Build Bed material testing - methods, results and reasons.

Hello Everyone,

I started testing a few different bed materials with the intention to find a good build surface for COLD building and then to see what happens to the best surfaces when heated, a little and a lot.

I get good results with a COLD Bed - Kapton tape stuck onto a 3mm Opal Polycarbonate sheet extruding ABS, only slight warp with most medium sized parts, but I recently changed the Polycarb sheet to Aluminium, still with Kapton on top and my results are not good at all. So why does it stick better with PC under the Kapton and not Alu?

I have yet to get ABS to stick to Cold PET tape, but I can't understand why if it's so good when hot? (Not tested Hot PET yet)

I have a new setup that allows me to change bed material really easily and get very accurate distance from my nozzle every time so it's quite a repeatable test.
So last night I tested 8 different materials in the space of about 1 hour. - I have a long list of surfaces I want to test from Paper to Marble and anything else I can get my hands on.

I know different materials have been tested and different results reported, so does anyone have a good list of what was tested and some reasons why they are good or bad surfaces to build on?

I also found this page - Build Materials
But it states not to update it, is it worth doing a new page with a nice neat list of materials and results? Anyone want to help me document it and test some other materials? I'm going to move onto some more exotic ones shortly.

For reference I have a 0.5mm Nozzle and I always have this 0.4mm above the build surface, can anyone comment if this is a good distance? if not what do you do?

Rich.

---

[richrap.blogspot.com]

According to a page in the wiki the Z-axis zero should be just far enough above the bed for a single extruded filament to fit under without touching.

Quote
RepRapSoftwareTweakingManual
Adjust the z=0

Move the nozzle out over the bed. By adjusting the z-flag and homing the bed using the stepper exerciser you can set the datum position for the bed.

The software (at time of writing) prints its first layer at z=0. Using the filament from the previous test adjust the z=0 height so that the filament can freely slide under the nozzle.

Test your setting by printing the widebar.stl border again. Use the callipers to make sure that the nozzle is not deforming the profile of the segment by being too close to the bed.

RepRapSoftwareTweakingManual wiki page

---

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

If you, or others reading this thread, haven't seen it, there's a forum topic over in the polymer group discussing theories on why plastic sticks or doesn't stick.

Theories on sticking/not sticking

---

I would think the high thermal conduction of the aluminum bed cools the extruded filament down too fast. try adding a couple more layers of kapton, which might act as insulation.

Thanks Buback

If I carry on testing more materials I will post my results in the polymer group section, thanks.

I did test 'normal' masking tape, it's no good for ABS, three different types of paper/card also didn't stick.

I do now have some PLA but have not yet tested it, I will do a bit more investigation on ABS first.

I'm not expecting to find a magic material that counters the warp without the use of heat, but it would be great to understand a little more about what makes things stick.

---

[richrap.blogspot.com]

I've been thinking about glass recently; specifically, why does it always feel cold when it has such a low thermal conductivity?

I know that that isn't really a scientific question, but it got me wondering anyway. I found some info at the corning museum of glass website regarding the thermal properties of glass. This was interesting:

Quote
www.cmog.org
Thermal conductivity is a measure of the ability to conduct heat through the body of that material. In an opaque material, thermal conductivity can be measured straightforwardly. In a transparent material such as glass, any measurement of conductivity becomes complicated by the fact that glass also conducts heat internally by radiation. Values for conductivity rise with rising temperature and above 400°C the contribution from radiative transfer is appreciable. Radiative transfer depends on the thickness of the sample.

Because of this complication, published figures for glass thermal conductivity show considerable disagreement. And because of this uncertainty, thermal conductivity in glass is not used extensively as a design parameter. Other factors are usually more important.

Thermal conductivity is stated in the English system in BTU (British Thermal Units) times inches of thickness per hour per square foot of area per degree F. Room-temperature values for glass lie between 4.23 and 10.10, well below the corresponding values for metals. However, in some applications glass equipment will transfer heat more effectively than metal equipment, in spite of the lower thermal conductivity of the glass.

perhaps this is why plastic doesn't stick to cold glass plates? I'd like to use glass as a bed surface, mainly because it is so flat.

Edited 2 time(s). Last edit at 11/09/2010 04:55PM by Buback.

Rich,
I have not read the plastic sticking or not sticking theory thread, so this may duplicate existing comments, or be all new. But here is my best shot.

When you put Kapton tape on polycarbonate, then aluminum, then glass, I suspect you will see the least warpage on the polycarbonate. My theory is that the polycarbonate has a little flex, so as a corner begins to pull up, the plastic underneath bends a little to follow the warped bottom of the piece, and stick holds on by sticking to a larger part of the piece, holding it flatter. With the glass and aluminum, the bottom is much stiffer than the ABS, and so when it starts to warp, the bed does not change at all to adapt to the bottom of the ABS, so that the critical sticking force is exceeded much closer to the center of the piece. With less area held by sticking, it is able to warp more than with a compliant bed.

If this theory is correct, gluing a high temperature foam or rubber (like the silicone rubber oven mitts and trivets) to the rigid base, then covering the yielding stuff with Kapton tape should allow the sub-surface of the bed to stretch or shrink to match the bottom side of the ABS and stick the whole way to the edge. If the compliance is balance between sticking to ABS all the way to the corner with being so mushy it does nothing to stop the warpage, we might find something superior to Kapton covered polycarbonate. Another advantage is that silicone rubber sheet over aluminum with heater wires/resistors will reduce heat loss where there is no plastic piece, and still warm the bottom of the plastic piece because it will add more insulation a trap heat under it, further helping to keep the parts from warping.

Mike

ABS doesn't stick to glass at any temperature. PLA sticks very well to hot glass.

Even if you can get ABS to stick to cold Kapton it won't be a strong enough bond to resist warping. ABS on a cold bed generates enormous warping forces. Even if you manage to hold it flat it will warp after it is removed.

---

[www.hydraraptor.blogspot.com]

Thanks for the feedback everyone -
Well, I had some fun testing everything I could find and here are my results -

My reference is Opal Polycarb and Kapton tape, for me that's really good.
In all tests I printed a 20mm Cube using the same Gcode each time and my Natural ABS is at 236 Deg C Same print speed (1000mm/Min) for all layers.

PET tape = Not great, initially sticks well, but slowly un-sticks before 5 layers are completed – started warping - Stopped

Sticky backed paper (180gm) = No adhesion at all.

Silk Card (270gm) = No adhesion at all.

Normal paper masking tape = No adhesion at all.

Chopping board (PolyProp) = Total disaster ! Hot-end heat seemed to bend the PP upwards causing it to weld into layer 2 - a total mess – Stopped.

Self Healing Cutting Mat = Very Good!, worked as well as Kapton, (maybe a little better) firm to remove – only downside was some marks on the surface and a tiny trace of surface melting, so it would not last long-term. Seemed to make a good light-weld with the ABS - Anyone know what these are made of?

Vinyl Tape = Really Bad, No adhesion at all.

Fibreglass Plasterers tape = Not bad, 0.75mm corner warp and easy to remove, not great underside as you may expect.

Tufnol 5mm (Carp) = Good start, but slight lifting towards the end, more warp than Kapton

Tufnol 5mm (Carp) Lightly sanded = Same as above, just with a matt underside at the end.

Limestone Highly polished (Cold) = Bad - No adhesion at all.
Limestone Highly polished (Heated) = Bad - No adhesion at all.

Granite (Polished and cleaned) = as limestone, not good.

So anything that is Glass-like seems really bad for ABS.

I read Nophead's blog about Copper and PLA and I was wondering if anyone has tested really well etched copper clad (So it's Salmon Pink in colour) ? It makes a big difference when you transfer toner to pre-etched (Pink) copper - Any thoughts?

@Nophead -Do your nice flat parts warp if you re-heat them again after they have been removed?

I am going to stick to PC and Kapton for ABS for now and move onto testing PLA and hot glass next month.

Thanks Bob for the pointer about nozzle clearance, I tested between 0.3 and 0.7 and my setup seems most happy between 0.4 and 0.5 clearance at the moment.

Rich.

---

[richrap.blogspot.com]

Quote

@Nophead -Do your nice flat parts warp if you re-heat them again after they have been removed?

I have never done that myself but I guess they would warp a bit if heated. I sold a set to Foton and he sold them on later (presumably after making moulds from them) and they were then warped, but not in the way they warp when being made. It looked like the sides tapered inwards towards the top. I have no idea how they got like that but I presume they must have been heated.

---

[www.hydraraptor.blogspot.com]

One of the reasons for warping is because the plastic cools too fast. Aluminium is a great conductor of heat, and so (IMHO) cold aluminium is sucking all the heat out of the plastic, making it cool faster.

Conversely, if you heat the Aluminium up (which is relatively easy), the heat doesn't flow out of the plastic as fast/much, so it cools slower.

While other substances will heat up to the right temperatures, Aluminium has the advantage that:

1. It is fast to heat.
2. It doesn't tend to deform much under uniform heating around the 4-6mm thick mark.
3. It's relatively cheap.

Buback Wrote:
-------------------------------------------------------
> I've been thinking about glass recently;
> specifically, why does it always feel cold when it
> has such a low thermal conductivity?
>
> I know that that isn't really a scientific
> question, but it got me wondering anyway. I found
> some info at the corning museum of glass website
> regarding the thermal properties of glass. This
> was interesting:
>
>
> Thermal conductivity is a measure of the ability
> to conduct heat through the body of that material.
> In an opaque material, thermal conductivity can be
> measured straightforwardly. In a transparent
> material such as glass, any measurement of
> conductivity becomes complicated by the fact that
> glass also conducts heat internally by radiation.
> Values for conductivity rise with rising
> temperature and above 400°C the contribution from
> radiative transfer is appreciable. Radiative
> transfer depends on the thickness of the sample.
>

Hey Buback,

Below 400 C, thermal conductivity should be quite low. Glass is very opaque to the radiation at lower temperatures, as we found in the glass nozzle thread.

In my experience, glass doesn't typically feel cold to the touch (compare glass cookware with metal cutlery). But thermal conductivity is not the only important parameter that affects coldness to the touch. Although less important than conductivity, specific heat can also be important, because it measures how much energy the material requires to increase by a degree. A material with a higher specific heat will feel cooler than one with a low specific heat.