Uneven/Shifted/Missing Layer Issues

Looking for some fresh input here…

[WHAT I KNOW]

I’ve have great results printing with my new TAZ6 over the last several weeks. Recently, some issues have popped up.

My latest issue is that layers are printing uneven or shifted as shown in the attached image labeled bad_print.png. Additionally, and possibly directly related, is that when I zoom in on the part in S3D Preview Mode, I see a layer completely missing right at the height where my first process ends and the second begins.

[WHAT I’VE DONE]

Tighten belts: X and Y belts felt a little loose so I tightened them up per the documented procedure
Extruder calibration: Extruded 100mm of filament at appropriate temp and speed and adjusted E-steps in firmware accordingly
Filament diameter: Measured filament diameter in several areas, averaged and noted the value
Layer height calibration: Printed 20x20x0.2mm calibration layer at 0.2mm resolution and adjusted Z-offset in firmware accordingly
Extrusion multiplier: Printed 20x20x20mm cube with no top/bottom is vase mode with appropriate test parameters and arrived at extrusion multiplier
PID tuning: Tuned my hot end and bed
S3D process settings: I input the values that I arrived at for filament diameter and extrusion multiplier

[OF NOTE]

I used a factory file from a friend who has a PowerSpec Pro 2 printer as a starting point (we are printing the same parts).
After loading his factory file, I ran the S3D Configuration Assistant to change printers to my TAZ 6 with a single extruder.
Rather than creating my own processes I simply edited those included in his factory file since we are printing the same part.

[QUESTION SUMMARY]

How do I correct the uneven/shifted layers?

How do I correct the missing layer issue?

What is the proper procedure for replicating non-printer-specific print settings from S3D processes that are tailored for printer models that are different from my own printer (as in the case where I am attempting to replicate settings for Layer, Additions, Infill, and Support)?

[ATTACHED SUPPORTING DOCUMENTATION]

bad_print.png

missing_layer.png

S3D.zip (contains the above-mentioned .png images, images of all of my relevant process settings and my TAZ6_PRINT.factory file - the object included in this file is not the actual object I am printing since that object is our own design and not meant for public distribution)

Someone please save me; I’m tired.

-calfer
S3D.zip (2.03 MB)


bad_print.png

First question with your gap between layers.

When you’re running multiple processes on simplify3d, do you have process 1 starting at 0 and stopping at (for example) 5mm and the second process starting at 5mm and unchecking the stop printing box.

It’d be very easy to accidentally say:

Process 1: Start 0mm stop 5mm
Process 2: start 6mm stop: never

Make sure your start and stop layers are common and not incremented.

jim2386,

Thanks for chiming in. The processes in question are configured like this:

Process 1: Start 0mm stop 2.2mm
Process 2: Start 2.2mm stop never

To make it a bit easier to assist in troubleshooting, I’ve attached a .zip file containing images of all of my relevant S3D process settings and my .factory file.

Welp, I at least solved your gap issue. Kind of an interesting problem that I’ve never run into, but here’s your explanation.

You set up your first process between 0-2.2mm which is fine, but you forgot to account for that your first layer is 150% thick

Therefore your first layer is 0.3mm, next layer would be at .5mm, .7mm, .9mm…2.1mm, not 2.2mm.

If you change your first process to end at 2.1mm and your second process to start at 2.1mm, you should see your gap close.

I’ve never run into this problem. After staring at your file for a while I’ve figured out why. Whenever I’ve done multiple processes, I’ve never bothered to change the first layer for each subsequent process to 100%, I’ve always left them at whatever the initial layer thickness the same for each process.

You can test this by resetting your factory file to 0-2.2mm and 2.2mm to end like which is in your zip file. Note, initially that gap has returned. Go into your 2.2-end process and change your initial layer thickness from 100% to 150% (150% matches your first process initial layer thickness). Go forward to the layer preview and note your gap is gone. So not accounting for the initial layer thickness caused the gap, but if you stay consistent with initial layer thicknesses, S3D can fill in the gap since the offsets are consistent. That’s why I’ve never seen this issue but you stumbled on it. Basically your layer math stop height was calculated incorrectly.

Either remember to include your thicker first layer in your stop height calculation, or make sure each process starts with the same initial layer thickness.



Other things I noticed in your file that maybe you’d care about:

  1. In process 1, you have 10 bottom layers and 4 top layers. In process 2 you have 4 bottom layers and 10 top layers. Your infill is the same throughout the part so there’s no reason to really have top layers at the end of process 1 or bottom layers at the beginning of process 2.

In process 1, set top layers to 0, in process 2, set bottom layers to 0. This will give you solid layers at the very bottom of the part and solid layers on top of the part.


2. Under the infill tab on both processes, you have “include solid infill every XXXX layers checked”. Again, there’s no real reason for this unless you’re just doing it just to see what it does. I’d highly recommend unchecking this box on both processes.

Once you have done these two extra steps, go to the print preview page and scroll through the part. You should see 10 solid layers at the bottom, 40% infill (arguably overkill) in the middle and 4 layers at the top (Personally I’d reverse this so you had 4 solid layers on bottom and 6-8 on top to bridge the infill. This would allow you to use a lower infill percentage like 20% or so.)

Anyway, I hope this helps. If anything doesn’t make sense, let me know. Great puzzler!

-Jim

Jim,

Thank you kindly for the informative reply and please forgive the late response.

Your explanation for my missing layer issue makes a lot of sense. In addition to learning from your information, I will make a habit of paying more attention when viewing the build process in print preview mode.

You say when you use multiple process and have configured a non-standard first layer height that you don’t change the the first layer height in subsequent processes. Would doing this not result in a noticeable layer size inconsistency in the first layer of each process as compared to adjacent printed layers on your part?

Regarding the other items you pointed out:

  1. While I did not create these processes, I can shed some light on why some settings were chosen. The part we are printing is a connector that requires a high degree of strength and structural integrity. We install these connectors with blunt force using a heavy hammer. The shape is unique in that, for us being fairly new to 3D printing, it isn’t readily apparent how to achieve the required level of strength without going overboard on layers and infill.

I’ve attached a couple of images of the model here. There are some challenging angles that require strong walls. Would you have any advice as to how to how to achieve our strength goal with regard to applying uniform walls across angles like those in the model? When we specify the number of perimeters are they applied uniformly to exterior angles of the elbow arms?

And, since there is a portion of the center cube that prints (~2.2mm) before the elbow arms begin to print, and say for example we specify 11 bottom layers and print with a 100% first layer height, am I correct to assume that there will only be one bottom layer applied to the elbow arms if we are using only one process?

As you can surmise I am having difficulty grasping how to print a uniform shell across all angles of the model.

  1. Again, the solid infill layers were an attempt at achieving our strength goal by the original creator of the model. Based on the above explanation and included images of the model, would you still recommend against these infill layers? And, again taking the above into account, what would your final advice be regarding our chosen infill percentage?

I really appreciate your time and insight.

-calfer
part_2.PNG
part_1.PNG

I will make a habit of paying more attention when viewing the build process in print preview mode.

Yes…always spend some time looking at the layer preview mode and slicing through all the layers. A few minutes there can save you hours of having to reprint a part for an error you didn’t catch!

You say when you use multiple process and have configured a non-standard first layer height that you don’t change the the first layer height in subsequent processes. Would doing this not result in a noticeable layer size inconsistency in the first layer of each process as compared to adjacent printed layers on your part?

Theoretically yes, but we’re talking about 1 layer that is maybe .3mm thick vs all the other layers that are 0.2mm thick. One layer that is 0.1mm thicker at the start of a process I couldn’t even notice. After all, a lot of what processes are used for is to speed up your print. we print very thin layers in areas with lots of curves or that needs lots of detail. We print thicker layers on vertical walls where details aren’t important. Basically, you are correct, but in actual application when you’re looking at a part from half an arms length away, do you really notice a 0.1mm different for one layer? Hardly…unless you’re reeeeeeeally trying. If it does bother you, I explained above how to mathematically take care of it so subsequent processes don’t have a thicker initial layer :slight_smile:


**Regarding the other items you pointed out:
\

  1. While I did not create these processes, I can shed some light on why some settings were chosen. The part we are printing is a connector that requires a high degree of strength and structural integrity. We install these connectors with blunt force using a heavy hammer. The shape is unique in that, for us being fairly new to 3D printing, it isn’t readily apparent how to achieve the required level of strength without going overboard on layers and infill.**

While infill does add some strength to a degree, what really adds your strength is the number of outside VERTICAL walls or “perimeter shells” as simplify3D calls them. Cura calls them “shell thickness”. Let’s stick with S3D for now since that seems to be what you’re using.

We want to increase the overall wall thickness. Assuming you’re using a standard Taz 5 or 6, your nozzle diameter is probably 0.5mm. This is important to know when dealing with shells. Each “shell” is one pass with the nozzle on the print. Example: If you set the shells as “2” you’ll have a 1mm vertical wall thickness. If you have 4 shells your vertical wall thickness will be 2mm = 0.5 * 4. Adding more shells will thicken your vertical walls and add strength.

I’ve attached a couple of images of the model here. There are some challenging angles that require strong walls. Would you have any advice as to how to how to achieve our strength goal with regard to applying uniform walls across angles like those in the model? When we specify the number of perimeters are they applied uniformly to exterior angles of the elbow arms?

What you’re missing here is the definition of “top solid layers” and “bottom solid layers”. Bottom solid layers are any horizontal or near horizontal layers on the bottom of the part NOT NECESSARILY touching the build plate.

Top solid layers are the opposite. Layers that are horizontal or nearly horizontal on top of the part NOT NECESSARILY the upper most layer and the subsequent layers before that. It’s easier if I use some pictures:

If I tell S3D I want 10 solid bottom layers and 2 solid top layers, look at what I get in the print preview:
thickwalls.PNG
Note that your ramped areas also have 10 solid bottom layers. (Ignoring your initial layer thickness increase for simplicity of math) , I set my layer thickness to 0.2mm, there are 10 layers, 0.2mm *10 = 2mm, my bottom walls are 2mm thick uniformly.

Now let’s change my top layers from 2 layers as in the last picture and change them to 10 layers:
thinwalls.PNG
Note that now all my top layers, even the angled ones, are 10 layers thick, or 2mm total wall thickness. Great! However, notice my left and right vertical walls are still 2 shells thick. Since our nozzle is 0.5mm wide, our vertical walls are only 0.5*2 = 1mm thick. Boo! I want uniform thickness. Since we agree that the top and bottom walls are 2mm thick each, we want the vertical walls the same thickness. The nozzle is 0.5mm wide so if we have 4 shells, 0.5mm x 4 = 2mm. What does that look like in S3d?
bottomsolidlayers.PNG
Ah ha! Now notice all outside walls are of uniform thickness. You should repeat these steps for 3mm wall thickness etc. This was all done in 1 process and does not require multiple processes.

Angus over at Maker’s Muse has a nice Youtube video giving some part strength 101 on how to build tougher parts here: https://www.youtube.com/watch?v=1NzDr1YAb8Q

And, since there is a portion of the center cube that prints (~2.2mm) before the elbow arms begin to print, and say for example we specify 11 bottom layers and print with a 100% first layer height, am I correct to assume that there will only be one bottom layer applied to the elbow arms if we are using only one process?

Hopefully my explanation above answers this question :slight_smile:


As you can surmise I am having difficulty grasping how to print a uniform shell across all angles of the model.

Taking the above into account, what would your final advice be regarding our chosen infill percentage?

Since this seems to be a small part, you could print this part with a uniform outer shell thickness and the standard 20% infill and print it again at 50% infill and try it out. Most strength comes from the outer shell thickness. Of course you said you’re hammering these blocks. Impact resistance is a function of the material too.

Here is a chart that helps choose that material to use. Note there is a line there for impact resistance:


TPU has the best impact resistance, but these are rubberized materials and not probably useful for your application. Nylon is an extremely tough material, but it can be a little more challenging to print with and is hygroscopic so you have to keep it very dry or you’ll need to put it in an oven to dry it out to be able to print with it. ABS or PET are your best bets for these parts. Polycarbonate is tough, but you have to print it at a very high temperature which can lead to other problems if you switch materials and you’re not careful. PLA is TERRIBLE for impacts, don’t use it in this application.

I hope this helps! If you have any other questions, please don’t hesitate you ask. I’m happy to help!


Jim

Jim,

You’ve been most wonderful in responding to my questions.

Your information is clear and easy to understand. I’ve enjoyed and learned from many of Angus’ videos but don’t recall watching the one you recommend so I’ll give that a go as well.

Thanks also for the material chart. While PLA may be labeled a brittle material, we have been very happy with its impact resistance. We use a 3lb hammer when installing these connectors and largely see zero deformation, cracking or other damage during installation. I will continue to tweak the design in an effort to maintain strength while reducing print time/material usage while using the knowledge I’ve gained from your replies.

Thanks and see you around the forums.

-calfer

Still struggling with the uneven/shifted layer issue…

I’ve attached a couple of images of the latest print.

In addition to the shifted layers, I’m also noticing that the top layers are showing gaps. I’ve specified 10 top and 10 bottom layers. Perhaps I should add a process to slow down printing for the top several layers?

Extrusion multiplier has been calibrated and set to 0.83 in S3D.

I’ve noticed that the extruder assembly has a lot of play in the direction of the TAZ 6’s Y axis. It doesn’t feel loose in a way that can be fixed by tightening screws but can easily be manipulated in that direction by hand. Not sure if this is a contributing factor or can otherwise be correct.

Anyone from AO have any input?

-calfer


I can answer the gap question easily.

Did you measure your filament diameter with a pair of calipers? Make sure your filament diameter is set to your actual caliper measurements under the “other” tab. If you’ve already done that and you still have gaps, go to the “extruder” tab and increase your extrusion multiplier by 0.01 or 0.02 and try again. This will extrude a bit more plastic per layer and will fill the gaps. Don’t add too much because overextrusion can be a bad thing.

You are currently underextruding by a tiny bit.


As for the layer shift, I think you have a hardware issue. The best way to check this is to print the same part but use CURA. If the layers still shift, you know it’s hardware and you need to contact lulzbot support. They have THE best support around. I’ve gotten return phone calls/emails at 11pm at night on the weekends. They are extremely helpful. They support cura though so do a print with cura and if the waviness is still there, shoot them an e-mail with pictures and attach your cura profile and they will step you through what to try there.

If you print with cura and the part is perfect, then we need to do a bit more pondering on simplify3d…however I really doubt its the software. I’ve been wrong before though :slight_smile:

Jim

Jim,

Thanks for the reply.

I am performing the required calibration steps prior to any printing. Filament diameter is being measured, averaged and input. Extruder is calibrated but perhaps, as you say, I just need to tweak it a minute amount to get it as close to perfect as I can. I’m using a 0.83 multiplier currently. I’ll give that a go, thanks!.

Interesting note about trying Cura as this is what I did after submitting my post over the weekend. I input the parameters to match what I was using in S3D (resolution, extrusion multiplier, shell thickness, bottom/top layers, infill, filament diameter) into Cura and printed the same part. The quality difference was amazing, however there is still room for improvement. The bottom right edge of the cube does not line up with the rest of the vertical edge and layers appear a bit offset in the extreme close up view. I’ve attached Cura pics here. I thought I’d better check my print speed in both slicers. I just looked and I mistakenly had Cura print at 120mm/s for this test while S3D was printing at 60mm/s. So, even with Cura printing much faster, results were still superior over S3D.

So, as I’d like to utilize the support options in S3D, I need to keep tackling the issue to narrow down the cause.

Do you have any experience with one slicer printing much better or worse over another using largely the same settings?

-calfer