Welcome! Log In Create A New Profile


Non heated bed question

Posted by B3ck 
Non heated bed question
September 18, 2012 11:52PM
I am quite new to the community. I have not saved up enough for a build of my own yet, so this might sound dumb to those that have already played with one.
Instead of doing a heated bed, could you use a base plastic that is cross hatched, with two or three layers that are slightly miss-aligned to give a sturdy bade that won't move once cooled and clamped down? And extrude the project layers at a slower rate, or with time in between layers to reduce the contracting and curving? Or even fan cool the project layers in between layers?
Re: Non heated bed question
September 19, 2012 05:25AM
That's subject to research, right? Undobtly there are patterns which make warping stronger or weaker.

That said, PLA can be printed without heated bed reasonably well and building a printer is an experience on it's own. A heated bed can always be fitted later.

Generation 7 Electronics Teacup Firmware RepRap DIY
Re: Non heated bed question
September 19, 2012 05:53AM
Fully cooling the plastic between layers is an interesting idea. The first layer of PLA usually sticks very well to blue painters' tape without a heated bed. It's the combined shrinking of the layers above that pull it off, or lift the tape. If each layer is cooled fully before the next layer is put on, there is only one layer's shrinking to effect those below.

My only worry would be a weaker layer bond, so more chance of delamination, possibly. Dealing with ooze between layers might be an issue. And slower printing!
Re: Non heated bed question
September 19, 2012 08:05AM
You need the previous layer to be somewhat hot in order to get good bonding between the previous layer and the layer you are currently printing.

You can see this effect even with just a cooling fan: parts printed with the fan on a high setting are much, much weaker than parts printed with no fan running.

tl;dr: bad idea.
Re: Non heated bed question
September 19, 2012 10:20AM
It also would create maximum stress in the parts and give the most warping. Warping is reduced by holding the entire object just below the glass transition temperature, not by cooling it before the next layer hits.

Re: Non heated bed question
September 19, 2012 11:51AM
Not that I doubt what you're saying Nophead, and I'm sure you've tried it out in the past (you have most things!) but if each layer cools entirely before the next layer is put down, surely there isn't any stress left in that layer? And with only a small force of the subsequent layer to pull the layer up.
I watched the program on BBC4 on ceramics the other night, I hadn't realised/thought about how tempered glass is made, which is to cool it very quickly so the outside of the glass hardens, so when the interior cools it is actually under tension, which gives it it's strength. If plastic works in the same way, and assuming layer bonding isn't a problem (and it may be a huge one, I haven't tried), this could actually be better for strength.
I am, as ever, probably talking out of my proverbial! Please allow some latitude before ripping the idea to shreds.
Re: Non heated bed question
September 19, 2012 12:30PM
So the bottom layer is flat and cools to a solid.

The next layer layer grabs hold of it when it gets to it glass transition and then shrinks from that temperature to ambient. If you remove the object at the point you have one flat layer and one layer trying to bend it, both equal strength, so you already don't have a flat object.

As you add more layers you have more and more stress trying to bend the layers below. That will lift the corners once you get to a few mm because there is is nothing strong enough to hold say 10mm of ABS that wants to be curved. The force is just the same as would be required to bend 10mm of ABS, very high.

PLA is just the same but its glass transition is very low so the effect is less until you get to bigger objects.

With a heated bed just below the glass transition the bottom layers solidify but don't contract, so they form a strong base to hold the higher layers flat without having much stress in them. When the bed cools the bottom layers contract, and if the object is thin it actually warps a little in the opposite direction. I.e. domed in the middle.

To make truly flat objects at arbitrary sizes you need to keep the whole object just below the glass transition temperature for the duration of the build.

Re: Non heated bed question
September 19, 2012 09:44PM
Wow, thank you so much for the great information. Allot to think about. And makes me more comfortable in going for this with a great community behind it.
Re: Non heated bed question
September 24, 2012 05:51PM
Just out of interest how steep is the temperature profile through the material in the z-axis? Plastics aren't great conductors of heat so I'd expect a fairly steep reduction in temperature as build height increases, which makes me wonder whether it's worth having heated beds operating beyond the first few layers? If in fact all the heated bed is doing (in the latter phase of the print) is heating the environment to slow the cooling of layers to glass transition temp then there are probably better ways of achieving this. It also prompts the question of what build temperature (not extruder or bed) is actually necessary in this phase as we certainly aren't achieving temperatures above glass transition in these later layers. Or is my thinking completely wooly?
Re: Non heated bed question
September 24, 2012 08:58PM
Not wooly, but...one problem that might not be immediately apparent is that when the heated bed temp drops back below the glass temp, the part tends to self-release from the surface of the bed, at least with ABS. If it incorporates any momentum from the print head, y-axis translations, or via mechanical vibrations after release, the entire part can become misregistered.
Re: Non heated bed question
September 25, 2012 03:55PM
Yes, the blue painters tape is one way to go. I prefer more porous card-stock myself.

I find both PLA and ABS work well, ditch the heated bed, it creates more temperature gradients, and thats a problem!
Instead of printing tall layers where you have a slow forming gradient of heat, and needs a hot base, instead deposit hotter filament and immediately cool it, so fast than the heat cools by the time it goes through 1-2 layers.

Protocol (PLA): 195 degrees C. Use a strong fan on maximum. Print directly onto card stock. Ensure adequate cooling (or slow down the print head). This also makes stronger and more consistent prints. Adjust the fan to match the amount of heat deposited per layer, i.e., a larger print area (3x3 inches say) requires less rapid cooling because it Z-stacks slower.

Alternative: PLA: 185 degrees. Use a base heated to about 95C.
Re: Non heated bed question
September 25, 2012 09:56PM
I have had very, very good results printing PLA at 210c straight onto heavily scratched-up acrylic. You basically go over the surface with sandpaper for a few minutes to get the surface finish all cloudy and scratchy. At that high temperature, I get NO warping and 100% surface adhesion. Getting the prints off is harder than with a heated bed though; you need to use a paint scraper and a mallet. The big problem is that acrylic is not totally 100% flat like glass is, and the larger the sheet of acrylic, the more warped it gets. Keeping the platform level is a challenge. IMHO, a heated bed is a nicer experience, and it yields a smoother finish on the base layer, too.

But still, it can be done! Just sand the surface of an acrylic sheet and print your PLA on it at 210.
Re: Non heated bed question
September 25, 2012 10:34PM
how apply applying a roll of thin polycarbonate to a flat piece of glass?

Re: Non heated bed question
September 25, 2012 11:18PM
Didn't try that. Not a bad idea, though.
Re: Non heated bed question
September 25, 2012 11:22PM
they sometimes have adhesive backed ones. PLA might make sense too. It was dumb of me not to think of it before:


thick tolerance part cost
0.007" ±0.050" 1104T311 $1.65
177.8 micron thick.... ignore the tolerance, its ridiculous. It's usually accurate to within a few microns thickness...

No adhesive on this one.

A mesh might be better if you are heating the base though, because of thermal expansion, the film would develop kinks.
Sorry, only registered users may post in this forum.

Click here to login