So I'm very interested in the results you're getting. Our machine has been printing since January, and we've been somewhat successful in printing large diameter (ca. 20mm)/large pitch (> 1mm) threads. We've not tried them in pressure vessels, but they work fine for small loads in mechanical applications.

    For smaller diameter/pitch threads, we've used 2 approaches:

    1. Same your suggestions above: print a smaller hole, and then drill & tap it. This works reasonably well down to M4. Smaller than that and the thread becomes 1-time-use. Once you take the fastener out, there's not enough meat left on the thread to go back. But for anything that's M5 or larger, this works pretty well in PLA.

    2. Print in a nut. This has been most successful: we embed a stop command at a predetermined layer in the gcode. then insert a nut into a cavilty, then resume the print. There's no reason why the nut couldn't be firmly cemented in the cavity to provide a pressure seal. The nut could be replaced by a custom-height threaded insert, to give more thread length. In either case, I would strongly suggest building in some extra safety margin to help with the stress concentrations in the sharp corners of the cavity.

    We're primarily using PLA and don't typically do much post-processing, if any. We're getting excellent layer adhesion, and have seen more failures across layers then delaminations.

    Looking forward to hearing more of your results.

    ReplyDelete
  4. John, just buy yourself a plumber's hydro test bucket. I got one on eBay for $50 a few years ago. In fact, it's sitting my garage and you're welcome to use it :)

    ReplyDelete
  5. Thanks for the support with the Kickstarter project Ben! It's interesting to hear your approaches to threads. I'd love to learn more about it, do you have a website/blog?

    QuantumG, thanks for the tip/offer. I will be safe....

    ReplyDelete
  6. This post is really valuable that designed for the new visitors. Pleasing work, keep on writing.http://www.3dprintinggeeks.com/

    ReplyDelete
  7. Interesting post. I Have Been wondering about this issue, so thanks for posting. Pretty cool post.It 's really very nice and Useful post.Thanks
    top-5-best-cooling-system-pressure-tester

    ReplyDelete