3D Printer General Chat

Discussion in 'Hobby Engineering' started by Spanos, May 26, 2021.

  1. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    So I'm busily designing up a set of 3d printed fridge slides to suit some unusual requirements in my 4 x 4. What I'm struggling to find is a suitable filament! The filament has to be UV resistant so I kind of narrowed the list down to ASA, PETG and Nylon. I have a preference for Nylon because of its low co-efficient of friction means I won't need any bearings in the slide (hooray!!). This isn't essential as I can print a rebate into the slides so I can insert nylon in as the contact surfaces.

    Now, because it is in a vehicle, it needs to be resistant to heat whilst supporting the load. The fibre (glass or carbon) versions of the filaments seem to have a much higher HDT but I'm struggling to find any useful design information on creep resistance.

    I'm leaning towards ASA at the moment as I can use interference joints between slide segments (max length I can print is 225 mm and the slide lengths are around 675 mm) and acetone to glue the segments together. However, I cannot for the life of me find any fibre reinforced ASA available in Aus. I'd like to keep the cost down to around $70/kg roll too which I think is gonna be a hard sell.

    Anyone got any ideas?

    Anyway, a quick and dirty analysis of the slides. I'll turn it into a proper analysis later on but you can see the loads are quite small (about 15% of the yield strength ASA). The slides are curved and not straight as shown either (part of the requirements to make the fridge fit.
     
    Last edited: Sep 18, 2021
  2. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Pics not showing?

    I have queried a mfg (a decently prolific one too) in the past about creep properties with their engineering targeted materials and they simply replied 'sorry we don't have this data'. I suspect that's not the whole story but it's typical of material manufacturers to stick to saying only what they think is needed to sell their material, and then farm out the rest to end-user and sponsored research. Plastics is particularly a dark-arts central for this.

    Anyway, yeah I think you're going to struggle to reach your price point on filled nylon, asa or even petg locally.

    A possibility is that you can use sheet-metal to reinforce the slide structure - either inserts or crimping sheet around a print.
     
  3. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    Oh yeah, I was mid edit on the post and then forgot to come back to it. haha.

    I'd probably go for a filled nylon over the petg because then I won't need to buy some teflon to use as the bearing surfaces.

    I do like the idea of using a bit of sheet and incorporating it into the print. If you look at the below, you'll see the three telescoping parts of the slide. The majority of the deflection occurs in the middle slide and not the end ones. I was planning on using filled filament on the middle one alone but i do like the idea of a bit of sheet. I'd have to work out how to do that though. Maybe just print a shell in nylon and then screw the sheet metal to the top and bottom flanges of the slide. That way I also don't need to worry about creep as the sheet metal will hold it's shape and resist any change in the plastic, since this is a low stress use anyway. The fully loaded fridge, fully extended and with a load factor of 1.5 is only generating 7.2 MPa of bending stress. Well below the yield stress of even my crappy prints.

    Oh, and I won't need to worry about the joins as much either! they can just be an interference fit and happy days.
    upload_2021-9-18_15-15-51.png

    The slides are curved because the fridge is too tall to reach into the bottom. This way, I can slide the fridge out and it will have a 30 degree angle from horizontal in it. It's an 80 L fridge which I picked up second hand.
     
  4. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    So the filled materials I've used usually finish abrasive. They have enough fill exposed after the process that for example the normally relatively slideable nylon isn't anymore, so I don't think your one-shot plan will work in any case.
     
  5. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    That's a good point, I hadn't considered that. So I'm thinking aluminium screwed to the top and bottom of the flange and nylon for the web and inner part of the flange. I think that's probably the best of both worlds then.
     
  6. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Doing It like this would be completely ambitious but I couldn't help myself from drawing after thinking about it.

    Basics:
    Parts:
    Plastic print
    TIG wire, 1.6mm (or MIG wire), 2 pieces

    Operations:
    1. Print one section of the base until the layer where the wire can be inserted then pause
    2. Insert wire reinforcements
    3. Finish the print
    4. Remove print
    5. Print another section until the wire can be inserted then pause
    6. This time insert the base you just printed with the wire, dovetailing it in place
    7. Continue print to finish
    8. Repeat.

    You would need to provide end caps that secure the wires (obviously the overhang has to sit on something so the end piece on that side would need to be different), and a little pretension would be good. Plenty of possibilities.


    Linear v1.png
     
    Last edited: Sep 18, 2021
  7. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    I like. Not a bad way to do the whole thing. I like how it will pull the joins together a bit. I might actually steal this idea and run some elastic through it rather than wire though. I calculated the force to push the fridge back up the slide when loaded and it was close to 50 kg's. Thats easy peasy for me but not for my partner. haha. So I was planning to run some elastic up it and i think the wire replaced with elastic will work awesome.

    I'll probably use aluminium screwed top and bottom flanges to give it a bit of sideways stiffness too. The slides are at 75 mm width at the moment to make sure it doesn't leave (i've calculated it) when its in the closed position and I'm 4 wd'ing.
     
  8. T1tan

    T1tan Member

    Joined:
    May 8, 2002
    Messages:
    6,217
    Location:
    2607 / caaaaanbwa
    upload_2021-9-20_15-3-55.png upload_2021-9-20_15-4-57.png upload_2021-9-20_15-5-40.png

    This is my first ever tinkercad project.

    It's a 5.25" drive bay blank, I've added a slot for a PCI card, to use with a riser cable. This should allow me to mount a front facing Four Port NIC if all goes to plan. The brackets are to apply pressure and should work with case screws.

    Before I submit this to the request thread, are there any obvious issues? I am completely naive to 3d printing. I'm assuming at 0.5mm for the brackets (red) there is a degree of flex.


    upload_2021-9-20_15-7-28.png
     
    Last edited: Sep 20, 2021
  9. theSeekerr

    theSeekerr Member

    Joined:
    Jan 19, 2010
    Messages:
    3,712
    Location:
    Broadview SA
    It's a little odd that every feature that ought be round is a hexagon, any particular reason?

    Those tall thin struts will be pretty fragile, I'd beef them up or at least fillet the base.

    In general you should probably try to work out what the maximum amount of structure you can include with a reasonable amount of clearance is, rather than trying to design a minimal structure. Every feature looks a bit thin.

    Still, I don't see anything there that's unprintable.
     
    T1tan likes this.
  10. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    Looks great!

    theSeekerr has some good points. I'd also point out that you will want to think about the nozzle diameter when determining the thickness of things as well. i.e. a lot of people use 0.4 mm nozzles so you would want to design the thickness of things in increments of 0.4 mm width, or whatever nozzle diameter you think is appropriate.

    How does it fit together? snap fits and captive nuts? You want to make sure there is enough material around the nuts to provide the require strength.

    Chamfers at intersections are really good to improve strength.

    How thick are the faces? I usually print with 8 x 0.1 mm thick layers for the external faces of things. If its printed in white then I can usually still see through the plastic. i.e. I can see the infill through the plastic in a normally lit room (so, not with sun shining through the object). Would that be a problem?

    Don't be afraid to brace items like those long struts. A k-brace would work well especially if you dont mind printing with some supports. i.e. the k is printed at 45 degrees so shouldnt need supports and the straight bar between them can have supports printed from the diagonals that can be easily removed.
     
  11. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Was kind of curious on various ideas in my previous post so developed it a little further

    Modelled up the parts more correctly then printed them... encountered a number of difficulties as expected but got it working.

    [​IMG]

    Being PLAI did try annealing it, but it turned out as I had expected. This print is smaller than the example and the wire is MIG wire. MIG wire comes off rolls which you have to straighten, in this case by hand. Straight enough to embed but the residual forces did warp the print slightly when it came up to soft temperature.

    Then came the last part which was securing the wire ends. After thinking about various options I settled on a rudimentary fold over. This is where then everything decided to fail. I had a thumbsuck figure of about 3mm of pull on the wire to be somewhat reasonable. I marked it and tried to get that on the vise. I got pretty close and began to wrap the end over and then... BOOM. The end piece I had chosen to wrap up onto the flange and that was a fail. With the forces involved in making the turn it simply ripped the flange off.

    This leads into another matter that I was thinking about before just doing the simple option. The wire essentially will cut through the plastic well before breaking the two aren't really that compatible. Methods to spread the load would be essential to help gaining material efficiency out of the method.

    The favourable observation though is the print-in lap join. I had to print a new vent to get this to work on my printer (the approach angle plus vent height limits the parameters- what you see is on the limit of my capability). I did some bend break tests on the partly destroyed item - and as seen below the lap join is fused well enough compared to the rest of the print, although the print was definitely light on - you can see that it's hollow with only a thin wall and low infill.


    [​IMG]

    Though it was quite finicky placing the join. You need good quality on the dovetail on both sides.
     
  12. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    Could you explain this again? I don't follow what you are trying to say.

    Did the flange you were tensioning fail, or the opposite side?
    Yes, different stiffness materials. Your right, the design will need to accommodate that specifically.
     
  13. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Flange is probably not particularly explanatory. Basically this was the geometry I was trying to bend the wire around, and due to the specifics of how I had grabbed it in the vise as well as just the amount of force I had on the vise grips I was holding the wire with, the layers across the weakest point just gave up. Not helped that the wall and fill was set right down at this point in time to get the speed up on the print.

    It's pretty dubious how well it would perform in other matters, more just a quick and dirty choice to finish a test piece.
     

    Attached Files:

    • term.png
      term.png
      File size:
      232.3 KB
      Views:
      10
  14. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    It looks like it has failed across the dovetail in the middle of the print, not the ends though?

    I think what you are saying is that you printed the part without tension in the wire first. Then tensioned it after the print was together. It's a decent experiment to do. If you've post tensioned it, then you've loaded the flange in tension (through friction) which has eventually failed. It's actually the reverse of what we want to do with the tensioning. We want to tension the flange with a stiff material so that the plastic is in compression. So you either have to tension the wire first, then print around it like you have, or, print the wire in a duct and then bend it over so that the transfer of load is only at the ends. This will put the flange in compression, so that when you load it up, the load is being equalised out (as you are stretching out the flange which is in compression).

    Edit: or maybe that's what you've done? I.e you've torn the flange off the Web due to the geometry at the ends? But, the photo of the failure is a tension fracture at the join? Or maybe the photo of the failure was when you did some bending tests?

    Its a good test, because now you have a measure of the immediate losses that occur after prestressing. This is an important piece of information to have when designing. Then you would also need to accommodate for long term relaxation due to creep (both in the wire and the plastic).
     
    Last edited: Sep 20, 2021
  15. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Yeah I haven't posted the other failure. The picture is just testing the lap join for strength, that was after I busted the end segment trying to lock the wire with some tension on it so the wire is just there, it wouldn't have been doing much.
     
  16. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    Aaah, yep. Now I think we are on the same page.

    So I take it that the wire wasn't being loaded in the tests that produced the photo of the failure then? As I assume the ends weren't locked into place?

    Looks like it failed in the top piece alone and not the bottom piece?
     
  17. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    It's kind of like train tracks, but in the breakage basically it's broken though all printed layers of plastic, rather than across any part of the lap join - including the small flat area above the wire. Below the wire only the dovetail is in place and that hopped from too much deformation. Although the dovetail is notably a bit shallow, I was also using a hammer at that point so nothing had much chance.
     
  18. OP
    OP
    Spanos

    Spanos Member

    Joined:
    Jul 29, 2019
    Messages:
    43
    Yeah, I'm really interested in the behaviour of the angled up section which is quite obvious with the zipper like finish. It looks like the printer has pushed into the completed plastic section and ironed the plastic a bit. It seems to have built a full strength joint! It's almost like you've done a full penetration weld at the angled part.

    Could you explain how you did that join? I assumed you've printed one part. Removed it. Printed another part. Then put both parts back on then printed the top part? I'm very confused about the sequence of steps that you've taken to do that?
     
  19. mtma

    mtma Member

    Joined:
    Aug 12, 2009
    Messages:
    5,925
    Well being not 100% fill it's probably more than strong enough even if it's not fused well as the main body is thin and part fill, but yeah the original thought would have been that you can't approach an adjacent object on a typical print head like mine if it was square. The fact that it can be used as a fusion joint is a secondary matter as far as design intent but certainly was a curiosity. It would be better to have something that intrinsically pulls up square assuming a tension could be pulled on the wire in the assembly.

    The primary reason why it's particularly squashed is that it's almost certain that the inserted 'under' component rides up slightly. Especially with the wires and stuff and if the 'over' component has any degree of elephant's foot it can cause some issues.

    P.S: Apparently in woodworking this is called a scarf joint. The more you know
     
    Last edited: Sep 21, 2021
  20. T1tan

    T1tan Member

    Joined:
    May 8, 2002
    Messages:
    6,217
    Location:
    2607 / caaaaanbwa
    Thanks for the tips :) but doing/confirming 0.4mm might be a tad tough at the moment. Functionally, does it matter enormously or is it primarily for aesthetics?

    Cheers

    Nik

    upload_2021-9-21_14-10-31.png upload_2021-9-21_14-10-52.png
     

Share This Page

Advertisement: