How to use a heated bed

Hi everyone,
I'm trying to use a heated bed to print PLA and I'm having a lot of trouble getting it to stick. I received some advice in this thread but it was pretty far off topic so I created this thread for this specific issue. Here's my current understanding:

1. Some people claim to get PLA to stick to a heated bed at 65 degrees celsius, but others need to heat it as high as 120 degrees.
2. It's important to clean the bed surface in between prints with denatured alcohol (methylated spirits) or ammonia.
3. The best surface for this purpose is removable glass sheets, 2nd best is kapton tape.
4. It's best to minimize draft near the printer, mainly to help the surface stay hot.
5. Once you follow these steps, your prints will be stuck so hard they will be difficult to remove.

The problem is, even following all these tips (except I've got my bed at 115 degrees, it doesn't quite make it to 120) I have a very low success rate, maybe around 20%. Even when I do get successful prints most of them aren't even stuck at the end of the print, they were just successful because I got lucky that the print didn't slide around. Since I'm following %90 of the formula, I would think my print would at least stick, even if not very hard.
Can anyone think of anything I'm forgetting? I am using glass sheets and cleaning them with denatured alcohol. Anyone else having this issue and can't understand why? I've gone back to printing an double-sided tape from MakerGear for now since I have a 100% success rate this way. In fact, I'm starting to wonder why everyone messes around with these heated beds when tape works perfectly well. I'm thinking I'll make removable corrugated plastic plates with double-sided tape on them.
Thanks for your time.

First off, for PLA on heated glass, you need to keep it at or below 80C. Going higher will cause it not to stick. Going cold on glass also will not stick.

The first layer needs to be printed quite thin. less than 0.3mm, and go slow. At least half speed.

If you have an aluminum print bed, you need kapton, or blue tape on there.

If you have a Red PCB heat bed, you can put kapton on the solid copper side and print on that.

You can also print PLA to cold acrylic covered in kapton.

For ABS, you can use the Red PCB @120C with bare glass, at any speed. The ABS extrusion temperature should be around 250C. When the glass cools, the part will already be free. Same for PLA on bare glass.

One more thing. I use Pyrex Glass AKA Boro-Silicate Glass, so your results might varry with other types.

Ahh, yes. I forgot to mention printing the first layer slow and solid. I am doing that as well. I've been testing at a 0.2 ratio, normal printing speed is 200 mm/s.I have tried a few different temps ranging from 65 to 115 (getting different answers from different folks). I will try keeping it at 75 the whole time to see if that works. Thanks for the reply. I'm not sure what type of glass I'm using, it's from the Michael's framing department.

I just use standard picture frame glass because I had some, PLA is sticking very well at 60C and I do clean it with Meths each print run. I tried tape and found it pulled off in places warping my prints. I print at a relatively slow 30mm/s and do the first layer at half that but keep my layer height at 0.4mm.

Are you using Raft if using Skeinforge because that gave me sticking problems so turned it off

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

200mm/s feed rate is huge! Bring that way down.

Quote

For ABS, you can use the Red PCB @120C with bare glass, at any speed.

ABS does not stick to glass for me.

Quote

The ABS extrusion temperature should be around 250C. When the glass cools, the part will already be free. Same for PLA on bare glass.

PLA does not auto release from glass for me.

Quote

One more thing. I use Pyrex Glass AKA Boro-Silicate Glass, so your results might varry with other types.

Evidently the type of glass must make a huge difference. Where do you buy Pyrex from?

---

[www.hydraraptor.blogspot.com]

jcabrer Wrote:
-------------------------------------------------------
> 200mm/s feed rate is huge! Bring that way down.


At a 0.2 ratio my first layer was printing at 40mm/s which, from the sounds of it, is too fast for the first layer, but are you suggesting that I need to bring down my normal feed rate as well? Why would that make a difference?

I find once I have removed the glass and let both it and the part get completely cold the parts just lift off

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

Quote
jcabrer
The first layer needs to be printed quite thin. less than 0.3mm, and go slow. At least half speed.

I used to print the first layer at the same height as the layer thickness, 0.3 and 0,4 mm, on heated bed between 10-20 mm/s and it worked fine for me.

Quote
nophead

Quote

The ABS extrusion temperature should be around 250C. When the glass cools, the part will already be free. Same for PLA on bare glass.

PLA does not auto release from glass for me.

It worked all the time for me. I wished there was a similar surface for ABS. It sounds like I should try pyrex too.

I bought my Pyrex from McMaster-Carr. Its not cheap at $27 for 8x8 inch, but it was worth it.

Ok, I tried a print at 75 degrees. First layer at 10 mm/s, the rest at 100 mm/s. Cleaned the glass just before the print, got a solid first layer. Window closed. My layers are 0.37 mm and my nozzle is 0.4 mm. I didn't try a smaller first layer because I couldn't find the setting in skeinforge. It started sliding around loose mid-print.
I've attached the stl I'm trying to print.
Here's a picture so you can see what the bottom looks like. Help!

How about including the gcode that you are printing.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

I have only ever used natural PLA. Perhaps the black stuff doesn't stick to glass. It looks very different. I have found some types of black ABS don't stick as well as natural ABS to PET tape.

---

[www.hydraraptor.blogspot.com]

I've printed black PLA from Ultimachine succesfully on heated glass with my bed set around 70C.

Quote
brnrd
I've printed black PLA from Ultimachine succesfully on heated glass with my bed set around 70C.

No way! That's the exact same PLA I'm using! What kind of glass are you using? Like I said I'm just using standard craft store picture frame glass. I guess about 3mm, but I haven't measured. Also, what is your normal printing speed? I'm going to try and get my setup as close as possible to yours.

Quote
brnrd
I used to print the first layer at the same height as the layer thickness, 0.3 and 0,4 mm, on heated bed between 10-20 mm/s and it worked fine for me.

What do you mean "0.3 and 0,4 mm"? You printed at both thicknesses at different times with success?

I am planning on using a heated bed with glass and from a bit of reading it seems that most of the picture frame glass are temepered, perhaps this has something to do with why your prints arent sticking? Perhaps find a local glass shop and ask for a small piece of each, tempered and none tempered and see if that is the case. Would be interesting to see.

Mines definitely not tempered as I do framing and never buy tempered glass, I have used so far black, purple, clear, red dark blue and dark green PLA from Vik here in NZ and they all stick fine

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

Just called Michaels and they said my glass is tempered. cubevertex_export.gcode.zip is the gcode used to print the piece in the previous image. Had to zip it to bring it down under the allowed max size.

I print on Kaptan covered aluminum at 65 degrees. I just finished printing 21 sets of Prusa bushings without a miss. Apart from temperature I use exactly the same settings that I use for ABS. When switching from ABS to PLA, if I don't scrub the Kaptan with ABS cleaner and then a dry paper towel, and I mean REALLY scrub, it won't stick at all. Leads me to wonder if maybe your cleaner or applicator might be leaving a slight film on the surface. It doesn't take much.
I noticed in your picture that you have some weird whitish thing going on between extrusions. Not sure what that is but it doesn't look right to me. It does however highlight gaps.... which shouldn't be there.

Just a note from my own experience. I can't print .37 layers with my .4 nozzle, .35 is the max for decent prints and it's much happier printing at .33

unicoder Wrote:
-------------------------------------------------------
> > I've printed black PLA from Ultimachine
> succesfully on heated glass with my bed set around
> 70C.
>
>
> No way! That's the exact same PLA I'm using! What
> kind of glass are you using? Like I said I'm just
> using standard craft store picture frame glass. I
> guess about 3mm, but I haven't measured. Also,
> what is your normal printing speed? I'm going to
> try and get my setup as close as possible to
> yours.
>

I bought window glass of similar thickness from a stained glass shop.

>
> I used to print the first layer at the same height
> as the layer thickness, 0.3 and 0,4 mm, on heated
> bed between 10-20 mm/s and it worked fine for me.
>
>
>
> What do you mean "0.3 and 0,4 mm"? You printed at
> both thicknesses at different times with success?

Yes, I printed them at different times. Skeinforge doesn't have the option to slice at multiple layer thickness.

unicoder Wrote:
-------------------------------------------------------
> Just called Michaels and they said my glass is
> tempered. is the gcode used to print the piece in
> the previous image. Had to zip it to bring it down
> under the allowed max size.


Have you tried washing it with dishwashing detergent to remove oil and grease from the surface?

As far as speed, I was printing the first layer at around 10 mm/s and the rest at 20-30 mm/s. I think 100 mm/s is too fast and leads to segment pause anyway.

So, you are carving at 0.37 mm and that is the height of the first layer which means that when your Z MIN switch activates you must be very close to touching the bed.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

I have switched to a 0.33 mm layer and that does seem to print better. I've never printed with ABS but I gave my glass a really good scrub first with dish detergent, then with denatured alcohol and wiped it dry with a paper towel. I set up skeinforge to generate gcode to print at 20 mm/s, with the first layer at 10 mm/s. I heated up my bed to 75 degrees and tried again, first layer looked good, but once again the print became detached part way through. I would be so so happy if I could get a print to stick so hard that I had the break the glass. I don't understand, it sounds like everyone else is having the opposite problem from me. Is there no one else who has trouble getting objects to stick to glass? Is anyone here successfully using tempered glass?

Quote
Bruce
It does however highlight gaps.... which shouldn't be there.

How do I get rid of the gaps, increase the object first layer infill flow rate?

I saw this on Thingiverse

Quote
http://www.thingiverse.com/thing:10227
I needed to be able to adjust it very small amounts to get the z-height just right to get a nice first layer that wasn't too hard to peel off, but still stuck.

This is the first I have heard of this. Where is the goldilocks zone for the start nozzle height? How can I tell it's at just the right height without a specialized measurement device?

BTW, Bruce, that unidentifiable "whitish thing" is probably from these guys

Since the first layer extrudes the same amount of plastic as any other layer the correct hight of the nozzle is the layer height. The only exception to this is when the bed is absorbent, e.g. PLA on masking tape and you want force it soak in a little. If is too low there is nowhere for the plastic to go and the nozzle will end up ploughing through plastic and the object will have a ridge around the bottom. If it is too high there will be gaps in the first layer if it manages to stick.

I calibrate it by rolling an 8mm rod under the nozzle until it just touches and the I know the bed is 8mm lower. Since it has to be correct to about 0.05mm I use a digital calliper to measure the rod. Mine is actually 7.95mm.

Two essential items of equipment to calibrate a machine are a multimeter with a thermocouple probe and a digital calliper.

I don't know if the glass I tried is tempered or not. One piece was from a kitchen scale, I would guess that probably was and the other was a sheet of 3mm glass from a glass shop. I expect that wasn't.

---

[www.hydraraptor.blogspot.com]

Quote
unicoder
Just called Michaels and they said my glass is tempered. cubevertex_export.gcode.zip is the gcode used to print the piece in the previous image. Had to zip it to bring it down under the allowed max size.

Are you sure it's tempered glass. Did they cut it to size for you? If they cut it with a diamond cutting tool, then it's just plain glass. I understand that you can't cut tempered glass with that tool. What I have is just plain window glass that the stained glass shop cut to size for me.

The wiki on tempered glass says the following:

Quote

However, the toughened glass surface is not as hard as annealed glass and is therefore somewhat more susceptible to scratching. To prevent this, toughened glass manufacturers may apply various coatings and/or laminates[which?] to the surface of the glass.[[/quote]

I'm using these 3M transparencies with thin 3M 2way tape to stick it to the aluminum bed for printing PLA, I compared performance of the transparencies, Kapton and Blue tape with PLA and the transparencies are king. (23oC ~65oC) sticks really good and super durable.

I wipe the table down with acetone before I print.

Quote

BTW, Bruce, that unidentifiable "whitish thing" is probably from these guys

Cute pooch's....... Laying there so contentedly, while you get frustrated out of your mind because your prints won't stick!!!!!

Anyway.....

CVBummer.jpg

I noticed some inconsistencies in the amount of fill that's being laid down.

CVBummer2.jpg

Maybe a bit of filament slip??? But even so, it still should have stuck.

Since all the glass plate people on the thread are having success using different types of glass, chances are the glass is not even the problem. Do have a means of monitoring the plate temp when the machine is running? Maybe the heater is cutting out. Do you an alternate means of reading the temperature?

I use one of these

CTCthermometer.jpg

We have 5 different machines and they show 5 different temperatures when set the same, so I use the gun and adjust the software settings to get what I want.

As for adjusting flow rates. I don't know. I'm old school. We use an older version of RepRap Host and firmware with acceleration turned off, and Skienforge 39. In the old school way of doing things: If my prints are sparse or thick as happens when changing to a different diameter filament I adjust the E steps in the firmware to compensate. I wouldn't know how to relate to the new stuff you people are using.
Like nophead wrote, if your Z homing is adjusted right then the the first layer should go down the same as any other. So if Z height is right and the first layer is sparse, then all the subsequent layers will be sparse as well. If the rest of the layers have good fill and the first layer is sparse then Z is most likely zeroing a little too high.
Speed shouldn't have much to do with it either. I print first layers at 57mm/sec and subsequent at 72..... that's without acceleration.

Maybe nozzle temperature? What are you using?

Edited 1 time(s). Last edit at 12/19/2011 09:41PM by Bruce.

That's not unusual, and does not mean it's not sticking. It's more of a nonuniformity. Filament cross section is round, and is layed down that way. It may be that your bed is not level, or that the x axis is moving up and down.

Good news!
I have printed 3 things in a row that have stuck very well. The only things I changed between my last failed test and these successfully runs were the following:
1. I got my nozzle a tiny bit closer to the bed, so now the filament is getting smashed pretty flat on the first layer rather than just going down as a round tube.
2. I'm using a transparency, attached directly to the heated bed with double-sided tape and heated to 65 degrees.
3. I sped up the first layer to 20mm/s normal feed rate is at 200mm/s

I wish I could have been more scientific and changed only one variable at a time, but there are way too many variables so that would've taken ages. Since I have this working so well, I don't know if I want to go back and try glass any time soon. The print has a perfect amount of stick to the transparency so I don't have to work very hard to get it off. Thanks for the tip!
Btw jcabrer, no my bed is not perfectly level, I'm still working on that but it sticks fine to a transparency without a level bed, so why not glass? I think something must be wrong with my glass. I scrubbed it so well though, I just don't get it. I couldn't get anything to stick to heated kapton either. Oh well, I'm doing very well with transparencies now.

couldnt stick PLA to heated kapton???? wow. I had to try and remove the kapton to get parts of sometimes when I was using that. And even then it sometimes just teared around the part and I had to damage the part so I could get something under it to prise it off.