this is a big one that i have been working on for a few months now (on and off). a further delay was due to me switching from onshape to fusion 360 (due to onshape's change of terms). it has required many sacrificial prototypes and a few calculations and modeling and includes a new 30mm fan adapter and a new nozzle/layer fan outlet.
new 30mm fan clip (v3.50) - with significantly improved cooling of the lower part of the heatbreak, which result in smaller retractions, less heat-creep and faster times to ambient temp.
the main improvements are:
- a more optimized air-flow cone while still maintaining a low footprint and weight, while maintaining a strong structure.
- an added venturi effect tunnel for better cooling of the heatbreak. tbh - i did not think the 30mm fan can create enough pressure/air movement for this to be effective, but i compared the same model - with and without the venturi and the results were very much improved (the complete comparison and results are below).
- indexing/insert tabs were added - so the model can now both grip the heatsink (even without a screw), and it is always indexed at the same height. i can provide a model without them - if anyone is interested (or they can be removed from the provided step file).
print settings: 0.2mm layer height, 0.4mm extrusion width, 15% infill, 4 perimeters.
!!! a left handed version will be provided as well over the next few days.
a new nozzle/layer fan outlet (v3.5L3) - with an improved airflow which is divided to different height levels based on the direction. the front part is aimed higher - just below the exit of the hotend (you want the filament to cool AFTER it has bonded to the previous layer and not before), and the sides are a more dispersed jets aimed downwards and forwards.
i recommend you print this in abs or pc/nylonx as it's longer outlets means it is exposed to more heat (i have printed without a silicone sock for testing purposes - but i do recommend using one).
print settings: 0.12mm layer height, 0.49mm extrusion width, 15% infill, 4 perimeters.
as mentioned before - i am making the move from onshape to fusion360, due to onshape's change of terms. so it might take me some time figuring out how to post the full models on fusion360 to a satisfactory degree. i still remain committed to making this fully open-source and it will be released (meanwhile - step files ARE available).
scroll below if you are interested in the results of the cooling tests performed on the various models (i have only included the select few relevant, but you get the idea).
i have added a 2 modified versions of the original layer cooling spout:
- an easier to print version, with thickened walls and a few fillets and chamfers (marked as v3.4).
- a lengthened version, which also features a lowered air flow for better cooling of the printed layers (marked as v3.41L). i recommend you print it in abs or some other high temp filament, since it is more directly exposed to heat, but the cooling performance is much improved.
print both versions with an extrusion width of 0.48mm for best results
also, due to a few requests i have added the following:
- a left handed version
- a 2mm lowered mount plate (for both right & left handed versions)
- a bl-touch mount (i have not tested it, since i do not have a bl-touch, so feedback is welcomed)
as always, all models include a .step file version for easy modification and were added and updated in the original onshape model (link below).
after encountering some problems with retractions on my e3d v6, and going through a rigorous testing process, i reached the conclusion that there is a problem with the standard cooling.
if you attach it as intended - most of the airflow goes to the upper part of the heatsink which is not very practical, as it causes the transition zone to be higher than it could/should be.
after designing a few fan holders and playing with the direction of the air - i settled on this design. it blows air to the lowest part of the heatsink and where the throat contacts the heatsink. this allows you to do shorter retracts (at least in my experience), and has eliminated any problems i was having with PLA related clogs due to the full metal path.
i then wanted to try a few different layer cooling solutions.
i started with the excellent fang by Lion4H (he was kind enough to supply me with a modded version for my earlier testing), you can check it out here - https://www.thingiverse.com/thing:2175956 and i highly recommend it for anyone trying a first layer cooling solution above all others.
my only problem with it, was that it was too focused - meaning that printing an extreme small overhang at faster speeds caused the problematic areas to receive very little wind for a very short amount of time.
so i went in search of other solutions (Diii, Ciii, prusa mk2, etc...) but none were giving me good results when printing at 100mm/s and above.
when the prusa mk3 was released i immediately started testing with it's new cooler - and after a few modifications (for better airflow, and a stronger connection mainly) i settled on something i was happy with.
being a big believer in opensource, i'm sharing the complete design via onshape - so you can basically do with it as you will (within the limitations of the license - which is very permissive). the fan holder was modeled in fusion360, and so was imported as a step file into the design. i hope to have the time to re-create the model in onshape and to then replace that current import.
you can check it out here - https://cad.onshape.com/documents/fa9d2c92fc5fa4cfced6d32c/w/724d95ccf4edcc5b3c95359c/e/ec243d25a6057a7f96d04ce0
.step files are also provided for easier modifications, and if another file format (supported by onshape) is needed - please let me know.
i also have an adapter plate for that fan holder that fits with the radial fang - if there's interest i can release that as well.
- the fan holder was printed at 0.2mm layer height, with 4 perimeters and 15% infill.
- the 5015 blower adapter was printed at 0.2mm layer height, with 5 perimeters and 20% infill.
- the modded layer cooler was printed at 0.12mm layer height, with 4 perimeters and 15% infill with an extrusion width of 0.49mm.
while the 30mm clip on for the hotend can be printed in PLA, Do Not print the nozzle/layer cooler in PLA, as i do not believe it will either stand up to nozzle temps, or be strong enough to be slotted together - PLA is just too brittle.
print this in PETG or ABS or something with similar strength (and temperature resistance).
mine are printed in PETG and have withstood nozzle temps of 270C and above pretty well for a couple of months now, YMMV...
to attach the blower bracket to the fan cooling block, you first insert the lower part of the bracket (marked 1 in the picture) to the lower slot in the fan block (cannot be marked on the image as it does not show). it is easily distinguishable because it has a straight edge and the one marked 2 has a rounded edge.
you then just press forward pushing the rounded parts together with a satisfying click. the hold is very strong.
if people find this hard to understand i might make a video, but it think it's pretty self explanatory (hopefully), to anyone who has put a 3d printer together (or is using one).
following is some information on how these tests were performed:
- the tests were conducted 3 times for each cooler and the results of those 3 tests were averaged.
- the 3 tests for each cooler were not performed consecutively, but each of the coolers was tested a first round and then this process was repeated for 2 more rounds in order to account for any mounting errors, or ambient temperature shifts.
- ambient temperature was 30C for the whole duration of the tests (the room is thermally isolated and kept at temperature. this was checked and validated by using 2 k-type thermocouple sensors across the room.
- the temperature was read by an arduino at specific intervals and validated visually - both on the printer's screen (e3d v6 thermistor), and via a k-type thermocouple that was inserted and glued (with thermal paste) to the heat block.
as can be viewed in the raw data - there is a major improvement from the default/original e3d v6 fan clip.
also, to my surprise, the venturi does seem to have a positive affect on the cooling result. the middle 2 columns are a comparison of exactly the same clip, with the single difference being the added venturi tunnel/tube.
this v3.31 was then further optimized with v3.50 being the result. it might also be improved in the future, but i am pretty satisfied with the results so far...