I believe it's aluminum or zirconium oxide ceramic. I recently bought some kitchen knives that have the same surface feel. I looked around on eBay, but couldn't find anything exactly right. Biggest I could find was maybe 5x5". I would think 1/16" think is about right. The adhesive is probably that 3M stuff everybody uses to hold down PEI sheets. I'll see if I can find some actual links... Edit: Here's a 6x4" alumina sheet. Can't tell the thickness, but you get the idea. THIN ALUMINA CERAMIC SHEET 6 and a half x 4 and a half INCHES No.: 27 | eBay
Sweet... yes and no. Early units have a polymer that has signs of thermal expansion stress shearing the material at the magnets. Later versions are prone to pitting on ABS. Alumina is great but not on aluminum. Alumina has very little CTE change compared to aluminum. It will try to bow and the alumina will just act like a spring. Now machining pockets in an alumina sandwich... excellent! Ps... .030" thick
Or you could avoid the whole thing and print on glass. You can use 5x7" glass from any picture frame. Throw in a few bulldog clips from the dollar store, and you're good to go. You still need the build plate to be white enough to reflect the light during leveling and z-gapping, but then manually z-gap with the glass plate on, and you're golden.
Excessive adhesion! Wow! Half the industry trying to reach this on a cold bed and you find it an issue! heheh
Same guy has 6" x 6" x .04" for $24. Ordered. LARGE THIN SQUARE 6" x 6" ALUMINA CERAMIC SHEET SUBSTRATE No.: 18 | eBay I'm going to try this with very selective use of trim tape. That should overcome the CTE difference by compensating for shear. It will affect the z-height, but not enough to care. BTW; you do know that the "Calibration" setting can be adjusted for a new default Z-gap. There is also a hidden button on the calibrate routine after "cal 2". I did test this and it works wonderfully. So if you have a bed that crashes into the nozzle, this is where it is set, but you have to take it through the entire routine.
I just wanted to say that alumina has a thermal shock resistance of about 100K. You can calculate it if you want. So if you print on this surface it should get micro cracks and chip off in small pieces over time. The plate I have is no alumina bed plate. It is somekind of thermoplast coating. Do a test, try to carve something in with a knife. If it is a ceramic substrate then you will not get any indentations on the surface.
I work with alumina in my professional life on a regular basis. That is one surface that I can depend on. Macor, however, does exactly what you explain. You can buy alumina in a "half hard" state (bisque) that allows you to machine it and later fire it again to get it fully hard. Alumina bisque is like a very hard chalk. It is useless in this state except for machining. When you re-fire it for full hardness, it shrinks again... a lot. Machining Macor is exactly that, micro-cracked chip removal. Dropping a hot alumina plate in cold water, however, yea, not a good idea Not planning on a heated bed as yet. The material on my plate is flexible... very flexible at the magnets. The big problem is the aluminum. It is dimensionally unstable as temperature changes. Alumina, on the other hand is one of the most stable at elevated temperatures. There is no way I can join them without something that gives. I could glue the magnets to a ceramic plate but at 1mm, just a bit thin for my liking. If this works as a surface, I'll get the thicker version and see what I can do with that.
I also work with alumina in my day to day job I am testing differente grades for thermal shock resistance, thats why i did mention it. But if you want a really good glue for glueing aluminium to alumina you should get your self HTK Ultramond, I do get pulloff strenghts of 100 MPa with this stuff. Would be great if it works with the alumina plates, it is cheap and you can get it everywhere.
The closest thing in CTE to alumina is Kovar... and a close second is Ti. I would never join aluminum to alumina directly. Over 6", that would likely curl more than the thin sheet could handle. What kind of shock are you subjecting the alumina to? We use it in ultra-clean vacuum environments at significantly elevated temperature. Any kind of alumina dust would be catastrophic in application. I do understand that braze operations will generate a lot of micro-fractures even with controlled cool rates. Most of what I work with is 96-98%. Below that I consider it "unstable" due to flaws. The science changes again when you get into ultra-pure over 99%. It is not too far fetched to actually order some very custom build plates for the Cube 3. Not sure what the MOQ would need to be. But at $100 for a 3D Systems replacement, it allows some leeway.
I just bought a 6x6" plate of Mic6 aluminum from onlinemetals.com, hoping to use it for the print bed on a cube-rap. (Mic6 is what Printrbot has been using for their heated beds since the Simple Metal, I think.) It's cast and then milled, so it's very, very flat, and supposedly warps less than any other aluminum alloy. Unfortunately, it's only sold in 1/4" thickness and above.... which ends up being waaaaay too heavy to mount on the cube3's tiny Y linear rail. I guess the Printrbot guys must have them custom milled out underneath to cut down on the weight.
Very flat and rigid is 7075 aluminum. It cracks when you try to bend it. Problem with aluminum is that it expands significantly as it heats up. As a heated bed, however, it would be stable since the temperature doesn't change. If you want light weight, stop by the hobby store and check out their carbon fiber sheets. Easily machined and tolerates heat. Probably still expands like aluminum but it will tolerate heaters.
The build plate on my cube 3 is much too soft (scratches and gouges) to be ceramic or composite. I always thought it was just a teflon sheet: McMaster-Carr
Anybody have any extra nearly empty cube 3 ABS cartridges or empty PLA cartridges they're willing to sell?
Yeah, that's the whole point of the Mic-6. According to the specs from Alcoa, the surface is machined to 0.5 micron smoothness, and you can continuously cycle the temp up to 600F and it won't change shape. Of course the specs I found for 7075-T6 listed the same CTE as Mic-6, and onlinemetals.com only gives you specific specs if you ask for them at checkout... which I completely forgot to do. I guess the big difference is the surface flatness out-of-the-box. 7075-T6 specs: http://www.crpmeccanica.com/PDF/aluminium-7075-t6-7075-t651.pdf Mic-6 specs: https://www.arconic.com/mill_products/catalog/mic-6.pdf
I've been doing a bunch of printing with my cube using simplify3d, mostly trying to find out the best settings in simplify3d to get good print quality. It's been challenging. The reason your bed is dropping like that is because for some reason simplify3d is putting in an M001 instead of M101. M001 tells it to stop printing, and when the cube is done printing it automatically goes to the park position (bed all the way down and forward), then its next command it gets is part of the regular printing so it raises back up and keep printing properly until it hits the next M001. What I did was in my simplify3d profile, at the bottom of the scripts page in the post processing, I added {REPLACE "M001" "M101"} . I've done a fair amount of research into bfb files and I'm sure it's meant to be changed like this instead of stripping it out, tho I may try that at some point. The biggest problem I've ran into is retractions, simplify3d doesn't make the bfb gcode the same as the cube print slicer for retraction and it's been hard to find an optimal setting. Here's my entire post processing scripts, just to help make it more clear (most was also from a cube pro profile): {STRIP ";"} {STRIP "G90"} {STRIP "M82"} {REPLACE "MATERIAL_TYPE1" "214"} {REPLACE "MATERIAL_TYPE2" "214"} {REPLACE "M106 S0" "M107"} {REPLACE "M106 S" "M106 P"} {REPLACE "M102" "M103"} {TOOL1REPLACE "M101" "M201"} {TOOL2REPLACE "M101" "M301"} {REPLACE "M001" "M101"} {STRIP "T0"} {STRIP "T1"} {STRIP "T2"} {STRIP ";"} {REPLACE "\n" "\r\n"} {EXTENSION "bfb"}
So adding this to s3d, saving as Bfb and then run into the encoder does the whole trick? Mind sharing your fff as well?
What I'm going to do is write up a "how to" about using other slicers with the cube 3, along with my fff and my kisslicer settings. Will probably do it on another board and post a link here.
The whole idea about casting is to get rid of residual stress within the material. Funny enough, the plate flatness means little when just the pressure of the nozzle can deflect the entire stage in Z. But you are right, 7075 is only as flat as the mill run unless you order it with special post processing. Finish of often optional. I'm pretty sure that is what 3D System's used. My other concern is what the leveling algorithm is looking for in the surface consistency. It knows when there is glue there. It purposefully avoids the magnet zones.
Willing to buy an empty (0%) ABS chip (chip only). Also accepting it as a donation (willing to pay shipping costs) Really really need it for the further development. If you have one, please help.
I genuinely need at least one more cartridge, or that bendy part with the nozzle on it. But i suppose i could do with chips reading ABS. I live in south florida.
Lots of measurements and then throwing them all away when they don't work and just trying some numbers. (It's what I like to call "engineering"!) I measured the thickness of my glass plate to be around 1.7mm, and the gap is consistently -3.2mm. I manually set it to around -1.5mm, but that was too close to the bed and it clogged up my cart! I backed it off another 0.4mm and now I'm getting better and more consistent flatness than with the regular bed.
