Corner Warping: 3D Printing Myths & Facts

Monday, June 10th, 2013

Let me begin by apologizing for the lack of pictures here… I will post some ASAP. However, I wanted to share some findings that have taken a good deal of effort to reach. There are lots of theories out there, and taking the time to test them is a brutal process. Some solutions to warping that are presented out there require altering your design… And I don’t like that at all. You shouldn’t have to compromise your part to prevent warping!

This brief post is meant to address corner warping specifically and debunk myths. The two myths in particular that I’d like to address are 1) enclosing the printer, and 2) printing faster…

1) I’ve finally gotten around to enclosing my printer. I’m glad I didn’t waste too much time on this… Because it doesn’t make a difference. I printed two parts… One without enclosing, and the other with. They both warped exactly the same amount. I must note that I didn’t fully enclose the top… But I could feel the heat build up inside… And had there been even a slight improvement… I would have gone the distance.

2) Printing faster is another myth… Not sure whose bright idea this is… but its kinda like trying to outrun a wave… Not going to happen.

So what works? Well, lets briefly look at what’s causing the warping. Think of each layer as a string with the potential for contraction. If the strings are cooling slowly and simultaneously (although at different stages of the process), then you can imaging them all pulling together and exerting more and more force on the part as the part gets taller and taller. If however, they shrink one at a time… They impose very little force on the overall part. So how do you get this to happen?

The simplest solutions… Print really slow! By slowing the print down to 1/3 speed… I took a part that was 4″x6″ by 1″ tall that was exhibiting significant warping, and eliminated the warping all together…without the use of a brim or mouse ears. Quite an improvement! I have started making different slic3r configurations which employ different speeds based on the part size. This has been a huge improvement.

Unfortunately, prints take a long time to print. That 4 hour print takes 12 hours. Big difference! But it comes out perfect. Alternatives to this might be to cool the part down as it prints with a fan, or ambient room temp. I will experiment with this and report back. Will need to see how this impacts the heated bed and the ability for the part to stick to the bed.

One other note in regards to this, I’ve found that printing external perimeters first is very important. Firstly, it often improves the finish of the part. Secondly, sometimes warping pulls layers apart slightly and if you run internal perimeters first, they are cooling while the external perimeters are printed… Which then experience the forces of those internal perimeter layers on them and sometime pull apart slightly… If you run the external layers first… They are contracting into the internal layers as they are printed and they don’t seem to pull apart at all. This last bit is theoretical… And not proven in any scientific way… Other than to say it helps!

Leave a Reply

You must be logged in to post a comment.