I dont think is is wise to modify nozzles or to use spacers to accomodate a 3mm gap. Mechanically, that might more or less work, but there are thermal aspects to it too: the thermal resistance to the heat source would increase, the heatsink to the ambient would decrease with more uncovered hot surface. And that unprotected surce would catch stray plastic etc etc.
Better try to find correctly dimensioned nozzles… but then the next question is: in ‘Mk8’ defined and respected enough to guarantee that they are (with or without a Z-offset) usable in the SM2?
Yes, the nozzles are different. Is there any aspect of this that calibration won’t fix? It seems perfectly plausible that calibration values are different when you change the nozzle for a given filament, but I find it harder to believe that there’s any other issue.
I agree - calibration will fix. Additionally, like I noted above, just screw the heat break in or out and then there’s no need for a spacer.
@dvdb The 3mm gap is only between the external hex shaped flange and the heat block, and the present of a 3mm gap there is immaterial and affects nothing. The only dimensional difference is the overall length which is 0.5mm shorter. You can leave it alone and just have the nozzle be 0.5mm shorter and change your Z offset, or you can thread the heat break in 1/2 turn and have it be exactly the same distance from the tip of the nozzle to the heat block.
It doesn’t affect thermal properties very much (assuming calibration). What it does not do, however, is to seal the threads against spiral leaks of molten plastic. Better to use a spacer as a very thick sealing washer.
That’s why the nozzle needs to be in contact with the heat break - it’s a face seal.
It may be supposed to be a face seal, but in practice doesn’t seem to reliably work as one. The disastrous leakage photos I’ve seen look exactly like there’s spiral leakage of molten plastic. There’s no locknut on the heat break, for example, so it could loosen from vibration. There’s no sealing washer (it would be metal) between the heat break and the nozzle. The face of the heat break isn’t polished; I can see tool marks on the one that’s part of my spare hot end. I don’t imagine cheap commodity nozzles have polished mating faces either.
I agree with everything you point out, but I’ll mention it’s a non issue on every other 3D printer that uses the same method of assembly.
By doing the final torque at 220C+ it will lock in place just fine and does not require a locknut.
The leak pictures you refer to I offered an opinion on above: it appears the factory incorrectly assembled the nozzles by screwing the nozzle all the way down to the heat block and not torqueing against the heat break. The fix for those would be to loosen the nozzle, screw in the heat break 1/4 turn or so, loosely tighten the nozzle against the heat break ensuring there is a gap between the heat block and nozzle flange, and then finally heat up to above normal print temp, then torque in place.
I also noted above that failing to assemble in that way does not guarantee a failure, and in many cases is not necessary. There are likely many ways to assemble a nozzle that will perform as expected, however there is a “best practice” of sorts that will minimize chance of failure without requiring additional sealing components.
Inserting a copper or aluminum crush washer to aid the face seal would be an easy, cheap way to make that even better.
1 Like
I’ve never been deep into FDM printing, admittedly, but this is the first time I’ve ever seen this instruction. It’s worth promulgating more widely, particularly for people that are swapping out nozzles.
I take it you do this on the nozzle as it’s mounted in a live and heated FDM head, yes? It might be expedient but I dislike it because of the way the heating block is mounted. There’s a slot in the FDM head that mates with flats on the heat break, so that aspect is relatively solid. The heating block, though, simply floats on the threads of the heat break and nozzle. Instructions should include a step to prealign the heat break with the slot and do an initial tightening in the correct position. Then, once the hot end is installed and heated, whatever remaining rotation that happens will be small.
The thermal coefficient of expansion of aluminum (heating block) is around 20-40% greater than that of stainless steel and brass. I don’t know the exact numbers because I don’t know what alloys are actually in use, but qualitatively it doesn’t matter here. What does matter is knowing that if the nozzle is only marginally tight at room temperature, it will open up a gap at heat, because the aluminum is expanding and will tend to draw the gap apart. Also relevant is that if you use a crush washer, it should be aluminum with its higher coefficient of expansion.
This instruction should also be extended to removing nozzles, which should be done hot. If the system is sealed at heat, the force on the threads at room temperature will be quite strong. There’s greater risk of damage in this situation.
I’d recommend referencing the Prusa instructions I linked above: https://help.prusa3d.com/en/guide/how-to-replace-a-heatbreak-heatsink-heaterblock-mk3s-mk2-5s-mmu2s_16104
It’s easy to see how the design decisions that went into the toolhead make adapting these instructions for use on a Snapmaker challenging. Likely why they decided to factory assemble these hotends and sell them as an assembly in lieu of providing detailed nozzle replacement instructions.
The nozzle should be dropped out of the toolhead some distance and the grub screw re-tightened. Then the nozzle heated to 250C, per Prusa, and a wrench applied to restrain the 22 in-lbs applied to the heat block. Then cooled back down so it’s safe, then the grub screw loosened, and continue installing and aligning the hotend assembly per normal Snapmaker instructions.
This sounds like an ideal situation for a dedicated-to-the-machine tool. By way of illustration, this is a torque-wrench handle, non-adjustable fixed at 2.5 N m (22.1 in-lbs). It would need a separate nut driver head for the nozzle itself. Wiha is high-quality and you pay for it. There might be something cheaper that’s adequate; I didn’t look.
1 Like
Amazon has this for 10 bucks.
However, pretty sure its not legit wiha. the listing says wiha. judging by the picture its probably a knockoff
Curiously, slice engineering has a handheld torque wrench for nozzle installation that appeared on amazon when i was looking for that thermal paste we discussed earlier.
however, its torque setting is different.
Is there a spec someplace that says the snapmaker nozzle wrench is 2.1 nm or is that number based on the prusa information and not necessarily going to apply on this machine?
2.1 was just based on prusa - I’m sure 1.5 would be fine. I’m also quite sure that a well calibrated hand can do just fine with a normal wrench, I’ve never torqued one with a tool. Tighten til it loosens, and then back half a turn 
I saw that as well. It’s got a different model number, so it’s possible it’s legit and just a previous model number. Or not legit and a clone of an old model number.
yeah i am not one for a torque wrench but i happened to be thinking about finding a 6 mm socket so i can work easier than the open ended wrench and it seems like a good idea!
A 1/4-in driver will work, like what the bit sets use for a driver. Surely you have one laying around.
I do. good tip thanks, didnt even think to try it
What slicing software and parameter are used?
I understand Luban is only for the 0.4 nozzle.
You can use Cura, Prusa, Simpliyfy3d
lots of profiles availalbe on the forums if you search around
I bought a set of nozzles from Amazon and had to adjust the Z offest down by about 0.25mm.
Has anyone else had to do this?
Are all MK8 nozzles supposed to be the same physical external dimensos?
Regards
Graeme