eventhorizon
Posts: 2
Joined: Tue Dec 17, 2019 5:07 am

Finally found the caused of so many bad prints??

Tue Dec 17, 2019 4:06 pm

Greetings all, I searched and did not see this issue reported, though I did see one which may be related that has been closed. I wanted to share this in case others may be suffering from what I was and not even know it.

Normally I trust my slicers on the nit level details of the slicing, so I do not tend to walk through every layer before a print (maybe I should), but recently I noticed a strange/problematic behavior. I have always used the drop to bed feature after importing a model, in the past and typically it will drop the model onto the plate from above and let it settle as gravity would have it (if it's above the mat, or askew slightly it will fall flat). When working on my own design though, I noticed the slicer was making a strange out of place line on the exterior wall. A vertical path was there, and it did not need to be. Example 1 below shows a normal 90 degree line as expected. Example 2 highlights something new in this layer (the previous layer did not have it) which is a strange ~83 degree line appearing out of nowhere in a space that nothing in my model should be causing it.
example1.png
example_part2.png
Looking VERY closely at the build plate, zoomed way in I was able to see what's happening. In the last example you will see that the object is a microscopic amount off level, I couldn't even see it naked eye without zooming this far in and lining the horizon up exactly. This behavior causes the slicer to make a LOT of strange/poor decisions, and while the prints usually work and you are not aware of it, internally t hings are messed up, to the point that I have had structural problems etc. because some layers would be half baked spanning 2 layers when parts should be all on the same layer.
example_part3.png
The workaround? If I use the place surface on bed option and pick any face on the bottom, it works as expected and the entire model lays flat. Looking at the same slice afterwards the stranger lines are gone, the surface is clean, and my print time has gone down. I am not sure how long this has been happening, but I now know why some of my previous print jobs went wonky on me. I hope this helps anyone else if they are pulling their hair out over inconsistent print results too!

To devs, I assume this is not the desired behavior of the drop to bed function? If not, would love to see a fix in the next version :)

TheBum
Posts: 118
Joined: Fri May 18, 2018 11:13 pm

Re: Finally found the caused of so many bad prints??

Wed Dec 18, 2019 12:46 pm

Drop to Bed has never adjusted angles of an object. Where it does come in handy is if you're manually rotating an object such that it ends up embedded in the build plate. In that case, Drop to Bed will "lift" the object to the top of the build plate without rotating it further. Also, there are many cases where you might not want an object flat on the build plate.

arhi
Posts: 424
Joined: Thu Oct 06, 2016 5:13 pm

Re: Finally found the caused of so many bad prints??

Wed Dec 18, 2019 5:01 pm

CTRL+L to select what face you want to be on the bed
gcodestat integrates with Simplify3D and allow you to
Calculate print time accurately (acceleration, max speed, junction deviation all taken into consideration)
Embed M117 codes into G-Code
Upload your G-Code directly to Octoprint
open source and unlicence

eventhorizon
Posts: 2
Joined: Tue Dec 17, 2019 5:07 am

Re: Finally found the caused of so many bad prints??

Thu Dec 19, 2019 2:59 pm

Really? So all this time I have just assumed it was adjusting anything that should be adjusted in the event of a "drop". So it sounds like all it really does is put the lowest point of the model in contact with the build plate? That is extremely misleading, at least to me, which is exacerbated by throwing in a actual "drop" animation to boot. Dropped things settle when gravity is introduced! :( That does at least explain it though. while I agree that you might not always want a full surface flat on the plate, I do often simply want it "on the ground", and gravity is a pretty efficient way of doing that without having to select any specific element (hence my assumption) I must have been REALLY lucky at hand-orienting askew models. Well, that and the micro-off-leveling does not make all prints fail, they just manifest strange methods to the end result. After experimenting more most models come out fine, but if you do a layer by layer inspection you will see strange internal or surface aberrations. In my example above there was no way to tell it wasn't flat on the table to the naked eye, it took quite a lot of zooming in to finally see it, but that little bit yielded a really poor final layer and the build time was quite a bit longer.

Cest la vie, I can honestly say operating "as intended" was not what I expected! At least now I know, and hopefully this thread will help others who might have been under the same assumption, or have a problem they haven't been able to figure out with odd printing behavior! :)

As a result I think I will propose a feature request, 3D Builder and other apps have a nice function called "settle" which does basically what drop does, but it lets the model settle on the plate as it would if it were in the physical world and oriented in that way (e.g., in my model it would have dropped it to the lowest point on the model as Drop did, and it would then fall to the table to remove the unnatural "hovering" of the up-end. While I know the slicer is not a "builder" I think the function fits the print-prepping use case quite well.

Thank you answering the question and educating me on the function! :)

Return to “Troubleshooting and Bug Reports”