Stainless wall thickness

Discussion in 'General Discussion' started by stop4stuff, Jan 5, 2011.

  1. stop4stuff
    stop4stuff Well-Known Member
    I designed a Flugel Horn to be printed in metal and used as a pendant... I paid careful attention to the geometry specifications (ProMetal) and designed the parts to have a minimum wall thickness of 2mm (the piece is 50mm in its biggest dimension)

    Here is the reject image;


    In the past I have had models successfully printed with no reported problems that have wall thicknesses a lot less than 3mm, but now my Flugel Horn is rejected due to having wall thickness less than 3mm... why?

    Even the design rules for stainless say upto 50mm models can have wall thicnkesses of 1mm, see; les-Stainless-Steel-3D-printing.html

    Losing faith here in stainless big time :(

  2. aeron203
    aeron203 New Member
    That part doesn't look too risky in terms of thickness, it's just that the inherent shape is not ideal. Remember there is some subjectivity involved in review and considerations beyond just the thickness. Your model has two larger masses connected by a 2mm neck. Try to picture the piece laying on a flat surface being brought up to right under it's melting point. It becomes fairly soft and rests unevenly. Any pressure exerted on the model is amplified by leverage because of the length of the neck. Just like you could sit on a thin branch of a tree if you're near the trunk, but slide out to the end and that branch will snap.

    I'd say submit it again and you will probably have it reviewed by a different person, but also you may have to take some artistic license to make it printable. Either distort the shape until the parts touch or add a connection just 1.5mm thick between the flare of the horn and the tube just before the curve to join the masses more directly. Filleting the edges of the connection would probably double it's strength while not changing the mass or appearance much at all. This would also help the bronze flow better during the infiltration process and reduce the likelihood of it breaking if they are brave enough to tumble-polish it with dozens of other random metal shapes.

    Perhaps you could add a loop for the pendant that also acts as support?

    This is great model, don't give up on it!
  3. stop4stuff
    stop4stuff Well-Known Member
    The model is for a friend's daughter, and is at about the maximum size required (see, It Arrived: Musical instrument models and Flugel Horn). I understand about the geometry and have already given alot of artistic license without the model looking like a trumpet... the WSF version (smaller wall thickness) has also alot of artistic license to make it work) I could just scale up by 50% to get the 3mm wall thickness, but then the model becomes 75mm long and, frankly, too big for a 'girly pendant'

    A support between the flare and tube would proportionately be the wrong size and look very wrong at 1.5mm (to scale the support would be about 0.2mm thick, and the model is all out of whack for scale anyway) the spec is to have the horn hanging with mouthpiece uppermost. The flare of the horn is 2mm thick all the way to just infront of the valves where the main tube is about 4.75mm dia. The forward-most loop of the tube starts at 2mm and expands to about 2.75mm where the frontmost valve tube is attached... the length of this 'free' (solid) pipework is around 50mm, much less than the 2mm rails on my Twin Rail mobius pendant

    There are several workarounds that I can try, however time is not on my side.

    Included in the workarounds are;

    - Redesigning for a thicker WSF model and try to get the material to soak up a low pelting point alloy (field's metal, 62 C melting point)

    - Have a 3 part mould printed in sandstone and cast using a Bismuth-Tin alloy (melting point of 138 C) that is used for GW 40K type figures.

    Also I have gone a different route and am waiting on a reply.

    Onwards and Upwards!
  4. GlenG
    GlenG New Member
    Aaron's observations are mostly spot on. But the major problem and probably the reason why the design was rejected has to do with the "green strength" of this part. Green parts have about as much strength as unfired potters clay. Asymmetrical parts are always challenging. Your horn design is way cool but, with that wide bell end supported only by that skinny loop of tube, it would be extremely difficult to de-powder and get it into the furnace without breaking. An anticipated failure rate of 2:1 or more likely 3:1 is the likely reason the job was declined.

  5. stop4stuff
    stop4stuff Well-Known Member
    Glen, the skinny loop is approx 4.75mm solid diameter where flare opens out... but I understand the situation... Shapeways won't get it printed, and I already have an explaination as to why... 3mm minimum on ALL wall thicknesses as agreed with their 'furnace boys'... hey ho
  6. aeron203
    aeron203 New Member
    It looks like this will not be a simple project. You have some very interesting solutions in mind. I know some people cast with Sandstone but I would be a little concerned with such a small part since Sandstone is notoriously inaccurate. expect 2% error at least, and the part "bleeds" outward from the designed surface by about 0.15mm. If the mold parts are joined by a 3mm section, printed together at the same orientation and snapped apart later that would help, I guess. I'd be really curious to see how it works, but since it is a gift and time is limited it seems a little risky.

    The only other thing I can think of is adding a temporary structure to the metal part, like a 3mm thick bar attached with very short thin connections so you can cut them off and polish it flat later. That casting will probably take as much or more effort to clean up, so this seems a little more promising if it can be printed. I suspect it might get through if reinforced properly.

    Either way this is turning out pretty interesting so do let us know what happens.
  7. GlenG
    GlenG New Member
    After 2 years (?) printing many thousands of SS parts for SW it was realized that in order to continue the service at a reasonable cost these new rules needed to be strictly enforced. Your flugelhorn design, and many many other small intricate parts can in fact be printed below the 3mm rule but the cost of such parts would have to dramatically increase. It's all about finding the balance point for good healthy commerce for all or most parties concerned.
    Also, one heads up! Most of the so called "low melt alloys" contain elements that are not suitable for skin contact. Bismuth is a heavy metal not highly toxic but is likely to cause a black stain on human skin. Some people may even develop allergic (dermatitis) reactions.

  8. aeron203
    aeron203 New Member
    As always, your input is very valuable to us. Even with the great customer service we'd be doing a lot more guessing without having an experienced expert available. In a business that prints basically anything there has to be some margin in order for things to work, and at some point a number has to be chosen, and that number appears to be 3mm.

    Still, it's important for us to discuss the other considerations because the reality is that the success or failure of the print and of the service truly is shape dependent. For example an 8 inch tall tree of 3mm elements would be nearly impossible to de-powder. 250 copies of a 3mm sphere would cost only $35 and that just isn't sensible for your business.

    I understand that we can't have a hundred Glens reviewing our models, and without a firm number the results are going to seems totally random to the customers, but I think putting more effort into educating people in addition to having a reasonable minimum will return maximum value and satisfaction to the customer and profitability for the business.

    Apologies to Paul for sort of hijacking the thread with this conversation but it's obviously not OT. Maybe it would be best to start a separate discussion compiling tips to help people think through the process before investing so much effort in their model, since it seems many people are still unclear about why their models are rejected after reading the current design rules.

    The horn would be amazing in silver!
  9. stop4stuff
    stop4stuff Well-Known Member
    No worries Aarron :)

    What gets my goat is that Shapeways have 'moved the goalposts' without informing customers first. And that has a direct impact about how Shapeways customers feel about the business.

    @ Glenn, quite frankly, I am at the stage with this design where I am prepared to pay Shapeways just to find out if the flugel horn ca be made successfully... I didn't say anything yet about my Baritone model did I, that got rejected way before the Flugel... the beefed out one got rejected too.

    But hey, like I said there are work-arounds... fwiw I'll be coating any LMP alloy cast models with a polyurethane varnish.

    I don't let knock downs get me down, but I will have my say.

    Like I said before...

    Onwards and upwards.
  10. GlenG
    GlenG New Member
    all of then horn designs would look great in Sterling! My best advice would be to print a master pattern in the highest res plastic that SW provides, do a little hand work to clean it up and then find a jewelry casting service to make molds and cast them in Sterling or bronze. This is not the most direct method and will likely cost 1-$200 for the first few castings. There is also a service provider in NYC that will print directly to wax on SolidScape machines. Cost will vary with the desired resolution. Again, not cheap but fantastic results.
    You have a great concept here Paul. It could be quite lucrative if you find the right market. It's all a matter of how much you are willing to risk.

    Good luck with it,
  11. stop4stuff
    stop4stuff Well-Known Member

    I have already paid my money for an advertised service... I now have a coupon for a little more money than I paid to pay for another model/more models that may or may not fail... sorry, I am trying to build up my own business and am not very happy with goalposts being moved mid flow. This one issue has now provided me with a reason to rethink and possibly push my plans in a completely different direction that may or may not include Shapeways.

    Silver would be nice, Shapeways do not offer that. From a master, I'd probably looking at 3 months turnaround for a line in silver at about £30 per piece, but that is not where I want to go.

    I have yet to talk to my friend about options, he likes the stainless items that I have already and knows his daughter would wear the horn in steel whilst she would not wear (and does not wear) typical jellewery pieces... It is the look and style of the printed stainless that makes the difference.

    Anyways, tomorrow is a different day and new emails about options should be incoming. :)

    Ranted a bit, did'nt I... time to party now...

    Party On!

  12. stop4stuff
    stop4stuff Well-Known Member
    Partying over and back to the serious stuff.

    Good news
    DMLS can be done with wall thickness down to as small as 1mm dependin upon the geometry.
    Bad news
    the Flugel Horn would com in at around the £300 mark as a one-off or maybe down to £90 per unit for 10 units.

    Please Shapeways get this hiccup sorted out with the 'furnace boys'
  13. GlenG
    GlenG New Member
    Just in case you have not looked into this here is a list of SolidScape service bureaus:
    Some of these providers should offer casting services as well.

    You might want to check out some of the websites dedicated to metal arts and craftsmen such as : <> This is US based but should have links to plenty of Euro orgs. Many links to services found on these sites as well.

    Last word, go take a jewelry making class at a local art center and do it yourself for next to nothing. Of course this might dip into your "party time" :rolleyes:

  14. stop4stuff
    stop4stuff Well-Known Member
    Thanks for those links Glen, I'll look into them. Other options for the future, sure, but no good for now. The thing is the Flugel Horn has a deadline and specific to the build is the stainless material from shapeways, in terms of finish and tactile qualities it is pretty unique which is what I have been designing for using and honing my 3d modelling skills. Traditional jewellery making is fine, however I do not have the space or resources for that, which is why the shapeways option seemd ideal until the goal posts moved.

    I don't know how life is for you, but I do have other issues to deal with as well as a potential 6 month time investment wasted. Last night's partying was just to get away from the stuff built up during the day, and I feel like tonight might end up the same.

    Always looking for the workarounds,

    p.s. In one discussion today relating to DSML, the guy was amazed that shapeways can offer stainless and accept models at such a low price.

    I was wondering maybe if shapeway could develope some software that automatically simulates the thermal dynamics of the print within the upload process so that customers know beforehand whether a model would be possible in stainless... oh & btw, my twin rail mobius pendant has 6 lengths of approx 100mm of 2mm solid tube unsupported with, as yet, no reported failures... idk what the probelm is with the horn :(
    Last edited: Jan 6, 2011
  15. Youknowwho4eva
    Youknowwho4eva Shapeways Employee Community Team

    I understand your frustrations, but in order to keep the price of stainless down, the change was needed. It's not for a lack of effort, or skill. It's just not price effective to allow parts under 3mm. I have many items myself that are now unprintable in SS that I'll either have to remodel or let go. I've read bits of the thread, but didn't see anyone suggest making your model the 3mm thick, and the areas that you want to be thinner and/or more detailed to tool by hand?

    I'm sure as the process evolves we'll see the 3mm shrink. Just right now, to assure that smaller items didn't fail it would cost everyone more money as each item would have to be evaluated on upload weather or not they reach the 3mm minimum to be able to price them differently.
  16. stop4stuff
    stop4stuff Well-Known Member
    I'd pay just to find out if the Flugel could be done as is without modification success or fail... it's the 'as printed' look that makes Shapeways (ProMetal) stainless attractive for this model (the model meets the ProMetal specification that I have)... post production is not an option, neither is having the model made in another material, it's an 'all or nothing' situation as far as this is concerned.

  17. gibell
    gibell Well-Known Member
    The design rules still indicate 1mm is OK for small parts?? Just to be safe, I just made all my metal puzzles "Show only" because they have 1mm walls, even though they have printed several times successfully ... It is not worth the hassle of somebody ordering them and having the orders rejected.
    Last edited: Jan 9, 2011
  18. Magic
    Magic Well-Known Member
    In addition to the wall thickness rule that seems to be generalized to 3 mm minimum, it seems that no more moving parts are allowed in Stainless Steel.
    I must confess that most of my Stainless Steel designs infringe one or two of these "new" rules.
    As a consequence, I think it is time for me the give Stainless Steel a break, until things go back to (what used to be) normal.
  19. fx2
    fx2 New Member
    If we take a look in the jewellery category, a lot of item are "breaking" the 3mm rule, including Shapeways own design. Does this means a lot of the existing jewellery designs can't be printed in stainless anymore ?
    Last edited: Jan 7, 2011
  20. virtox
    virtox Active Member Moderator
    If things really have changed, that more or less kills all of the jewelry design in metal..
    But I have received no official word that design rules for stainless steel have been changed?
    So I'm not about to change my design tactics until I do ;)

    Looking at the Flugel horn, it looks a lot like one of the typical "fail" examples in the initial design rules, the wine glass: les-Stainless-Steel-3D-printing.html :
    So far I have had no problems with items smaller than 5cm and using 1.2 mm struts, as long as there are enough of them to create an overall structurally sound design.
    (think sphere structures, circles, common supports etc)

    The mobius rail designs seem to be at the limit, long thin parts but still connected on more than one end, which seem to be the least trick, don't create masses connected by small items, but make sure "ends meet" ;)



    Last edited: Jan 7, 2011