After the printer was finally assembled, I was left with a mysterious aluminium plate, that was originally covered in a thick film that I mistakenly peeled off, assuming it was merely protection for its polished surface.
Browsing the manufacturer’s page for hints to what the film actually was to replace it, I found the image to the left, where the plate can be seen working either as an unheated surface to be used instead of the heatbed; or a printing surface to be used on top of the heatbed (there seems to be a red reflection underneath it, suggesting there’s also a heatbed installed). In either case, I shouldn’t have peeled of the film covering it – that now I know it most likely was PET tape -, as it functioned as a adhesion enhancer between print surface and printed object.
While looking for a solution created by my lack of patience, I found a great deal of information that helped me better understand the workings of the heatbed assembly and printing surface material.
Heatbed assembly information was obtained by rummaging thought the assembly instructions of several iterations of RepRap printers. Listed in the table and image below this paragraph are the design variations found in each set-up instruction:
=====================insulat.==██==heat spr.===printing surface========== reprap.org MDF ██ aluminium glass reprappro.com cardboard ██ aluminium glass 2mm + kapton tape nextdayreprap.co.uk plywood ██ aluminium mirror + kapton tape reprapworld.com MDF ██ aluminium borosilicate glass 3mm + kapton(ABS)/ "sticker"(PLA)
While the use of an aluminium plate (which I already own) as a heat spreader and a glass as printing surface seem to be a consensus, the insulation choice vary a bit, so I decided to do a little more research.
While MDF thermal conductivity (TC) lies between 0.12 and 0.15, the plywood value is 0.12 and cardboard is slightly worse than the other two contestants, with 0.2.
Cork, I found, is a better insulator and thus won’t catch fire as easily. The thermal conductivity of a sheet of re-granulated cork goes as low as 0.044.
I also found a list with other possible materials that would work, but for the sake of simplicity I’ll just start with the cork and change it if needed.
Even though the topic was briefly discussed in each of the previously mentioned print designs, I felt I should learn more about each of them before I made my choice. As I did, it became clear that in today’s 3d printing matured culture there’s a myriad of printing surface materials to chose from, each with it’s own properties and adequacies and I felt that to choosing I’d have to know a bit more about each of them.
Two good articles (richrap.blogspot.com | reprap.org) tackle this exact issue extensively, explaining what base materials are available and which ones work best for each printing material. Notice that some materials are meant to be used in direct contact with the heatbed presented in the form of sheets and boards (above, called base materials); while others are overlaying films meant to be applied atop of the the rigid material (above, soft materials), usually glass, as a thin film. I compiled the information from the articles in the tables bellow, for clarity:
name best for comment ------------------------------------------------------------------------------ Acrylic (10mm) PLA, ABS The hot nozzle can melt this surface! It can happen if not well leveled! Heat is mandatory for pieces wider than 100mm/4″. Glass/mirror PLA Only small pieces (less than 170mm/6.7″ of width) will stick. The print surface must be really well levelled and clean. Mirror helps check underneath first layer for problems. Cellulose based¹ Nylon Massive warp forces of Nylon can bend soft materials, though. Perforated board² ABS Good heat conductor. By printing a first 'raft' layer before the actual object, the ABS will melt into the holes and grip to the material. Works well for objects under 150mm/6″ of width. PEI sheet All! Expensive but very versatile, easy to use, durable and maintenance free. ¹wood, paper, compressed cardboard or impregnated cotton sheets ²only FR4 and FR5! These are glass-reinforced epoxy laminate sheets.
name best for comment ------------------------------------------------------------------------------ Blue painter tape PLA, Nylon Ideal for small or tiny highly detailed pieces. Rapidly cooling the material during the print improves quality. Not great for nylon, but it works. PVA/glue washes PLA, PET Example: 1 part PVA, 8 parts water. Apply over glass. Heat can be used after it's hardened. Gluestick Apply on print surface while cold and leave it to dry. Heating it for 30 min at 60°C improves success rate. Hairspray PLA, PET Chosen hairspray must contain 'extra hold' ingredients such as vinyl, acetate and copolymer. ABS slurry ABS ABS dissolved in acetone wiped gently on the surface with a piece of paper. Mix acetone and ABS in a ratio of 240ml/8g (8fl oz/0.15oz). See easier calculation. Kapton tape ABS, PLA A polyimide film that withstands temperatures up to 400°C. PET tape PLA A polyester film that can only withstand temperatures up to 180°C, making it less versatile than Kapton tape. On the other hand it's less expensive.
The PEI sheet look like the great choice, even considering the price. Unfortunately in Brazil it’s pretty difficult to get if you’re not buying at industrial scale, so I’ll have to settle for something else for now.
Balancing all the stuff I read and the things I have available, I picked the following set-up: cork as insulator, mirror as a base printing surface and – since I already own a 50mm roll of it and intend to start with just ABS and PLA -, kapton tape as the less messy (than glue, hairspray and slurry) and more versatile (than blue painter tape, PET tape or even cellulose or perforated boards) overlay printing material.
Here are the added costs so far (conversion rate 02/19/16):
name======================================local=======U.S.== Re-granulated cork sheet (45x60cm - 2mm) R$23,00 $5.69 4 20x20cm mirrors R$29,90 $7.39 -> 1 + 3 spare mirrors! Kapton tape (50mmx33m) R$48,00 $11.86 12 foldback spring clips R$5,24 $1.33 -> 4 + 8 spare -------- ------- totals: R$106,14 $26.27
This is how each of the materials chosen above came together to form the heatbed assembly:
On the first image, notice:
- That the cork sheet is attached to the heatbed with Kapton tape by the sides.
- Since the extra layer of cork made the springs – that hold the heatbed assembly up when in place – too compressed, a small portion of cork was removed from each corner of the sheet to allow the four springs to press against the heatbed directly.
On the second image, see the aluminium heat spreader getting fitted over the heatbed. Notice that an adaptation (a cutout) was made in order to prevent the aluminium sheet to touch the soldered terminals. That is a good idea for two reasons: the solder could melt if put in direct contact with heat; and the lumps formed by the solder would otherwise prevent the proper surface-to-surface contact between heat spreader and heatbed.
The third image shows the mirror surface installed, without the tape. Unfortunately it’s not wide enough to cover all the available heat spreader surface, but I think it’s good enough for now.
The fourth and fifth images show the kapton tape applied to the mirror surface. The soapy water technique (learned here) made the process very, very easy!
(…) the current standard (as of January 2015) when the PCB heatbed is used with a piece of glass as the build surface is to mount the PCB heatbed with the traces up. This improves heat transfer to the glass and the PCB heats up faster.
Meaning: if you the printer you’re assembling lack a printing surface above the heat bed, keep the smooth side up! If it doesn’t and you decide to orient the heating element up, be careful with built-in thermistors as they’re almost imperceptible but will cause an array of problems (see next post).