… with changing the printing order of each layer.
OK. I’m not a coder or anything like that, just a user.
I’m currently printing a stand for my AMS and looking at how it is printed. Currently, I’m printing a frame standing on the long side of it. The bottom is massive, but the upper side needs support. So each layer is printed like this: At first the support, then the left post of the frame, fast movement to the other side to print the right post, Layer change and it begins again.
The total print time of this part is 5 hours. Fast movement of this print is 21 minutes, 99% of this is this moving from one side to the other.
What if the slicer changes the order of support and part for each layer?
What if, the slicer would reevaluate each layer on its own? In this case, print the left post, print the support from left to right, print the right post and start a new layer at this point, so print the right post, print the support from right to left and then the left part.
Generally spoken: Include the support in the part itself, don’t print everything in the same fixed order. Print the walls once they make sense in order to safe time.
For single filament prints, this would improve the times a lot, as most fast travel times would simply vanish.
I hope you understood my crappy English and explanation.
I agree - a photo and a link to the model would be really helpful at the moment.
I understand what you meant, but there is probably a reason behind it: Support is printed at a different speed than walls. Even if you change order, the amount of time you would save would be minimal.
Avoid printing any support materials; this is the key to saving time and making things more efficient. There are settings that save you hours, not just 8%.
Cut out the object and insert joints to avoid support.
The right infill for the right print. It doesn’t always have to be slow gyroid. Adaptive Cubic, for example, is very fast and sufficient for certain objects.
Define modifier cubes to control this even more precisely
Sure, obvious things like speed, nozzle size, etc.
Of course, everyone already know all this. My argument is that there are significantly better ways to reduce printing time without compromising on quality.
As I said, I think it makes sense to tackle the big issues first, such as avoiding support, correct infill, and object cutting, before focusing on 8%*.
*I copied the number from OP, didn’t check it myself
Here is just one example of a video where someone summarizes everything I have written.
The part is on plate 2. Even on my H2C it has to stand diagonally to fit onto the plate. So it is impossible to print without support. And to make it clear: I’m not just talking about support. I’m talking about all behaviour in slicing. I can choose if I want to print the outer or inner walls first, or even start with the infill. But in all scenarios it starts to print with the support. So on many parts it moves all about the complete build plate to print the walls first (or last) and then the infill. Why can’t the slicer choose on it’s own, what part is printed next to remove unneccessary travel time? 8 % was just the estimated time on this part. I’m sure inother scenarios it can be way more time.
I’m aware that support is printed in a different speed, but changing the speed is not a real problem for the printer, right? But no matter what is currently printed, it prints. Moving along the plate is wasted time. That’s my point.
I know it’s got some holes in the back, but the AMS units partially vent their hot moist air out the bottom to whatever you’ve got in those drawers. Seems like a potential issue to me.
Within a colour within a layer - changing the direction of extrude moves or the grouping or order of sets of extrude moves to reduce travel moves is something that is possible to test out using a gcode post processor.
I’ve done a limited amount of this sort of logic to reduce travel moves by changing the direction of the next extrude move on another non BL printer and it does save a fair bit of print time.
So I understand what you’re getting at, but regardless of that, why don’t you print it without support?
Cut it up and add joints. You can then put them together later and glue them if necessary.
Then the print will only take 3 hours and 20 minutes. Instead of 5 hours. You also save a lot of filemant. That’s why I said, what good is 8% to you when there are options to save hours? Not that I am against such options, but my argument is: There are significantly better ways to optimize printing.
Here, the object is cut and joined. Stability is ensured, as the force only comes from above - and that’s what the side panel pieces are for anyway.
Another advantage is that if you want to print in multiple colors, only a small part of it is affected, or you can even print just that multicolored part. In the final print, I would only use more and bigger joints instead of as in the example image.
Thanks for the tip with the cutting tool. I really have to go more into detail with that slicer.
I’m glad, you got my point. If anyone could make this a feature, that would be great.
