New design rules for WSF Polished & FUD/FD -- Feedback Request

I also feel that both items are VITALLY important.
a) Some way to mark a model (or invoice) as Test Print
b) Getting more feedback during trial periods (or Test Prints)

I will say this.. the production team HAS contacted me a few times over the past couple of weeks asking if I would beef up certain aspects of some models. I am MORE than delighted that they contacted me... this kind of interaction is crucial to the success.

For what it's worth, I took one model from 0.35mm up to 0.45mm, and Service said that the new "wires" were very much improved.

======
I would rather NOT see the limits increased to 0.6mm arbitrarily. I am fully aware that some of my models are hopeless, and I'll be happy to modify or retract those, but some models have been successful even at 0.3mm. I'd like for us to have a discussion before those (or any new models) are rejected.

The biggest source of trouble to me is getting rejections after orders have been placed. We need the Test Print option!

I think it has already started. My FUD figures test printed has already been rejected for further production.

Nancy very clearly said above.. this is a DRAFT.

I do know that the production team is making an attempt at closing the gap on multiple re-prints. You may receive a notice that individual models are problematic, but from all I have heard, they have NOT implement new "rules" for all orders.

Then I guess you should maybe add a paragraph named "handrail" or something along these lines. I had a look at my two designs:
-the distance between the wire and the main body is less than 20 times the wire diameter.
-the wires only have to support their own weight.
-the wires are always on a solid main body that can be held without having to apply pressure on the wires.

However, I would much prefer no to have a arbitrary limit set. I don't like the wsf approach where everything below 0.7 mm (even 0.68mm) is rejected and then you have to reorder and lose one week.

I would much prefer that you set the rules as some sort of general guidelines (but NOT enforced WSF-like), and you could use them to tell some people you can't reprint their models for free, especially if the broken features are "out of specs"

I also have to tell I like the fact you took the time to show us a draft of the new rules to get some review from the users.

Anybody from Shapeways Staff still aboard ? I'm sure most people here would prefer to review and comment draft V2 before it becomes the new rule.

Hey guys, I'm mulling over your comments. The problem is that we get a lot of orders and need to find a way to scale. Reviewing individual pieces (rather than having a set minimum) would not scale, and in the end it would not enable us to chip away at the cost and lower the sales price for you guys. So in some sense it's individualized attention to parts and higher prices vs. "arbitrary" minimum that by and large weeds out prints that will fail and lower prices. Not an easy trade off, I know.

The other thing we've been doing is looking into other ways to strengthen. We made some tests and it looks like if you use square wires rather than round wires, it makes the wire a lot stronger. I think maybe we can work out some sort of 0.3mm to 0.6mm ratio, it would probably be very low (1:10?). Also this would be for free standing wires only, not wires used as sprues. A 0.3mm sprue supporting a heavy object (think about the part that connects the body of a helicopter to the propellers) would probably break.

Last, I agree that we need to give everybody better feedback during the process about the difficulties we had in printing.

Just curious, the maker of the FUD printer did not design the printer to just print FUD? Do they have a stronger material other than the brittle FUD? If the material itself is stronger, that should make life much easier?


CGD

Oh, one more thing.. re: Stop4stuff's question about support and resolution and the two layer problem:

the support's resolution is not the issue. The printer setting defaults to only supporting downfacing surfaces for FUD, and allows us to toggle whether we want to support both downfacing and upfacing surfaces for FD (which we opt for both to achieve consistent quality).

We can't 100% guarantee that your piece will be printed in the same orientation, but we will be optimizing the pieces to minimize the problem of having white and frosty layers. The one thing that may override that is optimizing for height. So say if you have a model of a drinking straw, if we orient the straw so that it's standing vertically, that would reduce the effects of the two-layer problem. However, the print-time would increase (and with that the price). If we lay it on it's side, so the straw will have a higher chance of having some whiteness on one side, but print time would be reduced. In that case, we would need to optimize for height--but generally for models that aren't outliers like this (where one dimension is way bigger than the other), we will aim to reduce the amount of downfacing surface (less whiteness, less support).

We are looking into ways to eliminate this issue all together, so regardless of how you orient it, this won't be a problem, rendering this issue moot.

Maybe it could be worth to review VisiJet® EX200 and MX Plastic Materials, to look if the strength can be improved with another build material without having to make new rules...

Bill, can you describe a bit more what you mean here? Do you have a picture?

As for others on this thread, if ANY of you have ideas on how to make wires & sprue structures stronger, I would gladly incorporate them into the design guidelines. Especially, if you have successfully printed and know the structure is not easily warped or torn.

Thanks!
Nancy

Hi Nancy,

Re: support material resolution.

I get that the wax is only printed on the underside (parts nearest the center of the earth) because it supports the rest of the model against gravity, so why does the wax supported side take on a rough pattern compared to the upper side?

Re: warping

Warping is a heat related issue, whether it be at print time, or more likely during the wax melt out process... either which way it can be resolved using hot water.

Cheers,
Paul

Perhaps there is a third alternative, and that is that those pieces which are deemed to be below a minimum size are supplied with the suport material still in place. I have had white detail supplied like this and removing the support stuff was not that difficult.

I believe I can answer these. This is what I have heard (Nancy correct me if I'm wrong) When removing the build material from FUD the only way to remove the rough pattern is with bead shot. To use the bead shot on a high detail model would remove detail as well. Then it would be Frosted Was Ultra Detail . This is what I was told by a production facility.

I'm not so sure warping could be corrected with hot water. When a part warps it does 2 things. It elongates and thins out. Using hot water wouldn't redistribute the mass. I'm sure for slight warpage it would do. But as it is right now, Shapeways wants to distribute finished products. They wouldn't want a disclaimer about fixing warpage with hot water, and I know I wouldn't buy something that I had to unwarp.

This fits along with what I said about the warpage. Perhaps a "Prototype Purposes" check box is needed for items that are your own?

Acrylic breaks where there is a discontinuity. So to cut sheet to size it is scored and then bent. The sheet will break cleanly at the score. This is what I found with my models. They were basically wheels on axles and they broke cleanly at the join between the two i.e. at the arrow in the upper diagram. If the join between the two parts is blended as in the lower diagram there isn't a discontinuity and is less likely to break.



I've uploaded a piece to test these and other ideas and I'm waiting for delivery to see how well they work. You are welcome to print some for yourself.

There's no mention of a 'polished' FUD here on Shapeways, is bead shot FUD even a possibility? From what I understand an oven is the only post 3D print process that is used to melt the wax.

I've corrected FUD warp with no more than hot water and a flat surface.. the surface tension of the water 'glues' the FUD to the flat surface as the water cools redistributing the acrylic material to where it used to be. (Hey, it work for me )

+1 to a 'Prototype Purposes' checkbox for models (even if is costs a little more)

@fx: And the HR200, which not only has some higher specs, but is also described as having better surface finish.

@stop4stuff: FUD cleaning was supposed to be "hands off", allowing better details and walls than other materials that required active mechanical cleaning. Confusing info about material handling, and FUD... there is a joke somewhere. ^_^

That's my understanding too Stannum. From the spec, and unless anyone says any different, the ProJet HD 3000 from 3D Systems is the machine that FUD & FD is printed on... "Hands-free wax support removal" is right there on the page :confused

So are we saying that all this problem with poor definition on one side has been caused by using abrasives to remove the wax?

This then begs several questions - how long have Shapeways known about this and why haven't they told us before?

And are they supposed to do this ?

When we were sold FUD we were told one of it's advantages was the support material was a wax that was removed by melting. So either something has changed with the material or the operators have found that the wax is not removed by melting and have changed the process.

Is this approved by the machine manufactures or is it just a kludge to solve a manufacturing problem?

This is not so. I have had pieces made in the same material which displayed the same white finish, but interesting enough it only became apparent after the pieces had be cleaned and degreased in an ultrasonic bath.
If you make a model which is basically two tiles mounted one above the other with the top one having a pierced design in it, and if they are built horizontally, you will get the pierced design projected on to the lower tile as a white texture.

Youknowwho4eva is the one saying FUD is getting bead blasted. If you look it up, other SW pages say melting, as repeated in this thread. Blasting was mentioned about Alumide, and one reason for it requiring thicker walls and having less detail than WSF.

Looking some more, you can find a text saying the support causes different texture in FUD (but not FD) because it's printed at low res (the FD resolution). Think about it, for each support "drop", you get 4 material ones (in FD it would be 1:1 match). That means a blob of support with even smaller blobs of FUD on top, so the group of 4 will be pushed up by a small "peak" where they meet and flow down towards the corners of the bigger blow below. Or some other irregularities caused by the mismatch (the push and flow is speculation, just imagining small gel like blobs being piled, but rather plausible).

Why does it look perfect until cleaned then? Probably the oil used in last cleaning steps forms a nice uniform coat that hides this sub-millimetre pattern. Surprise! Oil, not only wax, is involved in FUD production, and also freezers and ultrasonic cleaners, not just ovens. Problems glueing or painting? Now you know the culprit and why degreasers can fix it.