I am not going to say anything else about S3D after this post but feel it is very important that folks go into it (if that’s what they choose) knowing that support for infill is completely unacceptable. Here is a post from 2 days ago and resulting followup: http://www.forum.simplify3d.com/viewtopic.php?f=8&t=2255
This was right at the top of my list of things that were show-stoppers for this app. If you are printing Yoda and don’t care about infill, go for it. If you are printing parts that need more rigidity or the infill is visible through a translucent PLA this app is unacceptable. Infill is such a fundamental aspect of slicing that it is beyond comprehension that this has been a limitation for over 18 months with no signs of change. The stock user reply to “increase the extrusion width and increase the hotend temperature” as the solution are equally unacceptable and create more problems to deal with and do not actually result in a significantly stronger part.
Enough said, there are many other issues with this slicer and I encourage you to determine if it will work for you before spending your hard-earned cash. I’m happy to take further discussion off-line but this is my last public comment in these forums.
Yeah… I can see how having infill revealed with translucent PLA could be an issue. But I would guess that it could only affect 20% of prints. Its a pretty specialized need if you ask me… I print with ABS, so no translucent issues there. In my first experience, S3D with little tuning sliced a clean and functional part which neither Cura or Slic3r could produce.
As for rigidity, I rely on the perimeter shells of the object. If weight is a concern, I’ll print with extra perimeters and just enough infill to support the shells as necessary… the strategy has worked well in my limited experience.
Interior supports baffle me. When I think of “supports”, I think of something expendable and removed during cleanup. If the design is planned out, pillars could be placed in a hollow object manually to support the shape.
One unique feature of S3D which I haven’t seen in other slicers is the ability to vary print characteristics (ie. layer height, extrusion widths, infill percentage) in a print. This is achieved through the use of combining multiple print processes. The tutorialan interesting use case. I suppose if you really needed to do this with another slicer, you could cut and past g-code to create a custom file. But through layering processes it becomes a pretty easy thing to do.
Interesting point, albeit one that I’ve never had an issue with. I suppose I would like to see the ability to completely bond each layer to the previous rather than alternate +45/-45, but quite frankly if I need structural rigidity in a part I’m going to design that rigidity into the part, not depend on ANY slicer.
What I would really like to see is the ability to specify custom infill profiles.
This is specified as a single project so its one continuous print. Again, this could be possible with Cura or Slic3r if you generate two models and cut and past at the height required… but it could be really tedious.
Anyway… I don’t mean to dwell. I’m just a happy S3D user. (stepping off the soapbox…)
@kcchen_00
What I actually meant was that I would like to be able to generate a hexagon infill without having to model it into the part, not that that’s a big deal when using a full fledged CAD program like SolidWorks. Just draw a hexagon with the desired wall thickness, extrude it, pattern it, and trim to the appropriate faces. But this gets me back to just building in structural rigidity when I really need it.
This is exactly how I feel and wouldn’t have bought it had I known it was so limited. I really feel like that is a large part of why there was no demo. It seemed like such a basic feature I couldn’t believe it was so poorly supported.
Also, it appears S3d does not allow posts that question their position on infill options on their forum. I’ve posted, or rather submitted for post, questions about infill options and the posts have not shown up. All posts have to be approved by a moderator and it seems if they don’t like the question, or believe customers will not like the answers, the post is not allowed.
I’ve actually had my posts censored even though they were mostly questions and not really negative. In fact, 2 weeks ago I even contacted them directly (I have their direct email as I talked to them at length last year) and asked to re-purchase if they would honor the refund if I still had issues. I even stated the exact things I was going to test, asked them to provide input on what the best platform to test would be (OS X, Windows, or Linux) and stated that I was considering it for other parts and the infill weaknesses - which I realize are still there - would not be evaluated or factored in. Not only did they decline, they recommended that I “go elsewhere”.
I am a software exec and have founded 8 companies in the Boston high tech sector. If I had an ex customer willing to come back and re-evaluate my product I would bend over backwards to make them wildly successful. Think of the great story that would tell, and as you can tell, I’m pretty vocal about things I like (and things I don’t like when pressed into it). But then again, I would never have let a customer walk the way they let me last year. I opened the door many ways and times to give them a chance but it fell on deaf ears. I don’t hold grudges and was willing to give it a fair shake recognizing the known issues. I won’t ask again but if they come to me, I would still consider it.
I think they monitor the first few posts if you just registered a new account. I remember my first few posts seemed as though they were held.
Regardless, there’s a third party analysis of infill patterns. The quantitative analysis seems to show that rectilinear and hexagon patterns are fairly comparable in strength. Further supporting S3Ds direction and stance.
Biggest thing I’ve found with S3D’s infill is to turn off the ‘random infill placement.’ Leaving that active will make terrifically weak infill at anything lower than 90% infill, as you’re basically just bridging a bunch of thin wires without much beneath them.
I will admit, it’d be nice to be able to do an “X” pattern every layer. To be honest, I mostly print things solid (unless it’s exceedingly large, but often if I’m printing large I need a lot of structural integrity so it’s solid anyway).
I acknowledge that this work on infill is a good step in the right direction. However, it is not a compete story and has some significant omissions. Statements about what a slicer should or should not be doing for infill based solely on this work are misguided. One example - this work is for PLA only, each material has different properties so you can not make a general conclusion based on the results (and as you’ll read below, incomplete results) from a single material.
I’m not going to pick the research apart (this is based on my reading of the original work at 3DMatter) because I hope/anticipate this is just the first phase of a more complete analysis, but one glaring ommision is - what orifice diameter are they using and what is the relationship between orifice and layer height? The ratio of orifice diameter to layer height has a very large impact on strength and rigidity. The work does go into some exploration of layer height, and it does present the machine that made the test pieces on, but they do not explicitly say what nozzle size they used or the effects of nozzle diameter or the ratio of diameter/layer height.
The other larger issue I have is the lack of meaningful treatment on the effect of infill pattern on rigidity - and in particular, the effect of infill on anisotropic rigidity not only along Z but, particularly along X and Y. This is far more important in selecting parameters (infill and others) to tailor the mechanical properties of printed parts than is tensile strength. There are other areas that need more work and hopefully these will be addressed in the future.
I’ll add in and say that while I love Simplify 3D and its speed / beauty of the parts, the lack of a better infill system is pretty awful. My general rule of thumb is that if I don’t need strength, Simplify3D is the way to go. If I need any sort of part strength (particularly on the z axis), Cura or Slicer (depends on the part) are the way to go.
To add, I would be very curious to see that same effort (the one at the link) done with Simplify3D. If I have a bit of time this summer, I may task a student to run some stress tests on the parts that come out.
I have zero confidence that the results gleaned in the earlier link with regards to infill types generalize to what Simplify3D does for infill.
It would also be interesting to do a market study of slicer software. Id like to see which is most popular, and market shares for these in question (along with makerware, and etc.).
If you do perform some stress tests, it would be really cool if the strengthening effects of vapor polishing and/or acetone brush surfacing could be included.
I don’t know what there is to doubt… rectilinear is rectilinear in any slicing software. If you really want strong infill via S3D, specify 200%+ extrusion… @200% you get almost a solid wall between infill layers. I don’t think the 3D Matters report was to support any specific slicing software.
I’d also be interested on any tests with post processing (ie acetone vapor finishing, epoxy coating).
In my first experience, S3D with little tuning sliced a clean and functional part which neither Cura or Slic3r could produce.