I have ordered a 3D printer. The B9 Creator (http://b9creator.com/), to be specific. There's a four week lead time currently, so with any luck, I'll be printing by the end of March.
I'll be interested in your results!
What kind of resolution do you expect? By that I mean in comparison with service bureaus you and our other 3-D printing guys have used: For example, would HO scale parts need much cleanup? That sort of thing.
And PLEASE keep us up to date as you experiment with the printer!
Russ
It goes as low as 30 microns X-Y, and 5 microns Z. In theory. The X-Y is pretty accurate, as it's based on the projector resolution/pixel size. (It will also do 50 and 75 micron X-Y, by repositioning the projector. So you trade resolution for larger print area, as the image gets larger when the projector is further away, projecting larger pixels.) The machine is capable of 5 micron increments in the Z axis, but people are getting the best results between 30 and 40 microns Z. Things like light transmission through the resin and overburn can affect the actual resolution, and require a bit of fiddling with settings.
We had the 1.1 version of the B9 at work, and the actual difference between the B9 prints and our pro machine were probably on average about 5-10 percent. With enough fiddling, we could achieve prints that, when primed, took a good hard study to determine which machine did which. My hope is that the new 1.2 version will improve on that. The biggest difference is the 1.2 uses an HD projector instead of a standard definition.
I can assure you, this will likely be the first place I begin posting my results.
Interestingly, the biggest market for the B9 is jewellers.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fb9creator.com%2Fforum%2Fdownload%2Ffile.php%3Fid%3D362%26amp%3Bt%3D1&hash=024aecdcd635525679c728d9d36257d1fcaaf960)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fb9creator.com%2Fforum%2Fdownload%2Ffile.php%3Fid%3D357%26amp%3Bt%3D1&hash=8d50480084a0634d07928da0f6106208d2dcad50)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fb9creator.com%2Fforum%2Fdownload%2Ffile.php%3Fid%3D423%26amp%3Bt%3D1&hash=9b31a302a2a35fb744e61469a8e761494542e208)
(https://b9creator.com/forum/files/uploads/2014/09/3qn1c-P9070069.JPG)
But there are some modellers.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fb9creator.com%2Fforum%2Fdownload%2Ffile.php%3Fid%3D157%26amp%3Bmode%3Dview&hash=38521a94648018837bcca46b965acfae03650e1e)
This one is N scale:
(https://b9creator.com/forum/files/uploads/2014/03/3bs57-2014-03-26-15.19.32.jpg)
(https://b9creator.com/forum/files/uploads/2014/03/8tfc6-2271628crop.jpg)
1/300 scale
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fbrumbaer.de%2Fimages%2F3D%2FFlughafen%2FF5.jpg&hash=97d98cfddbcd5a3ac46295d28cd66d8cf52dfa73)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fbrumbaer.de%2Fimages%2F3D%2FFlughafen%2FF35.jpg&hash=ea3610536a7dac9994ab90c9fd498c7c6d7eb78f)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fbrumbaer.de%2FSMX.jpg&hash=6a8e87de45322385ba970027bb5ae0bab593961a)
Price just came down about $1000, too. $4800 for an assembled printer, $3700 for a DIY assembly kit. Strangely, their authorized distributor (which is where I ordered mine) is selling the kit for $3500.
This is very interesting (I'd say "riveting", but might have to stay in the corner). Still beyond my budget and skills, but definitely something to keep an eye on.
This machine was one of those tested in Make: Annual Guide to 3d printing. The blurb at the top of the page says .. "Tinkerer's machine capable of the finest resolution of printers tested".
Thanks for the additional info. This looks promising. -- Russ
Quote from: eTraxx on February 14, 2015, 12:08:15 PM
This machine was one of those tested in Make: Annual Guide to 3d printing. The blurb at the top of the page says .. "Tinkerer's machine capable of the finest resolution of printers tested".
Yeah, the resolution was always the deal breaker for home machines, given how small the items are modelers would need. Plus, getting to run one at work and see real-world results before buying was nice.
The samples made on the B9 looks awesome. So awesome that I wonder why anyone would buy a more than ten times more expensive solidscape printer.
A little googling led to this forum thread, and it seems that the folks at MaxMini really knows a thing or two about both the B9 and the solidscape printers:
http://www.zbrushcentral.com/showthread.php?182002-MaxMini-eu-s-sketchbook-meddling-with-toy-soldiers (http://www.zbrushcentral.com/showthread.php?182002-MaxMini-eu-s-sketchbook-meddling-with-toy-soldiers)
I love how many times the guy mentions having to exaggerate details. Preach on, brother.
It shipped!
I won't have time for it until after next weekend, but it's on its way.
When can we start placing orders?
-for you to make stuff for us! ;D
JK - Mj
It arrived this week. I'm in Chicago right now for Crystal Brush (wish me luck) so I won't start setting it up until probably Tuesdayish.
And here it is. I have a bunch of other stuff to do today, so I probably won't get to it until tomorrow. I probably won't be printing until later this week, if the assembly goes well.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2FIMG_0038.JPG&hash=6c5c02fc6df7711c3c1d102323eb15675c23e37b)
Couldn't wait. The box, unpacked:
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_001.jpg&hash=a059d719fa1aa746561af801159e8386df6d0277)
Nice! The hardware is pre-sorted into assembly steps.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_002.jpg&hash=879b39d7a20137c834f82a824962dd7e37cccaac)
Step One, completed.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_003.jpg&hash=760dfa67690792d1f4e559771d14136e72398ff2)
Step two, completed.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_004.jpg&hash=75928beabc069f440060a6b82b82fa7e574cfb10)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_005.jpg&hash=33937eff63602cadac0899e93f000a419fc2bc22)
I got lazy and stopped taking pictures. It is assembled, and currently running a dry run. I'm tired, and I want to be awake while it runs it's first for reals print (which will be nothing exciting, just some 1 cm cubes with 2mm thick walls) so it'll have to wait for tomorrow.
Keep us posted! I'm sure the initial steps will be boring but a month from now you'll be turning out gorgeous resin parts. -- Russ
I wouldn't have been able to wait. And neither could you! Cool, hope to see some good results soon.
It's printing!
I HAVE CUBES.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_006.jpg&hash=296b7cf2114a4ad178b64fde64af8586c24dfece)
So does that make you square? Okay, I'm going to the corner. But to be serious for a rare moment, there appears to be "flash" or "melted resin" or something at the base of the cubes. What is it and why does it form? -- Russ
It is uncured resin. I hadn't done any cleanup as of that photo.
I also have a second print. Pics will have to come later (lunchtime, likely) but so far, it's going well. Not quite the sharpness of detail I'd like, but since I'm using a custom resin mix, it'll take some tweaking of settings to dial in the material properties. Even so, for my first "actual" print, I'm excited. If I never got better resolution off the machine, I'd be more than satisfied with my purchase.
Second Print:
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_007.jpg&hash=959ae3f00fd0d6c274f95ed04307975dea6e36ea)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_008.jpg&hash=748aa741330bcfad84bce2649e127223cc6b6c42)
As I already noted, the details a re a tad soft, but overall, I'm happy. I left the supports on, as I figured some of you might want to see that. I'll trim and clean them tonight, and post some macro photos.
Unfortunately, I don't have the files here at work, so I can't tell you off the top of my head what the sizes of those details are. The only one I know for sure is the raised "stripe" on the crest on the back of the Farseer head is .3mm wide and .3mm proud of the surface.
Oh, and the resolution was 50 micron XY, 25 micron Z.
Pretty stunning, given how early in the game it is. -- Russ
Quote from: finescalerr on March 30, 2015, 11:41:33 AM
Pretty stunning, given how early in the game it is. -- Russ
indeed! And more so when you consider the price of $4.595,-
Actually, it's only about $3500, if you assemble it yourself. Plus $100 for shipping. ($5500 for pre-assembled, I think.)
Impressive!
Very encouraging.
Quote from: Bexley on March 30, 2015, 12:31:00 PM
Actually, it's only about $3500, if you assemble it yourself. Plus $100 for shipping. ($5500 for pre-assembled, I think.)
Wow, Even better!
Could you tell us a little bit more about the skills needed to assembly the machine?
It's pretty straightforward. The trickiest bit was getting the springs onto the sweeper. Otherwise, it's all nuts and bolts, one step at a time. You do have to install a PCB/Arduino, but it just bolts on, and all of the components that plug into it already have conectors attached. If you can handle Ikea furniture, you can assemble a B9.
Though, I should qualify that by noting I may have made an assembly error. The projector is on a sliding plate, and the XY resolution is set by moving the projector up and down. The plate has a device with a fine adjustment knob, and a spring loaded pin. The back of the machine has three holes in it, corresponding to each of the XY resolution settings (70, 50, and 30 micron). To calibrate the projector, you select the resolution in the software, then move the projector into the correct position by inserting the spring pin through the plate and into the hole that corresponds to the setting you chose. Then you project a grid onto the bottom of the vat (the container that holds the resin). You take a printout of the grid, and with a combination of the projector zoom and the fine adjustment knob, you align the projected grid with the printout.
However, in the 50 micron position, even lowering the projector as far as I can with the fine adjustment knob, and zooming all the way out, I could not get the grid small enough to align. I ended up removing the spring pin entirely (which really just acts as a quick reference point for roughly where the projector should be) and lowering the projector about another two inches in order to get the image to align. It doesn't seem to be affecting the printing at all, but after asking about this on the B9 forums, the guy who made it said there's no way it should be that far off. Though, others also reported having the same problem, but not quite as drastic as mine. In some cases, replacing the fine adjustment knob (which is a plastic knob on the end of a socket head screw) with a longer bolt gives the amount of range needed, but I would require a 3.5 inch bolt for this to work. When I get home, I need to post images of the projector setup, so the problem can be diagnosed. I am also going to run through the assembly steps for that part again, to make sure I didn't mess anything up.
Nothing is ever straightforward, is it? -- Russ
New images coming. The difference between the first print and second print is night and day. I am exceedingly pleased with the crispness of the detail. And this time, I'll have measurements to go along with the images. And a puzzle!
The lack of crispness was determined to be caused by light bleed. To lower the amount of bleed, I doubled the amount of additional pigment that I added to the resin I used in the first print. It's much better, but I'm going to try adding even more pigment to see if I can really crisp it up. (After a point, adding more pigment doesn't make any difference, and too much will prevent light penetration into the resin, causing undercuring and possibly print failure.) I also dropped the exposure time a hair.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_013.jpg&hash=840d3314daec7ad2e2123b0b0ed13c7b899b57e9)
Small machine guns, designed to modify HO scale cars for a vehicle combat miniatures game. Printed on sprues for molding.
The flash supressor/venting holes are .6mm in diameter. The stroke width of the letters in the copyright is .3mm wide. The stroke
width of the circle around the "C" of the copyright symbol is .15mm. Each gun is 13.7mm long.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_014.jpg&hash=815d0105bc1dc3e3c3c9255673a08763da6552f8)
Size reference.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_015.jpg&hash=e65d1f87525d560152452a7a7821b9c673be1907)
The hammer on the left was from the most recent print, the one on the right is from the first print.
And the puzzle: on the support/sprue blocks, the layer lines are clearly visible. However, those parts were actually at an angle. The guns themselves were aligned to be vertical, and the blocks with the text were at a 20 degree angle to the Z axis. So what I can't figure out is, why the lines are parallel to the blocks and not at a 20 degree angle. (And further, why they aren't visible at all on the actual gun parts, but very visible on the sprue blocks. Not that I'm complaining about that, as I don't care if they're visible on the blocks, so long as they're not visible on the actual parts.
Also, in the first image, the guns on the right appear a bit squished. That seems to be due to the camera angle, as they look fine side by side here on my desk.
Also, it should be noted that the text is a little soft on purpose. Since these are intended to be molded, I filleted all the corners on anything I didn't actually need to be crisp, to lessen the wear from demolding.
Also also: I have yet to try the highest resolution available. I've been using 50 micron xy and 25 z. The z in theory can be as little as 5 microns, though the layers are so thin it's hard to control light bleed. 25 seems to be the best, as higher resolutions in the z axis don't result in considerable gains, but increase the printing time a lot. For xy , there are three settings- 30 micron, 50 micron, and 70 micron. The trade off is that the xy is adjusted by moving the projector closer or further away, which alters the size of the print area. Increasing xy resolution decreases how much you can print at once. The 30 micron area is only 32.4mm x 57.6mm, so I'd like to get the leave of detail I want at 50 if possible, since I do want to be able to print larger models.
Unless I'm mistaken then, the resolution of the parts you've printed makes them about equal to HO scale molded styrene. If so, this is extremely promising and pretty darned impressive. -- Russ
Hard to say. Those guns are nominally HO scale, but they're pretty oversized and cartoony, so that they read well on the tabletop for gaming. The barrels are probably twice as thick as a man's arm at scale. If anybody has any "true" HO scale 3d files, I'd be happy to give 'em a test print. Nothing too large, though, as I'd rather avoid 20 hour prints for the time being. Most of what I have right now is 32mm figure scale or larger.
Looks as good as injection-molded to me.
Here's a HO Water Cannon I created for a friend. The 'post' under the flange is 1/16" dia. The small 'bolts' on the square flange are 0.1mm dia and 0.1mm long .. all the flanges are also 0.1mm .. minimum for 'embossing' ... just to give a size reference. This was to meet the constraints for FUD at Shapways. I'm interested to see hwo well they print. It's a .stl file zipped.
(this is pushing FUD to it screams in pain .. "That's all she can take Captain!"
http://www.etraxx.com/stl/single.zip
I'll give it a try when I get home tonight.
Hm. It's going to be tricksy, because the overhangs are so small. I may have to cut it in half. Actually, I'll probably try it both ways, cut in half, and with tiny tiny supports under the overhangs.
Didn't cut the pump, just added supports. I had to use some pretty fine ones to support some of the overhangs, so they may not be strong enough. Also printing another hammer and sprue of guns, for reference. I upped the pigment content again, closer to the upper limit of what I can put in. We'll see what happens. Got it running early enough that if I stay up a bit late, I should be able to clean and photograph everything tonight. If the coming rain storm doesn't knock my power out.
What is "rain"? That is something we in Southern California have never seen. Well, at least not for about five years. This year's annual total was only about 4 inches and I am very concerned about the future here.
As for your prints, I hope I speak for more than just a few people when I say this whole thing is fascinating and I very much appreciate your frequent and detailed posts.
Russ
Well, the power didn't go out. At least, the electricity. I fell asleep on the couch. I checked it briefly this morning, and everything appears to have printed. But I won't be able to say how successful the print was until I get home tonight.
I pulled the parts out and cleaned them. It looks like I had some small failures. I suspect that I need to back off the pigment amount a little. Making the resin more opaque decreases light bleed, which gives crisper details. However, it also reduces light penetration in general, making the layer cure slightly less. I think the initial layer didn't attach to the build surface well- it looks like it shifted a bit. Luckily, the subsequent layers adhered to that partial layer, so the print didn't entirely fail.
I put two of Ed's parts in the print. One fully printed, one partially printed. I think the partial was a victim of the initial layer shift. Interestingly, two of the supports I added higher up the model for stability didn't fail, so it only failed to print up until the point where the first of those supports joined the model. What was surprising was that it appears to have printed well from that point on. The successful one looks good, but it's so small, it's hard to tell how well it printed without some macro photography. I'm hoping those hex heads will actually e hexagons when blown up larger. Photos soon; the items are post curing right now.
The other parts are as crisp as I could want them for mold masters, so I'm pretty thrilled.
Picture time!
First, the layer shift. The parts in the center and right are Ed's, the one in the center is the one that partially failed.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_017.jpg&hash=e5dc53be5156d27fec7515a9c74fa1181473f97c)
The guns, again. The one on the right is the newer one, the left is from the last print.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_018.jpg&hash=48085e253a6f8ade350a1f9833e061bd8d74c46d)
And hammers. older to newest from left to right.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_019.jpg&hash=b9f25db6bd857830eaddfc4f40e7e604f1ff25bd)
Ed's part. Not the greatest photos. It's tough to tell in the image, but looking through a loupe, the hexagon heads are almost hexagons. Sort of halfway between hexagonal and round. It's also tough to see in these photos, but the tiniest little rivets are there too, but they're pretty soft.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_020.jpg&hash=30e4ace1e16a22001d451fa3b64a7adfd60aef5b)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_021.jpg&hash=446eb56b8bb78e9b08531ef81acf91486de6c3e3)
If/when I recalibrate to prinvat 30 micron xy, I'll toss Ed's model on there and see how it comes out.
Today's improvement is very evident. -- Russ
Just to show what B is working with here ..
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fimages57.fotki.com%2Fv139%2Fphotos%2F2%2F1709102%2F13658350%2Fphoto-vi.jpg&hash=36dcc28ea0c2da784a6ebfb9dc5bbda753a1019c)
The 0.3mm groove was put there so if someone wanted to use the water cannon without the little 1.6mm dia post they could snap it off at that point (or cutting guide?). Notice it printed quite well! The hex head is only 0.346mm across the flats .. this is TINY!!
I thought it would make a good practice object since it is scraping along the edges of the constrainst / minmimums for printing in FUD at Shapeways.
For those interested .. trying to create something in HO that 'sorta kinda' resembles the full-size water cannon, here is what I used as a model
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fimages40.fotki.com%2Fv1075%2Fphotos%2F2%2F1709102%2F13658350%2Fphoto-vi.png&hash=8d94094bc79fac9e115af42598783497b2b8e304)
This weekend, I'm going to try a long print- the fuselage for a little Orky airplane, which is about five inches long. The vat only holds a small amount of resin (well, probably over a liter would fit, but the fill line is low in the vat) so for longer prints, you need to be there to top it off. But if I have time, I'll likely try a 30xy print, and see how the water cannon does then. Also, to see if there's a noticeable improvement between 30 and 50.
Just hit start on my first really long print- a little over eighteen hours. It'll be done around 1 pm tomorrow.
Appears to have been printing without problems when I checked this morning. Though, I accidentally gave the machine a good whack when I was topping off the resin, so I expect there to be a line/shift in layers from the projector moving. We'll see.
Latest print. Came out mostly okay. The fuselage halves are about 4" long, for reference.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_022.jpg&hash=90359e0018c528ed9af8a9bcf4b9e6e9972b53fc)
Except some spots where the layers didn't quite attach to each other:
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_023.jpg&hash=ed9187d89b1f5b40112df7a38ffa2d52965de3b0)
Bexley, what is the material like? Brittle, or soft or?
It's... both? It sands nicely. It has a little bit of flex to it, but it'll snap if I try to bend it much, like urethane resin. I'm still waiting on my UV cure lamp so I can post-cure them fully, that might change things. But like most 3d printing UV curing resins (at least, those that at I've had any experience with) it's not going to stand up to heavy use/abuse.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_027.jpg&hash=c4285727003361e367586b5d87e290ac0a56b7cf)
Yowzah! neat stuff.
It appears the "delamination" was due to a lack of supports. Somehow I missed those areas when setting up the print. One of the areas did have supports, but needed more. Pretty rookie mistake, and I feel kinda dumb. But at least, it was operator error and not something hard to diagnose.
That is going to look so cool with wings and paint!
Is the "delamination" the area that appears to have a herringbone pattern or is that intentional? -- Russ
Russ, I think the delamination might be the area in the green circle, but I'm also curious about the patterning that to me looks like woodgrain, one sample area marked by the arrowhead. Is that what you referrred to as 'herringbone'.
Anyway, Bexley, is that texture on purpose? Seems like it is.
The delamination is in the big picture, lower right side panel with curved lamination.
Nice results from that printer.
8)
"Delamination"
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_023a.jpg&hash=1623412801883ec389cb8a1ebcf36ed29073bb31)
The patterning is from the layering. The sides are gently tapered, so it's more pronounced than in other areas.
Thanks for pointing that out, Bill. The piece looked so good to me late last night I didn't even notice the delamination. Maybe I thought it was intentional but how am I supposed to know what I think?
So what about the layering pattern? Would you have to sand it off? Is it fine enough that primer and paint would hide it completely? Would another attempt at printing eliminate both imperfections?
Russ
The delamination was my fault, I needed to add supports there. Essentially, the only thing holding that spot in position as the vat slid away to release the layer was where the layer attached to the rest of the model, so tas the vat pulled away that section was a bit loose.
The layering/texture just needs a light sanding. Mr. Surfacer might even be enough to fill it in, but I don't have any on hand at the moment. It's not that much worse in person than what I see at work on the very expensive machine.
I had my first failure last night. The PDMS (silicone release layer) gave out. Surprisingly, some of the parts still came out (pics later) but the area right over the worn bit was all wobbly and disfigured. This is the one real disadvantage of the B9 (though, as far as I know, all other printers in this price range have the same issue): the curing of the resin on the silicone wears out the silicone. This can be alleviated by placing models in different areas, so that the same spots aren't getting repeatedly worn, but eventually, the layer needs to be replaced. It's a simple procedure, especially with their most recent vat design, but it's another expense. I bought a second vat, though, so that I'll always be able to have one ready to go in case of a PDMS failure. The upcoming software update will supposedly double the life of the PDMS (instead of returning the vat to the same position after every layer, it offsets to a slightly different position each time, so that the same spots on the PDMS aren't getting hit every layer.
Wrecked PDMS- the clouding is from repeated exposures. The ripped up area in the upper left is what caused the print to fail. Replacing it is pretty simple, the coating is a silicone called Sylgard 184. Place the vat on a level surface, pour 66 ml of mixed silicone into the raised side of the vat. The raised side has a lip, so most of the silicone stays in that half, but a small amount will work it's way over the lip as it levels itself, forming the dropoff point required to release suction. (Older style vats required damming one side, pouring it in, moving it around, etc. Much easier now.)
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_028.jpg&hash=b3caa1f4c341fa1093f421966e4a3d9dfd00b2a3)
Failed print.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_029.jpg&hash=7b717058c4308d01787b73b8c5adc9d0531ed597)
:(
It is becoming clear that 3-D printing is a delicate process requiring a lot of experience. As with many disciplines, the operator is more important than the machine. -- Russ
Quote from: finescalerr on April 08, 2015, 12:21:37 AM
It is becoming clear that 3-D printing is a delicate process requiring a lot of experience. As with many disciplines, the operator is more important than the machine. -- Russ
That is my conclusion, too.
So I have given up the thought of getting a 3D printer myself. Even if it i less convinient, I think that in the end I will spend less time and money pr. part if I use commercial providers. And now that Shapeways has announces Frosted Extreme Detail with 16 micron layers, I am even more sure I will not sink several thousand dollars into getting a printer for my workshop.
A personal 3D printer is a hobby in itself, and I will rather spend my time researching prototypes and making the neccesary scale drawings and prepare artwork for etchings, 3D models for printing etc.
This said, I am impressed by the effort Bexley is putting into getting the most out of his B9. The results are already mighty impressing, especially considering that he has used the machine only for a few days!
His postings in this thread is a veritable gold mine for scale modellers that plan to get a B9. As always, the information you get on this forum is incredible.
Agreed with the others that there seems to be(like in everything) a lot more to the story!
It does occurs to me though that the "failed" prints might make for good junk, you might even be able to sell them as such. At the least they would make for good painting practice.
Paul
Where would I begin to try and learn basic CAD to draw parts to then send to shapeways?
Is this even possible to learn and have usable parts?
It would have to be something I could do in my spare time, formal classes wouldn't work (right now).
I can draft using paper & pencil but no idea about CAD.
I work at a technical college so I "could" get with our instructor here when I get stuck.
Is sketch-up a good starting point or should i just buck up and learn the real thing?
Of course spending hundreds on software does not appeal to me either.
Marty
Quote from: lab-dad on April 08, 2015, 06:13:30 AM
Where would I begin to try and learn basic CAD to draw parts to then send to shapeways?
Is this even possible to learn and have usable parts?
Is sketch-up a good starting point or should i just buck up and learn the real thing?
Marty,
it depends on what you're going to achieve. SU is a good starting point when it's about 3D-printing. And it's free. Whenever you're going to have some parts cnc milled / lathed - forget it. For that I'd prefer a professional CAD program. I think they start at about 1.000 € or something (Rhino 3D) - but: If you're working in educative sector for example Autodesk offers special free or low cost licences. Anyway - it takes time to learn those programs and you will easily spend much more time at the computer than at the bench.
Quote from: Hauk on April 08, 2015, 04:05:08 AMEven if it i less convinient, I think that in the end I will spend less time and money pr. part if I use commercial providers.
You nailed it.
Volker
I think a lot of the people here use Sketchup. I know it used to require plugins to get printable files, but that was a long time ago, and it might be able to export .STLs natively now. It's a pretty good program, though it does have some quirks to it at times. I used it quite a bit before learning Solidworks.
You can get a Student's copy of Solidworks with a DD 214 for $20. James Lincoln dropped by the other day and gave me a short and swift tutorial on using it. For now I will stick with SU. I learned from tutorials on YouTube. Combined with MeshLab and netFabb you are really not limited .. Mr Lincoln pretty much hit the mark when he compared SU and Solidworks to a Model T and a Ferrari. In the end, both will get you there but the Ferrari is more comfortable and faster.
I will hold judgement on Shapeway's FXD. The news release only mentioned the decrease in the Z-axis from 29 microns to 16 microns. That will of course help with the stepping hyou get .. that is a 55% reduction after all in layer height. There was no mention of any change in the X,Y resolution so I have to wonder about how that will work with the halved Z resolution. FUD has a minimum of 0.1mm which works out to a 254 dpi resolution for FUD. Thing is .. we are working in three dimensions so .. just how halving the Z axis resolution and keeping the same X,Y resolution will work .. who knows. I surely don't.
Here's another photo of the last bits I got from Shapeways .. this time stacked using Helicon Focus .. and ALL the warts show. Some of the roughness can be attributed to unsanded primer but not all .. pretty much showing the (by my calculations) 254 dpi of FUD.
Yeah, my experience with Sketchup was that it was great, and it it did a lot of cool stuff. I used it quite a bit, but after learning Solidworks, I was constantly running up against the things that SU doesn't do.
SketchUp is less suited to parts with compound curves. You can create them but the expensive programs do them easily and probably with greater precision. For most stuff we would create, SketchUp works quite well. Ed Traxler is our resident guru. He really knows SketchUp and can help you out of a lot of jams. Marc is an expert, too, and probably would help if he ever pops back in. He did take the trouble to create a basic SketchUp tutorial here and it definitely will get you started. It's in the stickies and easy to find. -- Russ
For all my CAD drawings for the laser cutter I use a simple free program called DoubleCad XT2.
I design all my kits in this program but they are only 2 dimensional.
For my 3D parts, I do everything in free SketchUp following Ed Traxlers tutorials.
There are some quirks in learning how to get those curves that Russ mentioned but there is help out there.
I'm just not into spending money for software if I don't have to. And my results are satisfactory.
Marty, the program Dave refers to is the free version of AutoCAD LT I e-mailed you about (but didn't remember the name). You will never need anything more sophisticated for the kind of stuff we do here. You should download it because it could be very useful for creating any part a laser can cut (like that stuff I drew for you and Tom at Crystal River cut). I and others here can answer any questions you have if you start to use it. -- Russ
Yup, DoubleCad XT http://www.doublecad.com/ (http://www.doublecad.com/) is put out by IMSI, the same folks who designed Turbo CAD many years ago.
And it works just like AutoCAD LT.
And another good part is that DoubleCAD can export your drawings in many different formats to work with laser cutters and CNC machines.
And as Russ said, it is all we really need for our type of modeling.
No more prints for a bit. I'm heading to San Francisco for Friday and Saturday, so no prints until probably Sunday or Monday.
Discovered some things. I tried re-running the last print that failed on a brand new vat. Unfortunately, I got the same result- the print semi-failed. Worse, though, was that it ruined the PDMS, pretty much the same as last time, though not quite as heavily. This was particularly distressing, since one of the things I want to do is produce masters for model kits, and not being able to print large items without detroying the PDMS, or wrecking it after only one print was going to be a problem.
Then, I realized that I was thinking about the build table orientation wrong. The parts of the print that actually succeeded were over the destroyed part of the PDMS. So while that is still a concern, it wasn't the reason for the failure. After some intense forum-reading, I learned that I should check the flatness of the build table. Sure enough, it was warped. The areas where the print was succeeding was flat, but where the parts weren't attaching properly, the table was a little high. It was still partially printing, though, because the initial base layer was large enough to cover both the flat and high spots. I tried a short print of a single item, right in the center of the high spot. It failed entirely- essentially, each layer would print, but not attach. Then the sweeper would wipe it off of the PDMS when the vat slid over. So the end result was a vat full of resin, with all the individual layers floating around.
When I return from my short trip, I will be bonding a thicker piece of aluminum to the build table, and then flattening it with sandpaper on a granite block. This was how a number of other users solved this problem.
As for the early destruction of the PDMS, I contacted the company directly to ask about it. It was most likely over exposure of the resin. It isn't the light that wrecks the PDMS, it is the curing of the resin. Ordinarily, each layer is cured just enough to stick to the build table, then lifted off. Light from printing subsequent layers continues through the current layer into previous layers, so that over time, all the layers are cured more as the print grows. However, that fuller cure does not occur in contact with the PDMS. So I likely need to lower my exposure settings quite a bit.
Hopefully, the above solutions will work. The parts that printed well at 50 micron slices look just as good as those I printed with 25 micron slices, as the models I'm trying to print are not high detail items. But the ~50% lower print time is nice, so I'd like to be able to get this working at that slice thickness.
Are these problems common with all printers or unique to the Be-nign? It really seems to be a finicky process. -- Russ
Not entirely. The not-flat table is uncommon. As for the exposure/curing problems, those are mostly due to me experimenting with 3rd party resins, which I then further alter with different pigment concentrations. The printer software comes with standard settings for B9 resins. B9's resin is about $110 a liter, the stuff I'm using is about $55 a liter. It's also a faster resin- about twice as fast as B9. The B9 stuff is designed primarily for jewelry, so it has a high wax content, and can be burned out for investment casting. That wax causes a little loss of crispness, though. That is the primary reason for using the 3rd party (MakerJuice SubG+) resin, though the speed and price difference are also big factors.
That said, B9 will be shortly releasing a non-burnable fast resin, which I will certainly try out once it becomes available. If it's as good as the MakerJuice stuff, I would have little problem trading cost for consistency and pre-determined settings.
And that said, the machine is intended as sort of an experimental machine. If it were a computer, it'd be an Altair 8800 (http://en.wikipedia.org/wiki/Altair_8800)- a DIY kit that attempts to rival commercial machines.
Bexley,
I will make a suggestion on the piece of aluminum you use to bond to the build table. Use "tooling plate" of "jig plate" This is aluminum that is cast and its surfaces are then ground. It is used for making jig and fixtures in machine shops and because it is cast it is much likely to warp when cut. You can get it from Mcmaster:
http://www.mcmaster.com/#standard-aluminum-sheets/=woukbw
How thick and how big of a piece do you need? I might have some...
Paul
Printing again tonight. I needed a ton of 1/35 bottlecaps for a diorama. Since they are so small, I re-calibrated for 30 xy. What's kind of interesting is, it's actually a faster print, as moving the projector that close increases the strength of the projection, meaning shorter cure times, even though I'm slicing at 25 microns z.
Still waiting for the aluminum plate I ordered. Since the 30xy build area is centered on the part of the table that isn't warped, it should still work.
Anyway. These might be too small to look like much more than slightly ribbed discs, but we'll see. I'll likely run another batch of them if I have no problems, and if that's the case, I'll give Ed's part a shot at the higher xy resolution.
Success! Photos later. The bottlecaps don't look like much, but under a 10x loupe, I can actually see the little ridges. I think I will need to redesign them to be a little out of scale (larger) so that they read in the diorama.
Also printed Ed's part, x3. All printed, and look pretty good. Ed- did you ever get these printed at Shapeways, and if so, do you have photos?
Bottlecaps!
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwyrd-games.net%2Fcommunity%2Fuploads%2Fgallery%2Fcategory_12%2Fgallery_13970_12_33466.jpg&hash=a2c11594dc7dd41f6b3a52ae0ceadacf96e051a1)
I can for the life of me not think of what theme for a diorama that needs a ton of bottle caps.
Perhaps it is a scene with an army of warriors like this?
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fideas.coolest-homemade-costumes.com%2Ffiles%2F2013%2F11%2Fcapped-crusader-drinking-in-the-name-of-art-96877-618x800.jpg&hash=6c9f676e259004de31211a8475f70a3e507e8f06)
It is a Fallout inspired scene. Post apocalyptic, bottlecaps have become the standard form of currency. I needed a pile of money.
I think any scene in Russia....would have this many bottle caps....or on one of the duty free ferries from Finland to Tallin. :P :D
New print. Fallout laser rifle, frag mines, and microfusion cells. 1/35 scale.
I'm particularly surprised that the thin (.4mm) rod on the gun printed. I considered leaving it off the model, and using brass wire. But I figured I'd give it a shot and see what happened.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwyrd-games.net%2Fcommunity%2Fuploads%2Fgallery%2Fcategory_12%2Fgallery_13970_12_316976.jpg&hash=b022bdda7cbe24633b0d510ae2b406733f66bf39)
Also, a 30xy vs. a 50xy. The 30 is on the left. I was happy with what I got at 50, but the 30 is noticeably better. It's just a shame the print area at 30 is only about 30x50mm, otherwise I'd never need to go back to 50xy.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fwww.counterclockwiseminis.com%2Fimages%2Fb9%2Fb9_035.jpg&hash=cc78c3b1cd54e53ea619e30f4af548400f4cae54)
The 30xy print really is nice. Thanks for keeping us informed. I'm impressed and a bit in awe of 3D printing. Perhaps I should start learning Sketchup or something similar.
Bill Martinsen
Quote from: Bexley on April 16, 2015, 11:42:50 AM
Success! Photos later. The bottlecaps don't look like much, but under a 10x loupe, I can actually see the little ridges. I think I will need to redesign them to be a little out of scale (larger) so that they read in the diorama.
Also printed Ed's part, x3. All printed, and look pretty good. Ed- did you ever get these printed at Shapeways, and if so, do you have photos?
Long story .. shortened. They sent me the wrong print that looked almost like my watercannons .. they not only re-printed them .. but my entire order! They are in-route and should be here Monday.
Also, I got word from B9 that they'll be taking orders on their new fast black resin next week. They haven't posted any prints done with it yet, but based on what they're saying, it'll behave a lot like my custom mix. More importantly, they've already done all the trial and error for determining printer settings for it, so I should be able to dial it in much faster. Not to mention, not having to mix the resin myself anymore.
Gorgeous prints. -- Russ
lousy photo of the water cannon. It would be better primered but gonna send these off to they guy I made them for 'as is'.
Been a while...
Got back to messing with the printer. While I was able to print just fine at 30xy, I really wanted to get 50xy working, as there are many future projects where I will need the additional space.
I got some (much more expensive) black B9 resin, to help with calibration. Unfortunately, I had pretty much the same issues. What I eventually determined was that since I had to lower the projector more than I'm supposed to in order to calibrate, the actual lamp power reaching the resin was reduced. In theory, I could compensate by various means, but nothing that wouldn't involve a lot of trial and error every time I wanted to switch from 30xy to 50xy. I contacted customer support (actually, they contacted me, after reading about my issues in the B9 forums) and it is likely there is something out of whack with the lens elements. I will be returning the projector and getting a replacement, which will hopefully solve my problems.
That is not the first problem you have had with that machine. I hope the manufacturer was a generous in helping with the other issues. -- Russ
Yes and no. It is likely the source of all of my other issues. Even the not-so-flat build table might have still resulted in successful prints at 50xy if enough power was reaching the resin. Before I flattened it, I was able to print fine at 30xy; at 50xy, it I had problems both before and after flattening. It's more that there were a lot of potential diagnoses, which all had to either be resolved or eliminated as problem sources before starting to look at bigger picture things/more unlikely things. And certainly, had I not jumped in using a third party resin with unknown exposure settings, I might have gotten to this point much faster. Bad optics is not a common issue. And technically, this would be Vivitek's fault. But I suppose, if I were using the projector as intended, being a little out of focus along one edge probably wouldn't be a problem, even if I were able to actually notice that in a 10' wide projection.
Ultimately, it's my own usual bad luck. The B9 forums are full of people who own machines (and in some cases multiple machines) who have never had a problem. (I suppose they could all be plants, though...)
And, the cost of getting ~$100k machine quality from a $4k machine I assembled myself. I knew going in it wasn't a plug-and-play device.
I've bee quietly keeping an eye on this thread. The model RR club at RPI has been discussing getting both a laser and a stereographic printer not only to help construct large numbers of models, but also to help recruit new student members.
Given their courses, most students have pretty heavy time restrictions so having potentially nifty tools that actuallyhave a lot of down time for whatever reasons is one of the reasons this is still in the discussion phase. (The other reason is the cost of tools like that).
I think at this point, it comes down to how much you want to tinker. The printers with high enough resolution for small scale models are either very expensive, or have a heavy DIY element to them. There are lower cost models that are more reliable, but they currently lack the resolution. Even Dremel makes one now, and you can get it through Home Depot (http://www.homedepot.com/p/Dremel-Idea-Builder-3D-Printer-3D20-01/205448581) (online only) for $1000 USD, but unless you want to print larger components without detail for hand-detailing later (not a bad thing) it's probably going to disappoint.
Hey, it's been a while.
It seems that my printer woes were based on operator error during assembly. The upside is, I am now printing again, and getting the best prints yet!
These are some custom bases for some figures I'll be casting in resin. (25mm in diameter.) They were printed at "medium" resolution, as a test. I will be reprinting them at higher res in the near future for molding. Unfortunately, the nice black resin doesn't photograph very well.
(https://scontent-ord1-1.xx.fbcdn.net/hphotos-xpt1/t31.0-8/12484766_1681481935455058_3273391833013197478_o.jpg)
And this is a Zbrush sculpt I'm working on for a 1/16th figure. I printed it last night, but didn't have time to clean and inspect the print before work this morning. (I did check on it, though, and it appears to have printed fine.) This is way over detailed, as I was having fun monkeying around with the sculpting, knowing full well that most of it was too fine for even top-of-the-line printers. The stitching should print, though, and some of the larger dents and damage. The texture on the helmet and leather will likely not be noticeably unless I printed at 1/6th or so, maybe.
(https://www.finescalerr.com/smf/proxy.php?request=http%3A%2F%2Fic.pics.livejournal.com%2Fjimbexleyspeed%2F50867845%2F2038%2F2038_600.jpg&hash=66e4b47526253f5499b80f6a09c73bfdca8d02bd)
The disembodied head looks terrific. Do I understand correctly that the printer wasn't the problem -- you were? I thought the table wasn't level and there were some other issues. -- Russ
Yes, it was (mostly) me. The table not being flat was a factor, which is why things improved once I lapped it. But the "shelf" that the resin vat sits on needs to be square to the back, and the two halves of the back need to be planar with each other. Somehow during assembly, these parts came out of alignment, so that the front of the machine was a bit further from the projector plane than the back. At the highest resolution, the build area is closer to the back of the machine, so I had no problems. But when I went up to the medium resolution, the build area expands. The parts toward the front of the machine were not getting as much exposure, and failing. Since each layer only gets a half second or so of exposure, it doesn't take much to cause the resin to not solidify completely. In the end, I bolted some angle iron across the back of the machine to keep it planar and rigid, so tightening the bolts to square the shelf couldn't pull things out of alignment.
Printed the head last night. I hit it with a bit of primer so I could get a decent photo.
(https://scontent-ord1-1.xx.fbcdn.net/hphotos-xaf1/v/t1.0-9/12644698_1688128328123752_7981894764879323923_n.jpg?oh=68dc99d88cc8e203cc0773aebf5da185&oe=57277270)
Nice! It retained a surprising level of detail.
Indeed. I got about what I was expecting. And once I put the head on the model, it was clearly a bit too small. I'm reprinting it now about 10% larger. Probably not enough to significantly impact the finer details, but it can't hurt them, either.
(https://scontent-ord1-1.xx.fbcdn.net/hphotos-xtf1/t31.0-8/12622079_1688406971429221_5983920437510995647_o.jpg)
Wow! Very exciting. Looks great...Allan
Approximately how tall is the figure on the saddle? -- Russ
When I saw you were printing a 1/16 figure I got really excited!
Dont get me wrong - it looks amazing, just not one I could use.
The level of detail is really great- I have one from shapeways that looks good in white primer. I guess I need to paint it!
Keep up the great work!
Mj