Cure for warping

How print better models
jmgiacalone
Posts: 26
Joined: Mon Oct 25, 2010 10:24 am

Cure for warping

Post by jmgiacalone » Mon Nov 01, 2010 10:24 pm

I would like to begin this topic by saying that since receiving my UP printer less than a week ago, I have been blown away by the quality of the prints, the speed of the machine, and the overall experience of using my UP printer.

There was, however, one major issue with the machine which I could not live with, and I know many others have also struggled with: Warping.

I followed all of the advice posted by darkhouse http://www.pp3dp.com/forum/viewtopic.php?f=3&t=96 and othershttp://www.pp3dp.com/forum/viewtopic.ph ... 2&start=30, but I did not expect the issue to be resolved so easily.

Perhaps at this point I should explain a little about my experience with 3D printing. Since April of this year, I have been putting together a mendelhttp://reprap.org/wiki/Mendel based repstrap machine. Like the UP printer, I have only printed ABS on this machine and have had to learn
the hard way how to deal with Warping for good. I went through several designs of heated build platform before coming to the design I have today, which allows me to print components up to 240mm x 160mm with no warping at all.

The magic answer is to print ABS onto Kapton tape, with the build surface heated to 140C for the first layer, then 120C for subsequent layers.

Due to the large surface are of this heated bed, I have had to construct a heating element directly powered by mains voltage (240V AC here in the UK). This gives me nearly 500W of power and heats my bed up to 140C in under 7mins. But I'm not suggesting the UP should have a mains powered bed as the build surface area is much smaller and my mains powered bed was expensive (and time consuming) to put together.

So with my previous experience of heated bed designs, I happened to have some Nichrome wire laying around with a suitable resistance/m (50 Ohm/m), I decided to see if there was a simple solution to getting my UP's build surface to the right temperature. So I unscrewed the UP's build platform and X axis cover to reveal the heating element and connections. As proven by the pictures on chris.e's blog http://tinkerworks.blogspot.com/2010/10 ... tform.html the heating element is a long way form the build surface. Conveniently, the heating element has a resistance of 7.6Ohms, which equates to 1400mm of my Nichrome wire. This fits very nicely stuck with kapton tape to the bottom of my UP's build plate. I knocked up a small protoboard which allowed me to bypass the UP's original heating element and plug in my Nichrome wire. After some careful routing of wires and screwing everything back together, I tested my new heated build surface.

I appreciate that the temperature now seen by the firmware will not be the actual temperature of the heating element and build surface (which could cause it to struggle to maintain an accurate temperature, although I obviously don't know what control method is used, PID or ON/OFF), however I was aiming for a build surface temperature of 140C when the firmware read 100C (which I think is the temperature the UP sets the build surface temperature to during prints).

Now I understand that pp3dp may have issues with me modifying my UP in this way (in terms of warranty), but all of my changes are reversible and I had to get rid of the warping issue.

And the result is that I can now print right to the edge of the build surface (even within 1mm of the edges) without the slightest warping of my prints. I preheat the bed to 60C (or so it reads in the software) before I start my prints, and I haven't measured the surface temperature as I do not have suitable equipment, but it certainly feels hot.

So back to the point of this topic, I would like to offer a suggestion as an improvement for future designs of UP printers (and perhaps as an upgrade to the current model). That is, to use a PCB based heated build surface simliar to this one: http://reprap.org/wiki/PCB_Heatbed. The benefits of using such a design are that the heat is spread evenly across the build surface, the heating element is nearer to the build surface making it easier to accurately control the temperature, and as a mass produced item, this solution would be very cost effective.

I also have a small request regarding the raft settings. As my build surface is now completely flat (as I am using kapton tape rather than paint), would it be possible to have an option to select a raft thickness of less than 2mm please?

Sorry for the long post, but I felt it necessary to explain the reasoning behind my suggestion.

Thanks,
jmg.
jmgiacalone
www.emakershop.com

ssozonoff
Posts: 24
Joined: Mon Oct 18, 2010 6:43 pm

Re: Cure for warping

Post by ssozonoff » Mon Nov 01, 2010 10:42 pm

This is great,

Would be nice to get that PCB in 140x140 to fit the UP!.

Source is found here http://www.thingiverse.com/thing:2172

Can I ask you where you are sourcing the Kapton tape ?

Thanks,
Serge

Chris.e
Posts: 190
Joined: Wed Sep 15, 2010 10:05 am
Contact:

Re: Cure for warping

Post by Chris.e » Mon Nov 01, 2010 10:59 pm

That is wonderful jmg!
I´ve been experimenting with adjusting the heat some the last days but haven't got so far yet :)
How long does the kapton tape hold up?

airborn2
Posts: 14
Joined: Wed Oct 27, 2010 12:46 am

Re: Cure for warping

Post by airborn2 » Tue Nov 02, 2010 12:22 am

Hi Everyone,

I posted elsewhere about warping as well...

As for Kapton and heating, if you go to McMaster.com web site and type in # 35475K63 you will get
a heating sheet encapsulated in Kapton already made up.
The sheet is a bit large at 6" x 6" but it is very thin at .008" so maybe you can fold over the edges.
It has a self adhesive backing and attached leads. The specs are 115 volts and 360 watts and
will heat to a maximum of 300 degrees F (or 148 C.)
It needs a separate controller and thermocouple to control the temp, but I have several that I will probably try out.
The cost of the Kapton heater is $64.77.

I'll keep everyone informed if I get one and how it works...

Ken

P.S. On the same catalog page are the same sheet heaters in fiberglass reinforced silicone. Might be something
to try as well for the "non-stick" properties of the silicone. I may buy one of each.

trebuchet03
Posts: 467
Joined: Sun Sep 12, 2010 6:52 am

Re: Cure for warping

Post by trebuchet03 » Tue Nov 02, 2010 1:10 am

airborn2 wrote:Hi Everyone,

I posted elsewhere about warping as well...

As for Kapton and heating, if you go to McMaster.com web site and type in # 35475K63 you will get
a heating sheet encapsulated in Kapton already made up.
The sheet is a bit large at 6" x 6" but it is very thin at .008" so maybe you can fold over the edges.
It has a self adhesive backing and attached leads. The specs are 115 volts and 360 watts and
will heat to a maximum of 300 degrees F (or 148 C.)
It needs a separate controller and thermocouple to control the temp, but I have several that I will probably try out.
The cost of the Kapton heater is $64.77.

I'll keep everyone informed if I get one and how it works...

Ken


Something to note about those heaters... 300F MAX is not how hot it will get ;) You will need to heat sink it (which the bed will do except for the parts hanging over).

(to everyone else) If you don't have a temperature controller - you're likely to burn the mat up... Especially at 10W/in density (that's a lot of heat o.0).

I have a similar mat in a machine at work - we use a PID controller (http://www.omega.com/ppt/pptsc.asp?ref=CSC32Series) with a thermocouple attached on with polyamide tape (which is what the Kapton brand is). That controller is overkill for this though :p



As you'll want to replace the Kapton... You'll get more bang for your buck with the silicone version. Alternatively, you could wire two heaters along the edge and keep the stock heater in place. Cold spots will still exist though :/


OR... with all this modification afoot.... you could just bump the temperature up on the stock heater :p
Engineer - Mechanical, Mechatronic, Reliability, Etcetera
One Offer
Mass Producer

Chris.e
Posts: 190
Joined: Wed Sep 15, 2010 10:05 am
Contact:

Re: Cure for warping

Post by Chris.e » Tue Nov 02, 2010 8:37 am

The magic answer is to print ABS onto Kapton tape, with the build surface heated to 140C for the first layer, then 120C for subsequent layers.
Do you mean that you manually lower the temp after the first layer or that because of getting off the surface it automatically becomes about 120C?

ssozonoff
Posts: 24
Joined: Mon Oct 18, 2010 6:43 pm

Re: Cure for warping

Post by ssozonoff » Tue Nov 02, 2010 10:10 am

I have just ordered various "types" of Kapton from McMaster.

Both Sheet and Tape to experiment with.
I would also like to make the PCB heatbed found here http://www.thingiverse.com/thing:3919 but dimensioned 140x140 for the UP!

Thanks,
Serge

Gilius
Posts: 148
Joined: Sun Oct 17, 2010 12:33 pm

Re: Cure for warping

Post by Gilius » Thu Nov 04, 2010 9:25 am

What do you mean lower to 120C from 140C? You mean manually?

jmgiacalone
Posts: 26
Joined: Mon Oct 25, 2010 10:24 am

Re: Cure for warping

Post by jmgiacalone » Thu Nov 04, 2010 10:46 pm

When printing on my reprap machine, I can control the build surface temperature. So I set the temperature to 140C for the first layer of my print, then I program the temperature to 120C for the rest of the print.

I of course do not have the ability to do this on my UP!
jmgiacalone
www.emakershop.com

Gilius
Posts: 148
Joined: Sun Oct 17, 2010 12:33 pm

Re: Cure for warping

Post by Gilius » Thu Nov 04, 2010 10:53 pm

That's rather odd because the heat deflection temperature of ABS is 104C. 140C is well into the melting point of ABS. 120C is on the high end (as far as I know the melting points of ABS vary between 105C and 125C)

trebuchet03
Posts: 467
Joined: Sun Sep 12, 2010 6:52 am

Re: Cure for warping

Post by trebuchet03 » Fri Nov 05, 2010 4:32 am

Gilius wrote:That's rather odd because the heat deflection temperature of ABS is 104C. 140C is well into the melting point of ABS. 120C is on the high end (as far as I know the melting points of ABS vary between 105C and 125C)

Stratasys ABS M30 has a deflection temperature of 96C - remember that this is deflection under load following ASTM D648 http://www.astm.org/Standards/D648.htm... You'll need to go a fair bit hotter for it to deflect under its own weight given how well supported it is on the build platform.


While "melting point" may be relatively low - it doesn't flow well until much higher temps.... For industrial processing, "melt temperature" will be somewhere in the realm of 230-260C with lower temperatures the deeper into the machine you get (front/middle/rear barrel) - the melt chamber has beads which have low contact surface area....

Stratasys runs their material liquefier somewhere between 260C and 290C (depending on material and equipment).
Engineer - Mechanical, Mechatronic, Reliability, Etcetera
One Offer
Mass Producer

Chris.e
Posts: 190
Joined: Wed Sep 15, 2010 10:05 am
Contact:

Re: Cure for warping

Post by Chris.e » Fri Nov 05, 2010 8:41 am

Could it be an issue with the center and edge temp? I don't mean the 10c difference, more that the center may be around 100 and the edge 90, would it be better that the center was 120 and the edge 110? I think what I´m trying to say is, is there a specific temperature where most of the shrinking in ABS occurs? Like it´s relatively stable from 20 to 95c and then it expands up until 100c and then it´s pretty stable again? If you know how I mean? :)

Gilius
Posts: 148
Joined: Sun Oct 17, 2010 12:33 pm

Re: Cure for warping

Post by Gilius » Fri Nov 05, 2010 9:43 am

Interesting. I will try to use my heatgun to keep the temperature around 120C when my Up arrives.

Chris.e
Posts: 190
Joined: Wed Sep 15, 2010 10:05 am
Contact:

Re: Cure for warping

Post by Chris.e » Fri Nov 05, 2010 10:49 am

I tried the heat gun trick and it worked for me although I didn't have enough patience for longer prints.

Linkreincarnate
Posts: 137
Joined: Wed Aug 11, 2010 5:37 am

Re: Cure for warping

Post by Linkreincarnate » Sun Nov 07, 2010 2:44 am

What about printing a shell around the object you are trying to print? The shell could help keep a buffer of warm air around the print. Anyone think that might work?

Chris.e
Posts: 190
Joined: Wed Sep 15, 2010 10:05 am
Contact:

Re: Cure for warping

Post by Chris.e » Sun Nov 07, 2010 10:48 am

Linkreincarnate wrote:What about printing a shell around the object you are trying to print? The shell could help keep a buffer of warm air around the print. Anyone think that might work?
Is that what the "stable support" means? Think someone said that here although not sure. And there was someone else here who had done corners for his large prints on the reprap, although the search function times out for me now so cant find it right now.

Kurt
Posts: 6
Joined: Tue Aug 31, 2010 11:18 pm

Re: Cure for warping

Post by Kurt » Sun Nov 07, 2010 3:46 pm

Hi There

That was Nophead on his blog serch for some corners like it hot

Cheers

Kurt

jmgiacalone
Posts: 26
Joined: Mon Oct 25, 2010 10:24 am

Re: Cure for warping

Post by jmgiacalone » Sun Nov 07, 2010 7:14 pm

Hi all,

I was thinking of trying to print with added walls around my parts, but I don't think it'll work. Because the UP prints a raft, the walls would be on the same raft as the part so also subject to warping. Unless the wall were positioned far enough away, but that isn't necessarily practical and would probably not have any effect at that distance. Plus the fact that I hate printing sacrificial material.

I could be wrong of course.
jmgiacalone
www.emakershop.com

cyberbadger
Posts: 125
Joined: Sun Sep 12, 2010 12:26 am

Re: Cure for warping

Post by cyberbadger » Sun Nov 07, 2010 11:34 pm

Hi all,

Is there some trick (besides heat) with kapton tape? I have tried a few times, and the print invariably slides off the platform while printing...

-cyberbadger

trebuchet03
Posts: 467
Joined: Sun Sep 12, 2010 6:52 am

Re: Cure for warping

Post by trebuchet03 » Mon Nov 08, 2010 12:47 am

Gentlemen (and Gentlewomen?), a simple solution.

Preheat with a little blanket. While you can technically get the software to read 100C (or whatever) faster, continue to preheat for 15-20 minutes. I used the cleaning cloth that came with me laptop (eee).

Image
http://4.bp.blogspot.com/_ChGrYiAXl9M/T ... G_1640.JPG

I just printed a low X-Y aspect ratio flat object/part with zero warping using the green paint. 130mmx135mmx7mm, square corners and a couple holes near the center. Solid Fill, 4 layer walls, normal quality.

It was also a chilly morning - 17.5C when I started printing...

Job finished and I let it cool. I've been letting all parts cool completely (or at least close to ambient) before separating it from the raft. It separates much easier and the plastic isn't soft during clean up.


I'll take a probe to it later and make some plots ;)
Engineer - Mechanical, Mechatronic, Reliability, Etcetera
One Offer
Mass Producer

Locked