Posts: 165
Joined: Wed Jul 19, 2017 9:37 am

Problem with Holes in PETG

Fri Feb 01, 2019 3:51 pm

I'm not sure there is a good fix for this, but here's the problem:

I use PETG a lot for it's strength and (relative) ease of printing. It's one annoying property is that it tends to stay soft & sticky for a while, so it strings like crazy without the right settings. On a couple recent models, I've discovered a new issue with vertical holes. The latest model has 5 perimeter passes. It prints the outside perimeters just fine on the first layer, because the model is large enough that each pass has time to cool before the extruder comes around again.

Before it fills in the bulk of the first layer, it prints the perimeter of the hole. The hole isn't overly large, so each pass doesn't have much time to cool before the extruder comes around again. Every time I try to print, one or more of the perimeter passes for the hole sticks to the extruder and gets ripped up off the build plate. I'm using a thin film of glue stick for adhesion, and it works great for the full model, but it can't keep the tiny half molten pieces of the hole perimeter stuck down against the forces of the next extruder pass.

Although the general advice for PETG is not to use cooling, I've found it essential for smaller models. However, if I use cooling for the first layer hole perimeter, I have to use it for the entire first layer, and I suspect that may cause adhesion issues for the overall model.

I don't know of any way to tell S3D to use cooling for just a specific section of a layer. I'm already using a 40% 1st layer speed, and a 50% outline underspeed. I could try to slow the outline underspeed down further, but I'm not sure if that applies to only the 1st outline pass, or to all the perimeter passes. I think it's just the outer/inner most perimeter pass, which won't help.

Any tricks of ideas on how to lick this problem?


Posts: 351
Joined: Tue Mar 06, 2018 10:35 am
Location: Philadelphia PA Area

Re: Problem with Holes in PETG

Fri Feb 01, 2019 8:12 pm

what is your main speed set to. I never go over 45mms for large models and 35mmS for small stuff sometimes even 30. If this small part is a stand alone piece you can tray a skirt.. On the additions tab set the distance from part to 0 and try 2 or 3 outlines with one layer. You will have to cut this extra plastic off but it is a good way to hold tiny stuff to the bed.
Good luck with it!

Posts: 165
Joined: Wed Jul 19, 2017 9:37 am

Re: Problem with Holes in PETG

Fri Feb 01, 2019 8:43 pm

The part is good sized, and doesn't need a skirt, although if I turn on cooling, it might help with any warping. The problem is the perimeter of the hole is small. I fiddled with some parameters & it's doing better now, but I may experiment with cooling.

I slowed the first layer & perimeter way down, but that made a mess. All of my carefully tuned retraction & wipe settings were optimized for a higher print speed.

Posts: 51
Joined: Sat Feb 20, 2016 1:14 am
Location: Germany

Re: Problem with Holes in PETG

Sat Feb 02, 2019 3:34 am


indeed you should not use cooling for PETG, especially for the first layer.
Only (longer) bridges need cooling.
Instead of (active) cooling you should print more objects to give the layers enough time to cool down.
For a good adhesion you should use Kapton as build surface at 57-60°C bed temperature or additionally a one layer brim for very tiny parts.

I exclusively use PETG for over 2 years with great success.


Posts: 165
Joined: Wed Jul 19, 2017 9:37 am

Re: Problem with Holes in PETG

Sat Feb 02, 2019 9:21 am

I've found my success of PETG prints is highly model dependent. Certain features & geometries require changes in various settings, much more so than PLA.

The problem with the holes is that the perimeters are printed as a small separate item after the main perimeter, but before the first layer fill, all in the same layer. Ideally, the hole perimeters need different settings from the main outline & fill, but S3D only allows changes to speeds etc. on a per layer basis. Printing multiple items still works by layers, so that won't help. The model isn't small, it's the small feature size of the holes and the way the layer is sequenced that is the problem. If they did a small perimeter pass & went off someplace else for a bit, the problem would go away. In fact, I've also seen small defects from the same issue with the first layer infil in corners. With a diagonal infil, the corners have very short lines printed one right after another, and you can have the same problem with inadequate cooling.

I've managed to adjust things so that I have a functional part, mostly by slowing the first layer & perimeters WAY down. I can move on, but it still has visible defects where one pass of a hole perimeter got partially ripped up by the next one. While I'm developing my next model, I think I will try a little bit of cooling on the first layer & see what that does.

TN Yankee
Posts: 11
Joined: Mon May 30, 2016 8:10 pm

Re: Problem with Holes in PETG

Sat Feb 02, 2019 7:52 pm

I understand your frustration with getting small details in the 1st layer to adhere. In the case of your holes, consider:
1) Additional perimeters for the first layer, if not appropriate for the whole part (i.e. a separate process for the 1st layer if necessary). Print these from the outside in, so the largest perimeter is laid down 1st.
2) Eliminate the small holes from the 1st layer, likely via the CAD model. Start them from the 2nd layer on, & drill or ream out the hole as post-processing. This also works on top layers if the perimeters around holes isn't desired, though bridging concerns may arise.

Posts: 165
Joined: Wed Jul 19, 2017 9:37 am

Re: Problem with Holes in PETG

Sun Feb 03, 2019 9:52 am

Thanks for the ideas. It hadn't occurred to me to modify the CAD model.

Adding cooling didn't make much difference good or bad. No issues with warping, but the defects remain.

I'm already printing the first layer quite slowly, from outside to inside. I think adding perimeters is just going to increase the probability that one of the perimeter passes will get ripped up.

One issue may be that I'm lifting the extruder at the end of each pass, and that may be adding to the problem. I may turn that off for the first layer. Also, as I mentioned, the slow speed messes with all my careful wipe, retract, etc. settings to minimize stringing. If I do the first layer as a separate process, I can adjust those settings to get good adhesion at the slow speed, and then go back to my normal settings for the rest of the model.

As I mentioned, it's not just the holes. Regions of 1st layer fill that have multiple short passes (like diagonal fill in a corner) can also rip up the first layer. I've started on another model, and it has defects in the first layer fill, even without a hole.

One thing that helps is to make sure the extruder nozzle is really clean before it begins printing. With PETG, there always seems to be a bit of melted gunk on the nozzle at the end of a print, and that just increases the likelihood that it will stick to a bit of the first layer and rip it up.

I suppose I can also play with my build surface. The glue stick works great in all other regards. The prints aren't warping (even with the cooling fan on), and they fall off the print bed once they've cooled. My recollection from larger prints is that the glue stick is more aggressive than hairspray, and prints can be difficult to remove. That's why I prefer the glue stick for smaller models. My print bed is glass with Kapton, but for hairspray & glue stick, I use the bare glass side.

I think I'm going to spend some time just printing 1st layer tests to get things dialed in just for that. My latest project is a lot fussier about print quality, and is quite small (~ 15 mm square). I already know that will require slow speeds and/or cooling. The good news is that I can run a lot of different parameters fairly quickly with sequential printing, and I don't have to worry about how close together they are to avoid collisions.

Return to “Troubleshooting and Bug Reports”