Serious Customer Service Issue

Discussion in 'General Discussion' started by Pilgrim1908, Jun 28, 2012.

  1. Pilgrim1908
    Pilgrim1908 New Member
    I've just been told that Shapeways has decided not to refund sub quality prints - ie those where due to the failure of the Shapeways staff to align the model correctly in the print tray the model is marred by visible stepping and knitting. Please note this is the result of the actions of Shapeways staff and is not a problem with the materials or printing process. This is a big blow as I always had the confidence to order knowing that if there was a production issue Shapeways would correct it. Until this issue is addressed I'm sorry but I will not be reordering, and I would urge others to do likewise.
  2. ZoeBrain
    ZoeBrain New Member
    A print this bad can only be classed as "defective".


    Suggestions about "light sanding" don't work. WSF, like all nylon, resists sanding very effectively. It's both strong and flexible.

    Covering the entire model with fine-grain putty, then sanding, can work, but takes many person-hours for a result not as good as it would be if printed in the correct orientation.
  3. stop4stuff
    stop4stuff Well-Known Member
    Before everyone jumps ship, please could you share the model and material.

    By definition, 3D priting is an additive process whereby a model is built up layer-by-layer, some of that layering may be visable in a finished model, however there are steps that can be taken to reduce the visability of the layering that are independant to the print orientation. This may not be of much help to you with this issue at this time but it might help you and others reduce the risk of 'unhappy prints'

  4. ZoeBrain
    ZoeBrain New Member
    Material - WSF.
    Model - l


    Compare with what is stated at


    Such a "worst-case" is one thing. This print is quite another.
    Last edited: Jun 28, 2012
  5. ZoeBrain
    ZoeBrain New Member
    For comparison, here is a model printed acceptably., and of the usual quality for WSF we've consistently had in the past.


    And the finished product:


    See the problem?
  6. stop4stuff
    stop4stuff Well-Known Member
    Hi Zoe, are you pilgrim1908?

    The only image of yours I see is the one with the coin, the layering is as a result of the additive process, and I haven't a clue what size the coin is so cannot say if the layering is normal or not.

    [edit]doh that's the shapeways image :blush:

    Last edited: Jun 28, 2012
  7. pfeifferstylez
    pfeifferstylez New Member

    Usually, linking' on forums attachments doesn't work.

    And, since I don't have an account over there, even in the WingsOf War forums I can only see a thumbnail.

    Attach the pics here, or use an image hoster.

    Last edited: Jun 28, 2012
  8. ZoeBrain
    ZoeBrain New Member
    No - just a regular customer. I've re-ordered in the past, confident that in the case of the odd defective prints like this one, I'd get a replacement after a delay.

    Even when 22/25 models in one recent batch came through as bad as this one. It's no longer an unusual occurrence.

    Shapeways has had in the past perhaps $10,000 of business from these aircraft. That rapidly-growing business will evaporate if this is the new standard.

    If the models were printed with tail or nose up (aligned with z-axis) this problem shouldn't happen - and I suspect shouldn't be as bad as it is looking at the resolution of the z-axis printing anyway. The terracing effect is in 0.2mm layers at least, comparable to minimum detail resolution. It obliterates the finer details.
  9. ZoeBrain
    ZoeBrain New Member


  10. stop4stuff
    stop4stuff Well-Known Member
    Zoe, a basic concept of designing models for 3D printing is to make the faces of the model as small as possible, smaller than the minimum detail, half the size of the the minimum detail is good, 1/3rd the size is better. Yes, this becomes a juggling act between the quality you want and Shapeways triangle limit on upload, but the efforts are well worth the end result.

    You may ask. 'But, why must I make the faces so small? The model looks good on screen.' - even in the Shapeways image of your model, I can see faceting. The printer translates faceting to layering when the model is sliced up - the bigger the gap between vertices, the bigger the slices and the bigger the layering (or stepping).

    Here's an example, the top image shows the stepping, the middle image shows the faces, and the bottom image shows the scale against a 1mm grid - Bracelet ID = 65mm, total triangles = 615,456 - ideally this model should have had double the number of faces to reduce the stepping further, but like I already mentioned, its a juggling game and the print is more than acceptable.

  11. ZoeBrain
    ZoeBrain New Member
    If that were the problem, the results would be consistent, would they not?

    Same model, same material, but (assumed) different orientation of printing.


    Last edited: Jun 28, 2012
  12. stop4stuff
    stop4stuff Well-Known Member
    Yes, exactly - the results are consistent and in the case of a model such as yours, the faceting-converted-to-stepping is prevalent when the model is printed wing surface up because of the size of the faces that make up the wing surface although admittedly, the previous image of the stepped wings does look duff to say the least compared to the latest image. What did SW CS say about the print?

  13. Pilgrim1908
    Pilgrim1908 New Member
    Hi - no Zoe isn't me :)

    We both use Shapeways regularly to print model aircraft, and are part of a much larger community. I have over a year of ordering experience and own something in the region of 50 models, although the total number ordered by the rest of the community must be several orders of magnitude greater.

    This is not a complaint relating to the material used. We understand the limits of the materials and processes. The general consensus is that Shapeways WSF is adequate for our needs, and indeed this has proven to be the case, with problem prints being reasonably few and far between. When there has been this kind of problem Shapeways have identified the cause as being orientation in the printing tray

    However, in the past when there has been an occasional problem such as identified in Zoe’s pictures, Shapeways customer services have been excellent in offering to reprint. This has provided us with the reassurance to continue to order with confidence, and I have used precisely this argument to recommend them to other members. One of our members has now been informed that Shapeways are no longer prepared to reprint these “miss prints”. This has shaken our faith in Shapeways to say the least.

    The problem is one of quality control, not an intrinsic problem with the printing process or materials used. Shapeways appear to be saying they are happy to provide sub standard goods.

    Apart from the questionable legality under EU consumer protection legislation, this is in my mind a very poor example of customer service and business practice. Shapeways can rectify this by either getting it right when they print, or as they have done in the past, reprinting the “duff” ones. If they do, then I’m more than happy to keep placing orders.


  14. ZoeBrain
    ZoeBrain New Member
    I'll quote:

    If they replaced the WSF "worst case" example with the two photos above, to show how bad it can be and still be considered an "acceptable print", that would be one thing. I think it would scare away customers though, seeing the huge variation in quality. This is more than just "print lines" that may be filed or carved off.

    Customers have tolerated such bad prints in the past, knowing that at least they'll get replacements, when they submit photos of the items. This new policy is ill-advised.

    It also may contravene EU customer protection laws regarding "fit for purpose" unless only the worst examples are used in advertising.

    When they're good, they're very good. The rough surface can be dealt with, it's inherent in using WSF. Go to FUD if you need to. What can't be dealt with is the Russian Roulette of "do I get a good print or not" when there's no refunds.
  15. stop4stuff
    stop4stuff Well-Known Member
    Hi Ken & Zoe,

    I understand the situation now and have sent an email to, hopefully, the right person who'll be able to help you out.

    Anyhow, have either of you tried a high-poly model to see if that does make the difference? The method I use for near-smooth (i.e. reduced stepping) WSF models is what I've learned through the experience of modelling, buying, and refining the model and the method works. Perhaps, I should create an aircraft and put my money where my mouth is ;)

    Another little 'gotcha' - Shapeways is US based these days, so EU legislation isn't gonna work, haven't a clue how US advertising standards work though.

  16. ZoeBrain
    ZoeBrain New Member
    In other words, if Shapeways would just print the models with the fuselages aligned to the z-axis, there'd be few or no problems, right?. A simple fix.

    With their increased scale of operations, and subcontracting to 3rd parties of mixed competence, this appears impracticable though. The differences are clearly visible to the naked eye too, a simple QA check would stop such things from leaving the production facilities. This too appears too hard.

    I really feel it for the Customer Service people. So far, they've been permitted to do an excellent job. This new policy ties their hands.
  17. virtox
    virtox Active Member Moderator
    @stop4stuff Sorry, but I have to disagree on the faceting as the major stepping issue here. Stepping occurs on shallow angled surface that nearly align with the horizontal plane.
    And is most prevalent on the bottom of the prints, as far as I have seen. But this is usually regardless of the number of faces.

    In the case of curved surfaces, when it nears the horizontal plane, it has some influence, but mostly on the distribution of print-lines.

    For example:

    Combie View.jpg

    Whether the model is "perfectly" round or more faceted, the shallow angle parts will show lines, the shallower the angle the clearer the lines, because the horizontal distance will be larger than the vertical distance.

    If the model is faceted, the chance of a larger near horizontal area increases and the stepping might be more regular/pronounced.
    But in that case we are talking major face size, meaning much larger than 0.2 mm then. Looking at the airplanes that is not the case.
    The preview render has a tendency to exagerate this on small models.
    On the rounder version, the stepping will be non-linearly distributed, which might obfuscate it more.

    @stop4stuff, looking at your example the facets are smaller than the print lines, so I doubt making the faces smaller would improve anything.

    Back to the airplanes, yes I think it is the orientation that is the MAJOR factor here, but I am surprised the WSF material was ever suitable for these models.
    I would think the detail or ultra detail would be far better suited?

    Some more speculation:
    Another factor which may explain differences between orders may be which printer was used to print it, I believe there are quite a few different printer models used for WSF.
    And it would depend on the settings made by the operator.

    I can not tell how large the complete model including all sprues was, but if this was very large, the chances of it being printed flat will increase, and might even end up in the printer for large models (which according to some page I read :blush: ) has a coarser vertical resolution/more stepping.
  18. ZoeBrain
    ZoeBrain New Member
    As for your offer - good idea. Contact me privately and I'll send you some 3-view plans of a relatively simple aircraft to model in 1/144 scale, plus photos for details. Profit margins are small on these items though, they're very price elastic. No designer is going to get back in profits the time it spends to build the model, it's very much a labour of love. Too much competition from metal and resin models.
  19. stop4stuff
    stop4stuff Well-Known Member
    Yes Virtox, see my image of the red bracelet earlier - the image clearly shows a flat zone with stepping around it - however from experience sub 0.2mm faces reduces the stepping to the printer accuracy (0.1mm layer thickness with 0.05mm laser positioning accuracy for the EOS P100) and yes I have received models with major (0.3mm+) stepping - these were 'low' poly models with 1mm+ sized faces.

    I understand what you are saying about shallow inclines pronouncing the stepping, however Zoe's last image should have stepping on the fuselage of the plane as pronounced as on the wings in the image previous to that.

    The model Baritone Horn that I made & had printed in wsf has curves in all three axis, the faces work out to be about 72 triangles per square mm and the model has virtually no visable stepping.

    As for the printer settings used by different printers, surely that should be dictated by Shapeways, and if the settings are different to Shapeways specification (as in noticable stepping different to 'as advertised') then the model re-printed?

    Anyone got a choice of aircraft they'd like to see?
    (I have the full set of The Illustrated Encylopedia of Aircraft to work from) - PM on its way Zoe.
  20. Pilgrim1908
    Pilgrim1908 New Member
    They're EU registered and trading here so they're covered by EU regulations.

    Shapeways B.V.
    Commercial Register
    Eindhoven no. 17239507
    VAT no. NL/820321394B01.

    Look, we dont want a witch hunt, we want a return to a reasonable standard of service.