Glass sheet

Hi people,

Does anyone know where I can purchase large sheets of glass (say 1200mm x 700mm)? Ideally borosilicate as it needs to withstand temperatures up to 150 deg-C plus. If you can suggest alternative glass types then that would be quite helpful too and where I can get the glass in sheet too.

Thank you

---

The (getting better) Noob

Zapadee zoopidee.

Why not contact your local glass dealer? They must be able to help you further.

---

http://www.marinusdebeer.nl/

There are no local glass dealers which provide custom made borosilicate glass sheets here in France. Most are in USA. I just want some good suppliers.

Thank you

---

The (getting better) Noob

Zapadee zoopidee.

Temperature is only a problem for ordinary float glass if you heat or cool it quickly and unevenly. If the glass is for a 3D printer heat bed, and you put it on an aluminium heated bed, then as long as the heating is even and/or aluminium is thick enough to spread the heat evenly, ordinary float glass should suffice - and it will cost less and be much easier to obtain than borosilicate glass. Just don't spill water on it while it is hot!

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I agree on the sentiment. I put in a 3mm alu heat spreader on my glass with thermal paste interfacing the glass and the heater. After 300+ hours of printing it is showing no signs of fatigue. Although the hardest I have pushed it was 120c.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Hello 3DPrintingNoob,

we can offer you borosilicate glass at you specifications.

please contact us via our contact form.

www.filafarm.com

Edited 2 time(s). Last edit at 03/18/2015 12:13PM by Filafarm.

---

Onlinestore: www.filafarm.com & Galerie

As glass is the most widespread building material of the 20th century, it's a fascinating topic.

I live in France too so i can help you with stuff.

France has a very long tradition of unbreakeable and advanced glass, at my grans house there is a black plate of unbreakeable glass from 1954 or 1939 or something. Arc International in the biggest euro manufacturer of pyrex goods. Glass is to french people what AA batteries is to japanese. it's great for chemistry and cooking.

this google search : verre borosilicate site: fr

this result:
SOUS FORME DE FEUILLES

Dimensions de 2300 X 1700 mm
Épaisseur de 1 à 60 mm


Please tell us your intentions for the glass, it's good when you are an inventor to know how to source strange materials locally and how to question professional companies so they give you informations... and i am abit curious why a noob would use a 2 meter big piece of heatproof glass?


[www.google.fr]

Edited 1 time(s). Last edit at 03/20/2015 05:18AM by skynetprinter.

Hello,

When you say aluminium plate underneath the glass surface, how is the aluminium heated? Nichrome wire? Heatbed attached beneath? Because if you put an aluminium plate then glass plate on top then wouldn't there be a lot of heat loss. i.e. the heatbed is at 100 deg-C but the glass wouldn't be. I'd say it would be at 80 deg-C at best.

Skynetprinter: Merci beaucoup de m'envoyer cet info! C'est très utile. The reason I would use the large heatproof glass is because I am undergoing a project right now where I need to design and manufacture a large-scale 3D printer. A large build plate is a necessary option in the design so I need to look into it.

Thank you!!!

---

The (getting better) Noob

Zapadee zoopidee.

I would recommend not using glass at all!!!

It will take away your budget from other elements of the printer... The best method with glass and ABS for example is taping PE based kapton tape to the glass.

Mais voyons, Glass is heavy it will strain the motors more and be less responsive.

Makerbot use PE and PET and HDPE and epoxy based substrates. In fact, i think you could take 2 kilos of epoxy resin and a fibreglass sheet and pour it yourself onto a flat surface with some glass fibre/carbon fibre and you will have a much better substrate than glass.

Alu is indeed cool, you can use 50 or 100 watt resistors directly to make the heatbed. count about 150 watts per A4 sheet of paper size. use arctic silver adhesive if you want to glue it directly to the aluminium instead of drilling.

sandblased glass is maybe ok, but PE is more sticky to ABS plastic because it is miscible with it, so you can adjust large prints to stick directly to the printer.

perhaps try some varied polyethylene based boards. Your project sounds interesting. this is cool too: boots industries printer. its' based on open source design.

You would heat the aluminium from underneath using a Kapton heater or silicone heater. There is a Chinese company on alibaba that will manufacture a custom silicone heater for you at low cost. If the bed is very large, use multiple heaters. Aluminium is a good conductor of heat so there will be negligible temperature drop between the underside and the top. The top of the glass will be a few degC lower than the aluminium.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

PS power resistors are a poor choice for a heated bed IMO because you won't get good heat distribution unless you use a very large number of them, which will cost more than a silicone heater. For a large format printer, the bed should be either fixed or move in the Z direction only, so the weight of the glass should not be a big issue.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

no plastic sticks to glass, only kapton tape does. for PLA i use paper adhesive which is awesome, the PLA sticks straight to the aluminium fairly well too. you don't need to use only one print bed too... it's a myth. you can have .2 mm gaps in the print substrate the print will be fine. you can use des bosses under the substrate and holes in the alu to keep all in place if you wished, the print would be the same, the main issue is perfect flatness of substrate and stickyness using paper or abs acetone mix.

Quote
skynetprinter
no plastic sticks to glass, only kapton tape does. for PLA i use paper adhesive which is awesome, the PLA sticks straight to the aluminium fairly well too. you don't need to use only one print bed too... it's a myth. you can have .2 mm gaps in the print substrate the print will be fine. you can use des bosses under the substrate and holes in the alu to keep all in place if you wished, the print would be the same, the main issue is perfect flatness of substrate and stickyness using paper or abs acetone mix.

I print PLA right on glass...care to explain that? The reason glass is used is because it is very flat, much flatter than any aluminum you are going to buy unless you want to have machine work done to it.

---

Please add your printer design to [reprap.org]

Quote
tjb1
I print PLA right on glass...care to explain that?

So do I.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Oh right sorry. I didn't try PLA with glass. I mean that no plastic glues are recommended for glass, silicone is. I still wouldn't see the logic in having a 700-1200mm piece of glass on a printer. to be safe it would have to fairly thick or in multiple pieces, and it couldnt print ABS, it would be heavy and less responsive to move. Alu is a great option for the heatbed anyway. it doesnt stick PLA that well enough for prints as far as i can tell.

I was trying for a week to print ABS on different surfaces, and i tried 2 kinds of glass directly at 100 degrees and adhesion was not good enough for a 4cm square base object. I have mentioned that HDPE should be ok as a substrate, but it isnt, i just tried to heat and glue hdpe and abs together, it didnt work half as well as with PET. so PET and FR4 are the only 2 polyethylenes tha ti have heard stick for abs.

Glass has some good properties as a 3D printer heat bed:

1. It is very flat.

2. It is very hard, so not easily damaged e.g. by a head crash or when you prise the print off it.

3. It Is very cheap to replace if you do damage it.

4. It is cheap enough to have multiple glass print beds, so that when a print completes, I can start a new one on a different bed without waiting for the first one to cool.

This last point obviously wouldn't apply to a very large bed because it would not be safe or easy to handle such large sheets of glass.

To print ABS on glass, the usual approach is to paint ABS slurry on it first.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

What kind of glass sticks very well to ABS to have very good success rates with even large parts of 10-20 cm?

Different kinds of glass have different adhesive properties with ABS.

I tried an Ikea mirror, which stuck well enough to do 4-5mm of ABS prior to warping,

Then i tried a picture frame glass, which didnt stick at all.

Both of these i tried multiple times with ABS juice. So it seems different glass has different adhesion, and it's not very clear which one is best.

Most people advocate kapton tape on glass, which is made of some kind of polyethylene.

Hello guys,

Thanks a lot for all this info. I have a small-scale printer for testing and ABS wouldn't stick to the glass surface even with Kapton tape. The bed temperature was at 100 deg-C and the glass was an oven window glass about 4 mm thick. It prints PLA perfectly but not ABS. The heatbed is attached directly beneath the glass print surface.

I am going to post a photo of the layout of heatbeds and print surface shortly. Let me know what you think of the setup please.

---

The (getting better) Noob

Zapadee zoopidee.

Okay so I attached the heatbed and print surface layout that I plan to follow (really simple):

The print surface is 1300 x 800 mm and 4 mm thick (could be glass or aluminium).

Under the print surface is a total area of 1200 x 600 mm of 300 x 300 mm heatbeds (so 8 beds total).

The beds will be silicone heat mats.

I want to connect up all the beds into a single circuit (the use of an SSR will most likely be necessary for connection to the board). I hope to attach the beds onto the bottom via an adhesive (most likely); however, which method of attachment, I don't know.

Let me know what you think please!

---

The (getting better) Noob

Zapadee zoopidee.

Even if you decide to use glass for the print surface, you will need an aluminium base to attach the silicone heaters to, and distribute the heat uniformly.

Using a single circuit should be OK if the heaters are identical and of equal power. The risk if that if the different heaters have significantly different resistance and therefore different power, then you will get uneven heating. Use of a thick aluminium heat spreader will help, but may not eliminate the problem.

Unless you know that the heaters have very low coefficient of resistance with temperature, the most thermally stable configuration is to connect the bed heaters in parallel. I suggest connecting the thermistors in parallel too, then if the temperatures differ between thermistors, the reading will be weighted towards the hotter ones, which is safer than the other way round.

With so much heating power, IMO mains voltage heating is the only practical solution. Get advice from a qualified electrician on insulation and protective grounding. If the bed will be moving, then you need to choose highly flexible high voltage cable (such as the cable used for multimeter test leads) and make sure it has adequate strain relief at the ends. For such a large printer, I suggest either a non-moving bed (e.g. delta configuration) or a bed moving only in the Z direction (e.g. CoreXY configuration).

Regarding ABS printing, as well as using ABS slurry to get good adhesion to the bed, it is very important to protect the printer from draughts. Some people use a draught shield around the printer. For the same reason, if the printer has a bed moving in the X or Y direction, you should not use a high speed for non-printing moves. Including a brim around the part also helps adhesion.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hello,

1. Why would I need the aluminium base to attach the silicone heaters? Wouldn't it just be enough to attach them to the glass? The glass would just conduct the heat from the beds wouldn't it?

2. I will be using the same heaters throughout (equal in power and resistance, etc.)

3. Would the beds need to have a low temperature coefficient of resistance?

4. I have to conduct an FMEA for the heatbeds configuration (i.e. in parallel and series). The fact that the thermistor won't record the average temperature but rather hottest temperature may not be enough. Would you suggest an alternative or is this the most efficient way? If this is the simplest and most efficient way then it's accepted.

5. The bed will only be moving in the Z direction so there shouldn't be a problem.

6. Is ABS slurry basically a mixture of ABS filament and acetone? How would I mix the ABS? Would I need to chop the filament into little pieces?

Thank you!

---

The (getting better) Noob

Zapadee zoopidee.

Quote
3DPrintingNoob
1. Why would I need the aluminium base to attach the silicone heaters? Wouldn't it just be enough to attach them to the glass? The glass would just conduct the heat from the beds wouldn't it?

Glass is a poor conductor of heat. A glass bed is thin enough to conduct heat reasonably well from one face to the other, but it won't conduct well in the plane of the bed. The silicone heaters won't produce heat uniformly over their surfaces, and they won't produce heat right at the edges, especially if you have small gaps between the heaters. So I strongly recommend a heat spreader. Maybe borosilicate glass would stand up to the uneven heating, maybe not. Also, if you were to break or damage the glass for whatever reason, you would be faced with somehow removing the silicone heaters and attaching them to a new piece of glass.

Quote
3DPrintingNoob
2. I will be using the same heaters throughout (equal in power and resistance, etc.)

They may be nominally of equal power, but do you know what the manufacturing tolerance on the resistance is?

Quote
3DPrintingNoob
3. Would the beds need to have a low temperature coefficient of resistance?

Not necessarily, if you connect them in parallel. With parallel connection, the hottest beds will take less power as they heat up. With series connection, the hottest beds will draw more power as they heat up, which is not a good thing even though thermal runaway is unlikely. But the heaters may have a low temperature coefficient of resistance anyway, depending on what the conductors are made from. Ask the supplier.

Quote
3DPrintingNoob
4. I have to conduct an FMEA for the heatbeds configuration (i.e. in parallel and series). The fact that the thermistor won't record the average temperature but rather hottest temperature may not be enough. Would you suggest an alternative or is this the most efficient way? If this is the simplest and most efficient way then it's accepted.

From an FMEA perspective, series connection of the heating elements has the advantage that if one bed heater goes open circuit, the whole bed shuts down. So you will definitely realise that something is wrong. Whereas with parallel connection, ideally you would monitor the thermistors individually so that you can detect when a heater has failed. An alternative would be to monitor the total current drawn by the heaters.

Quote
3DPrintingNoob
5. The bed will only be moving in the Z direction so there shouldn't be a problem.

That helps, but you still need to provide strain relief at both ends of the cable, and ensure that the cables can't chafe on anything. And I suggest you choose highly flexible cable.

Quote
3DPrintingNoob
6. Is ABS slurry basically a mixture of ABS filament and acetone? How would I mix the ABS? Would I need to chop the filament into little pieces?

Yes. You need only cut the filament into small enough pieces to fit in the bottle that you are mixing the slurry in. Or you can buy solvent pipe cement, which is much the same thing as ABS slurry.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
Glass is a poor conductor of heat. A glass bed is thin enough to conduct heat reasonably well from one face to the other, but it won't conduct well in the plane of the bed. The silicone heaters won't produce heat uniformly over their surfaces, and they won't produce heat right at the edges, especially if you have small gaps between the heaters. So I strongly recommend a heat spreader. Maybe borosilicate glass would stand up to the uneven heating, maybe not. Also, if you were to break or damage the glass for whatever reason, you would be faced with somehow removing the silicone heaters and attaching them to a new piece of glass.

1. Good point regarding the glass replacement. Quick question, how would I implement the heat spreader onto the build surface? I'm not entirely sure how it's used, all I know is that it is used to spread the heat evenly but how? No idea.

Quote
dc42
They may be nominally of equal power, but do you know what the manufacturing tolerance on the resistance is?

2. I will need to find out the manufacturing tolerance. How much of an effect would a small difference in resistance create?

Quote
dc42
Not necessarily, if you connect them in parallel. With parallel connection, the hottest beds will take less power as they heat up. With series connection, the hottest beds will draw more power as they heat up, which is not a good thing even though thermal runaway is unlikely. But the heaters may have a low temperature coefficient of resistance anyway, depending on what the conductors are made from. Ask the supplier.

3. Okay, I will ask the supplier. I think it's safest to use a parallel connection then.

Quote
dc42
From an FMEA perspective, series connection of the heating elements has the advantage that if one bed heater goes open circuit, the whole bed shuts down. So you will definitely realise that something is wrong. Whereas with parallel connection, ideally you would monitor the thermistors individually so that you can detect when a heater has failed. An alternative would be to monitor the total current drawn by the heaters.

4. Yes, it would be easy to detect a fault if series was used. The effort required to monitor each thermistor shouldn't be a problem considering there would only be about 8 beds.

Quote
dc42
That helps, but you still need to provide strain relief at both ends of the cable, and ensure that the cables can't chafe on anything. And I suggest you choose highly flexible cable.

5. I will look into implementing highly flexible cables with strain relief.

Quote
dc42
Yes. You need only cut the filament into small enough pieces to fit in the bottle that you are mixing the slurry in. Or you can buy solvent pipe cement, which is much the same thing as ABS slurry.

6. We have more ABS and acetone than we need so that shouldn't be a problem.

Thank you

---

The (getting better) Noob

Zapadee zoopidee.

For the heat spreader and build surface, I suggest an aluminium plate about 5mm thick. Attach the silicone heaters to the underside. If the plate is sufficiently smooth and flat, you can use the top as the build surface. Or you can clamp a glass sheet to the top. Make the plate slightly larger than the heaters, so that you have a margin around the edge, giving you room for the bed supports, the clamps for the glass, the cable terminations and strain reliefs, etc.

A difference in the resistance of two heaters of a few percent will give the same percentage difference in heating power. So 1% or 2% difference in resistance shouldn't matter. OTOH, a 10% difference would matter.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Why would you use silicone heat beds on aluminium instead of resistors? resistors have pretty precise resistance, and you can also use a number of reprap heatbeds, which can go from 24 volts.

PET and FR4 both seem like good printing substrage, because you can remove them and bend them to get the plastic off. makerbot uses pet sheet instead of glass.

Quote
skynetprinter
Why would you use silicone heat beds on aluminium instead of resistors? resistors have pretty precise resistance, and you can also use a number of reprap heatbeds, which can go from 24 volts.

PET and FR4 both seem like good printing substrage, because you can remove them and bend them to get the plastic off. makerbot uses pet sheet instead of glass.

Resistors have very localized heat and require many to get even heat and the person may not be using a standard size bed and once you get past the size of the MK style it's rare to find anything that isnt Kapton/Silicone.

---

Please add your printer design to [reprap.org]

I see. Okay, I have been looking at aluminium plates and they are pretty easy to find so that shouldn't be a problem. And the silicon heaters availability is no issue either. The only problem will be the flatness of the plate because if it's really large (as it already will be) then by supporting only the 4 corners, it would tend to create a well in the middle of the plate, if it's heavy enough. Even more so when heated. The glass clamping might have to be taken into consideration for this.

So then I will just have to ensure that the resistances of each bed is almost equal.

As usual, thank you.

---

The (getting better) Noob

Zapadee zoopidee.

Quote
3DPrintingNoob
The only problem will be the flatness of the plate because if it's really large (as it already will be) then by supporting only the 4 corners, it would tend to create a well in the middle of the plate, if it's heavy enough. Even more so when heated.

You may wish to put some ribs under the aluminium plate, between the heating pads. Alternatively, build a rigid sub-frame from e.g. aluminium extrusion, and support the heating plate on that using threaded pillars, fitted around the edges of the plate and also between the heating pads. You may be able to use the threaded pillars to level the bed too.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

It would depend alot on the lattice micro structure of the aluminium.

The problem would be attaching the ribs, could it be done by welding the ribs onto the bottom of the plate?

Is the sub-frame suggestion regarding a sort of "web" structure beneath the plate?

I would probably be able to select the specific aluminium class I would like.

Thank you

---

The (getting better) Noob

Zapadee zoopidee.