Better cooling algorithm for small parts

[duaneh]

The current Kisslicer has the ability to specify a miimum time per layer which slows down the head if a layer is otherwise going to be too fast. This doesn't really help all that much, however, because the head remains in contact with the piece and keeps it hot (and fluid). Thus the next layer is printed on a hot, wobbly previous layer, and the whole thing becomes a mis-shapen blob.
A better solution would be to print the layer at the normal speed, then move the head away from the part for the balance of the layer time, thus allowing the part to cool and leaving a firm platform for the next layer to be printed on. If the printer has fans, moving the head so that the fans are over the part would speed up the cooling process, but just getting the head away from the part for a time would make all the difference.

[plexus]

I find the "slow down on small areas" feature to be excellent. so often I have a print that ends with a small area at the top. without this feature I'd have to manually slow the printer down to finish the print. the other option is to use "prime pillar" which prints a pillar along with the model so you get what you are talking about: the printer alternates between the actual print and the pillar. however this really slows the print down and consumes material and results in having to dispose of more plastic. so I use the slower speed on small areas feature first along with active cooling (fans) and if that is not enough I will use prime pillar. thing is, I've never had to use a prime pillar. so I am on the opposite side of the fence as you in that I think the slow down feature is excellent and addresses a major printing issue.

try using prime pillar. it will do what you describe but print a pillar in the meantime. i think just moving the nozzle over, waiting and then back to printing wont be effective because of ooze.

[duaneh]

I use Prime Pillar, but it is still WAY too fast for, say, a 3mm post 1" high. If I could get it to do 3 prime pillars that would be about right. You are correct that the ability to slow down the print is helpful, but for very small vertical features it just doesn't do enough, even with prime pillar.

[PenskeGuy]

This situation was discussed many times on the original forum. All suggestions were ignored.

My experience is the same as the OP. The current method of slowing the feed rate when printing narrow cross sections is counterproductive. Nice idea on the surface but not exactly sound, when examined in more depth. Far too much conduction is happening when the head stays on the print for a longer time.

"Air Printing" was suggested. To counter the ooze, a very minimal, continuous retract while the orbit was happening would cure it easily. If anything, slowing the high speed feed rate between sections and keeping the printing feed rate constant would be a better "slow down" approach. Get the head off the part in a normal amount of time and move to the next location at a more leisurely rate; thereby allowing the printed part to cool more before the next application... partly because it won't be as hot in the first place.

As is noted, the Prime Pillar feed rate is too fast and the operation is over too soon to have allowed sufficient part cooling. Only thing that I have found to be reliable is printing multiple parts, sometimes invoking the Min Layer Time right at the very tip of a conical cross section. Even then, I can watch the tip wobble around following the head as it moves. Not solid enough, but it seems to null out by the time the layer is complete on that portion.

[duaneh]

That is actually one thing CubeX got right. Their slicer would move the head away from the part for 3-4 seconds (I don't recall any ooze problem) whenever the part got too small. It worked great. There are a lot of things i didn't care for with the Cube slicer, but I have to give them credit for this one.
Hopefully Jonathan will consider adding this one - it would make such a big difference on parts with small features.

[plexus]

Yes, I agree too. I just printed a part using KS that had thin walls. one wall was under my set 15s layer min and so it printed very slow which actually messed it up. it was long and narrow. i do find the min layer print time reduced speed to he helpful when the area is more regular (X and Y about the same) but it appears when the area is thin, as you said its counterproductive.

[duaneh]

Yes, I have found that the only way to get a good print is to print a 'sacrificial' part such that the minimum total area of any given layer is > 6-10mm2. Not a great solution, and it wastes a lot of filament, but it does preserve the integrity of the part.