Sand blasted glass, no tape

When I bought my fully assembled printer it came with PET tape on the glass (Borosilicate glass). I even bought a spare to use with tape on it.

Well it did not take me long to destroy the tape on both glasses. While it would be easy enough to get some more tape, I had seen a video about a person that sandblasted a glass (very rough sandblast like used to remove rust from car parts) and he then used a light slurry of acetone and ABS to coat it and it worked.

Well I also sand carve glass (awards, glasses for weddings and such) so I decided to try it (I could always put the tape over it if it failed). Well it worked better than I expected except that the print bed has to be very level and the head must be height adjusted just right. I imagine the tapes give you a little buffer in this area. It's been working really good with no parts coming loose.

Well I printed an assembly jig for my uncle, I had been printing smaller items and re-coating the glass after each (most of the coating right under the item gets pulled off with the item printed) and this meant that around the center I have kind of a build up or thicker layer of ABS/acetone mix. Well when I printer the larger jig, it took me 45 minutes to get it off the glass. I used pliers, screwdrivers, a hammer and finally heated the plate back up on the machine, then hit the back of it with a torch until the ABS softened a but and I was finally able to pry it off. I can vouch for Borosilicate glass's strength, this sheet is very thin and I though for sure I would break it before I was done. Plain glass would have broken at the very beginning.

Anyway the sandblasting works great. Just need to clean the gunk off the class occasionally. This was easy, just pour a bit of acetone on it and scrape off with blade from razor knife then wipe with acetone and paper towel. Re coat with ABS slurry and print.

After I built my Mendel90, I went in search of a cheap piece of glass to use as the print bed. I was already using window glass on the Huxley fairly successfully so was thinking of doing something similar. I visited several of the charity shops in our village, browsing through the second-hand picture frames looking for something suitable. The largest one I found had a sheet of anti-reflective glass which has a very fine textured finished and was big enough to make new beds for both the Mendel90 and the Huxley.
I was a little unsure whether it would be any good for printing but it is undoubtedly the best surface I have tried.
I never need any tape. I wipe it clean occasionally with window cleaner.
PLA sticks firmly at 60 C and just falls off the bed at 30 C - no force required.
I do sometimes use ABS juice for printing larger models in ABS, just to stop the corners from lifting.
I'm pretty sure that the fine texture on the anti-reflective glass is giving something for the plastic to key into at a microscopic level, but not so much that it won't release the part as it shrinks when it cools.



richgain

I'm pretty sure that the fine texture on the anti-reflective glass is giving something for the plastic to key into at a microscopic level, but not so much that it won't release the part as it shrinks when it cools.

Exactly. Now my sand blasting is more coarse than your anti glare but it's less so than coarse sandblasting used to clean rust off car parts.

There is also a cream etchant (EtchAll) that gives an even finer etch into glass that's easy to use. You simply paint it on, wait 10-15 minutes and wash it off with water. I bet this would work as well, at least to some degree.

Heated PET tape on glass works so well and it's easy to find that I don't see the point in going through all this trouble to avoid using PET. The PET tape on my printers have given me several months of hassle-free printing without changing them. Parts stick during printing and comes right off upon cooling.

Well for me, I managed to destroy the only 2 pieces of PET tape I had within a week, I could not get it locally and would have to place a minimum order with my normal tape place so that's why I tried the blasting. I had nothing to lose.

While I have a good supplier (except for the minimum orders) I would like to know where people get their PET tape.

Are you sure that's PET and not Kapton? PET tape is usually green and it is thicker than Kapton tape which is usually amber. It should last you a very long time unless you sand it. I highly recommend using the PET tape as is. If you use it as is, extruded filament will stick when the bed is heated to a surface temperature of 110 C and then parts it will release upon cooling. If you sand it down as Prusajr instructed in one of his videos, then extruded filament might not stick unless you coat with dilute ABS in acetone. The problem with this is that it can end up stuck to good that it won't come off even when cooled. Then you can end up ripping the PET in the process of removing the part.

You can order 2" wide 3M tape from Lulzbot.com (http://www.lulzbot.com/en/72-3m-pet-tape.html) or directly from 3M (http://www.shop3m.com/70006431145.html?WT.z_bynt=1). You can also buy generic PET tape from the Chinese discount web shops.

Well I was told my tape was PET, it was clear (very, very slight blue tint), I destroyed it by accidentally dragging the head into it while hot (ripped it). I was trying to get the Z access just right.

My tape was 200mm wide (same as the glass). so it was 1 piece not multiple strips no that it matters.

I like the 2" wide tape since it's easy to replace a section if it rips.

I adjust the bed clearance with a sheet of printer paper which is typically around 0.1" thick between the bed and the nozzle. I use it as a feeler gauge. This protects the bed surface from the nozzle. After homing the z, pull on the paper. If you can't pull it, then the nozzle is too low. If you can pull freely without any resistance, then it's too high. If you can feel some resistance, then it's just right. When leveling the bed, raise the nozzle by 1-10mm before moving to another spot.

I just use picture frame glass heated to 60C PLA sticks perfectly every time, I wipe it before each print with methylated spirits. No tape to rip no sanding required, as soon as the glass cools prints pop off.

I don't understand why people do all this crazy stuff when PLA sticks just fine to glass

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

I don't understand why people do all this crazy stuff when PLA sticks just fine to glass

Well I am using ABS exclusively right now.

NelsonRap Wrote:
-------------------------------------------------------
> I just use picture frame glass heated to 60C PLA
> sticks perfectly every time, I wipe it before each
> print with methylated spirits. No tape to rip no
> sanding required, as soon as the glass cools
> prints pop off.
>
> I don't understand why people do all this crazy
> stuff when PLA sticks just fine to glass

I could not get my yellow PLA to stick to 60C glass using methylated spirits or acetone or anything else. I was only printing small parts. The solution for me was blue tape but I would prefer sandblasted glass if it worked.

Quote
Gibbedy
I could not get my yellow PLA to stick to 60C glass using methylated spirits or acetone or anything else. I was only printing small parts. The solution for me was blue tape but I would prefer sandblasted glass if it worked.

Have you tried raising your bed temperature a bit? If you have your thermistor mounted under the PCB heat bed, your reading will be higher than actual surface temperature. When I print PLA, I set my bed to 70 for the first layer and then lower it to 60 for the rest.

Anyway, this thread is about ABS. So, we should go back to the topic.

I dont want to hijack this "ABS" thread but for PLA use hairspray. I've been using this for a while with excellent results!

Check out my blog for more info: http://gearforge.blogspot.ca/2012/10/wont-you-stick-to-my-cold-bed-please.html

What is funny is that, when dry, it actually makes the surface finish of my print bed looks similar to richgain's anti-reflective glass, but in this case, the texture probably doesnt have much to do with it.

Edited 2 time(s). Last edit at 10/30/2012 09:33AM by thecrazy.

@Gibbedy - where were you measuring 60C? On top of the glass or directly on the heated bed?

To be honest I slap some kapton directly on to a MK1 and print right on that. Sticks every time. If I ever get away from this 'bad habit' which will probably eventually destroy the MK1 (some of the resist has chipped off, revealing shiny copper beneath ) I'll probably got the hairspray route - easy, especially for that cheap chinese PLA.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

You'd be better off with PET instead of Kapton. Anyway, the MK1 is so thin that it's difficult to make it flat. This is probably why your nozzle is running into it. Having glass on it is so much better.

Actually, nozzle crashes are very rare. When I replace the tape more and more of the resist peels/flakes off... Probably due to inferior manufacture.

I don't really bother with 'proper' first layer heights anymore - I just adjust the Z-screws manually at the start of the print to get a good height.

But I'm still chewing my way through my first roll of kapton/polyamide. Probably because I replace it very infrequently.

Edited 1 time(s). Last edit at 10/30/2012 11:15AM by akhlut.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

Well, I just found a reason to NOT print directly onto glass. I printed a design and the glass was sitting on my desk cooling off and I heard a PING sound, I though it was the plastic part popping off the glass but the glass had a big chip broken out of it and it was firmly attached to the part (see photo).

I guess I will order some PET to use.

tmorris9: that was regular glass right? you wont have this kind of problem with borrosilicate glass

I've had pretty variable results with PLA directly on bare glass. Ultimachine PLA works perfectly for me pretty much every time so long as my leveling + height and bed temperature are perfect, plus the glass is spotless.

I started using some PLA I obtained a few months ago and not only would it not stick but it seemed to have an oil mixed in that required much more Windex to clean the glass, and even when everything was optimal it would try very hard to not stick, to the point where I would see bubbles on the bottom surface of the severely curled print where the first layer would just sort of repel from it.

A diluted PVA solution seems to work amazingly well. I'm still playing with the ratios to balance drying time, hold, and print removal, but using a bit of a premixed batch of 30ml of PVA glue (regular Elmer's) + 120ml of distilled water has produced really great results. I'm able to heat the bed to a lower temperature than the values that worked for me before (60C vs 85C, as measured by the thermostat under the PCB heater + bare smooth borosilicate glass, which helps tremendously with bridging at low heights), print my first layer at 60mm/s instead of 30, and this is with the cheap PLA. The bottom is still nice and smooth, unlike printing on blue painter's tape, but with a slightly more matte finish instead of supergloss. Seems to last a good while, too, went through about 60m of 3mm filament over the course of two weeks in producing a variety of prints before they started lifting and curling like the bad old days. I haven't tried it with Ultimaker PLA yet but anything that helps me print in a more vendor-agnostic fashion is very welcome.

thecrazy Wrote:
-------------------------------------------------------
> tmorris9: that was regular glass right? you wont
> have this kind of problem with borrosilicate glass

It's Borosilicate glass.

The only thing I can think of is that sand blasting it may have weakened it. The chunk that came off was stuck to the plastic part (the rest of the part popped off by itself) and I had to work to break of the piece stick to the glass.

tmorris9 Wrote:
-------------------------------------------------------
> Well, I just found a reason to NOT print directly
> onto glass. I printed a design and the glass was
> sitting on my desk cooling off and I heard a PING
> sound

If I read that correctly, you took the hot glass off the print bed and put it on the desk to cool off.
If that is the case, I would expect some significant thermal stress on the glass. Maybe that contributed to the chip forming.

@Mazaw
Since writing my original contribution I have been trying out some cheap Chinese PLA which has proved much more difficult to print successfully with. In the end, I resorted to blue painters tape which was all I had to hand.


I'm interested in the PVA method. How do you apply it to the bed, and how thickly?
richgain

Edited 1 time(s). Last edit at 10/31/2012 11:46AM by richgain.

Quote
richgain
If I read that correctly, you took the hot glass off the print bed and put it on the desk to cool off.
If that is the case, I would expect some significant thermal stress on the glass. Maybe that contributed to the chip forming.

I've done this many times with window glass and I never had one crack. I've also placed the hot glass on an aluminum plate to cool it down faster without a problem.

Also, tmorris9 was using borosilicate glass which is better at handling thermal stress.

OK, that's good to know.
My glass is taped to the bed with Kapton around the edges. I've never removed it.

richgain Wrote:
-------------------------------------------------------
> I'm interested in the PVA method. How do you apply
> it to the bed, and how thickly?

After mixing the solution in a water bottle, I keep it nearby. When I'm ready to apply, first I clean the glass as sparkling as possible, then shake the bottle and give it about 2 minutes for the bubbles to settle.

Then I pour a bit at the center of my glass build surface, to a spread of about 6 cm (I wish I use a syringe or something to measure exactly how much goes on there). Then I use a camel hair brush and paint outwards from that circle into a square with the edge about 1cm back of my glass edge. Even at the 1:4 ratio I use, it's still somewhat thick, so the brush strokes will leave trails behind but I find it will level out on it's own. Sometimes I might see a bubble or two, but they seem to pop on their own and I ignore them.

I try to keep my machine as level as possible with respect to earth with the help of some leveling feet were equipped on my vertex foot pieces*, as well as with respect to it's own axes, but after about 20 minutes I can see it does seem to want to flow to a particular corner. Maybe at my layer height (0.2mm) it's within tolerance and I've never had issues, but after it dries you may wish to re-level the bed with respect to the axes.

Drying takes a long time for me, perhaps because of my high-humidity location, but when it comes time to reapply the fluid I kind of write off being able to use the printer until 6 or so hours later. I haven't tried using a fan to help it along, or keeping the heat on, which I suspect would help it dry faster.


* They're cast urethane ones I got off ebay specifically for leveling help, but they're somewhat poor dimensionally and needed lots of shaving and teasing to get the rods through. In addition to that feature, they're intended for a "bling" type of setup where you can pop 5mm LEDs into the back of them and make them glow through. I wanted to hook them up to my X and Y step signals, but I figured I need to get my machine in order before I worry about Blinkenlights and now that I'm printing reasonably well I don't want to mess with a winning formula. They can be found here. I've forked the Prusa github with the intent of adding vertex leveling to the existing feet, but since I was working in Slic3r 0.8.3 I stopped because I was fudging my numbers too much and I later learned of the dimensional issues which explained my problems. If you want to check them out, they're here: [github.com] , changed frame-vertex.scad and frame-foot-level.scad to use an M4 bolt and added the M4 dimensions in metric.scad , but now that the i3 is out and takes a radical departure from the Mendel frame, I don't think I'm going to put out a pull request so my motivation to make it nice and pretty is kind of dead. I'm trying to make a second printer with 100% personally printed parts and I'll definitely want to finish this before then.

The one thing I've found with dilute PVA is that you don't want to skimp on it when you apply it.
I follow a similar procedure to the above pour some in the middle, spread it out with a brush, I usually do it with the heated bed on, so it dries a bit faster.
I'm not as enamored as some are with it as a bonding agent, in fact I'm not sure it's a chemical bond at all, I suspect it provides a rough layer giving the PLA something to grip onto.
IME the pieces pop off much like PLA on glass when everything cools. This is very different to the dilute acetone/ABS mix with ABS which is definitely a chemical bond and even tiny amounts of it will adhere the print extremely tightly to the bed even after cooling.
I use dilute PLA for my repraper.com PLA which seems to resist adhering to the same clean glass that my Ultimachine filament has no issues with.

I have been using a sanded (grain 80) ikea mirror for months. I use a heated bed but my cheap just would not stick without tape. No acetone wiping worked. Sanding it made my super cheap pla stick but detaches after the bed cools. On the other hand high(er) end pla is extremely hard to remove (knifes and force needed). I have not printed on an unsanded mirror on the latter pla

So a solution for your problem might vary, hence there are no one size fits all solutions here.

for PLA im using pva on a mirror tile.

From clean I brush dillute 50/50 pva glue on the bed very thinly and allow to fully dry, and repeat maybe 3 or 4 times, when i notice it getting shiny or thin I reapply. I level the bed with everything at print temps to a paper thickness at 0 holding the paper and at .1 allowing paper to move, then use z offset to set my initial layer for nice prints. As I apply more and more pva over time i notice I either need to clean the glass and start again or increase my z offset