So ... 3d Printers....

[RandyT]

Nice parts!

I decided long ago that I would no longer consider a bed slinger machine to be viable without automated mesh leveling and a dual Z-axis.  Additionally, my Ender3 V2's all got upgraded filament feeders mounted directly above the hot-end assembly and that crazy Bowden tube was cut down to a couple of inches of filament guide tube.  This makes it a quasi-direct extruder.  Since doing this, prints have been much more uniform and reliable and I have only had to replace a nozzle once due to a severe clog, and that was probably due to overcooking the filament during prolonged pre-heat cycles (don't do that )  I've literally yet to have a failed print otherwise.  The modifications increase the cost of the printer pretty substantially, but IMHO, they are worth every penny.  Also, touch screens are over-rated for displays that small. They just limit the amount of information one can display or make it difficult to operate

Corners can be weird on a lot of automation processes.  Especially when higher speeds come into the equation.  With CNC cutters, it usually manifests as burning of the material, or rounding of the corner to account for the abrupt change in direction at speed.  It's all physics.  The heavier the head and the faster the change in direction, the more tricky this can be to overcome, as the travel must slow down.  As you found, it seems that in the case of 3D printing, the effect of the ramp down to the corners results in heavier extrusions/drool at those features.  Ramp speeds/acceleration/deceleration settings can be altered to give an acceptable result, but sometimes the design of the part is better altered a bit to avoid the problematic features if possible.

It's definitely nice that you got the filament change command to work.  But another way to do that for those parts would be to just have a black program and a green program.  It would require generating the g-code for the complete part and from that, creating separate files for each color through some manual editing, but it would negate the need for any babysitting of the machine.  Maybe something for others who want to do this type of thing to keep in mind if that capability isn't already present (or working properly) on their machines.

[lanman31337]

Over the holidays I upgraded my Ender 3 - Dual Z drive, direct drive, new allmetal hotend, printed a fan shroud. I bought an Ender 3 V3 SE, and purchased an HP 705 G4 to run multiple instances of Octoprint and cameras.

[pbj]

Over the holidays I upgraded my Ender 3 - Dual Z drive, direct drive, new allmetal hotend, printed a fan shroud. I bought an Ender 3 V3 SE, and purchased an HP 705 G4 to run multiple instances of Octoprint and cameras.

Can it print pirate figurines?

[05SRT4]

Can it print pirate figurines?

I need to dust off the ole resin printer, whatch ya need?

[lilshawn]

all you bambu owners with an A1

[pbj]

Screwed right in.

This part is unobtainium, other collectors I know are happy to see this come to fruition.  Found out one of my mosfets is blown on my $300 repro MPU.  But that’s another thread.

Well, well, well...

https://www.siegecraft.us/presta/index.php?id_product=79&controller=product&id_lang=1

[Ond]

Never thought I'd post in this thread. Stuff happens, things change. So I have a side hustle to my part time job  (I am an Access Technology Specialist) working in a electronics hobbyist shop generally one day a week. There we sell a large range of things, you name it, everything from Raspberry Pi stuff, Arduino, security cameras, retro gamer hardware and .. 3D printers.  To sell these things I at least had to learn something about them. We have variety of demo units including the Ender 3 V3 SE as well more expensive models, both filament and resin. Point is, I'm printing stuff off in the shop on the Ender 3 which just happen to be for my Vpin  .  I know quite a few folk use Fusion 360 for their design work but I prefer 3ds MAX just because I've been using it for so long.  Apparently the Creality Ender-3 V3 KE  is the sexy-go-to machine right now, I'll have a play with it and see...

[Ond]

Even on superfine mode filament printers are still coarse to my eye. But this print will be refined quite a bit for a project.

[Fursphere]

Even on superfine mode filament printers are still coarse to my eye. But this print will be refined quite a bit for a project.

Use a smaller nozzle, like a 0.25mm.   However, this makes print times increase dramatically on larger parts.  You can get down to 0.07mm layer heights, which is almost resin-like quality. 

[BadMouth]

https://ellis3dp.com/Print-Tuning-Guide/articles/index_tuning.html

3D printing is an endless progression of "could be better" and diminishing returns.

...and there are limits to how good an fdm print can look.

[Ond]

Even on superfine mode filament printers are still coarse to my eye. But this print will be refined quite a bit for a project.

Use a smaller nozzle, like a 0.25mm.   However, this makes print times increase dramatically on larger parts.  You can get down to 0.07mm layer heights, which is almost resin-like quality.

Thanks, yeah that print was done with a standard 0.4mm nozzle which is what our shop demos run. We have resin printers on display but work health safety rules mean they don't actually get to print anything.

https://ellis3dp.com/Print-Tuning-Guide/articles/index_tuning.html

3D printing is an endless progression of "could be better" and diminishing returns.

...and there are limits to how good an fdm print can look.

Ha hah, well filament printers have their uses I guess. By the time I'm finished with that print you wont think it came from one.

[lilshawn]

honestly, i don't really care what it looks like, as long as the part the comes out of the printer is functional.

i mean, yeah i print out decorative items once in a while, but for the most part i'm printing replacement parts for something broke, or a part to achieve a function. if it does that function, it's good enough for me.

my first layers could be better, my attention to preventing warping could be better, my corners could be square-er, my elephant's foot is very slight, but i trim the edges anyway... i occasionally get under extrusion on my outer layers...but also really, why waste the time and effort and filament testing out 2000 different fixes for stuff when it's "good enough" for what i need.

as long as my 3d print can hold up the rod, or hold the circuit board, or skim the scum, or drain the water... good 'nuf!

picrel... new output board adapter endplate for an amplifier. one that no one will ever see but me when i plug the stuff into it and then basically never again until it breaks... if it's ugly...who cares... if it's half ass structurally sound and it's holding the board... 

[RandyT]

It is possible to make pretty parts with FDM, but it involves printing with the loathsome ABS and finishing with chemical vapor.  I experimented with this a bit.  It's a fair amount of work and really only good for one-off "artistic" works, and that which does not involve fine details.  IMHO, resin printers are a better choice for pretty things, so long as they need not be too tough.

But for relatively strong and functional parts, it's hard to beat FDM.  I would venture a guess that today there are millions of little 3D-printed pieces of plastic in the world, doing jobs that otherwise would have been impossible for the amount they cost to produce.

[Ond]

Like I said, FDM printers have their uses.. We get some customers coming in that print simple functional parts for things and the only thing they care about is "how fast can the new model print". Those same customers are only printing in the toughest plastics as well.  A few of the guys in the shop have their own resin printers so I get them to print me the occasional "pretty" things when needed.

I just wanted to be able to understand the end-to-end process of design to print stages in 3D printing, at least enough to speak convincingly to customers anyway.  Finishing techniques for printed things like Randy mentioned interest me. Not just to smooth prints but paint them as well. Anyone got any examples of painted stuff that looks slick? For example, I see really nice resin printed figures but I never see any painted or detailed?

[nitrogen_widget]

had a watermain break in the basement last month.
Happy New Year.
printers survived but i had to fit a basements worth of crap into 3/4ths of the basement to make room for contractors.
all fixed.
cost more than i could of ever imagined.

so no printing for a while until i throw stuff out and make a real shop.
not going back to an old bench with a yoga mat on it and a cheap table top green house as an enclosure.

going to build more shelves and throw more stuff out so it looks less like a storage locker with a path and more like a shop.
lol.

[Ond]

I've not had a good week.  It doesn't really matter why.  *sigh* Anyway, I'm thinking of buying me my very own 3D printer!  I watched a Creality K1 working at speed in the shop the other day, very impressive!  A bit pricey though, so I think I'll get me a V3 KE.  They seem to be the best value for money. Has anyone got one?  Are they good? Alternatives?

[RandyT]

I've not had a good week.  It doesn't really matter why.  *sigh* Anyway, I'm thinking of buying me my very own 3D printer!  I watched a Creality K1 working at speed in the shop the other day, very impressive!  A bit pricey though, so I think I'll get me a V3 KE.  They seem to be the best value for money. Has anyone got one?  Are they good? Alternatives?

I don't own one, but the V3 KE seems decent considering the price.  The important bed-slinger upgrades, like auto-bed-leveling, direct extruder and dual-z-axis drive are already there.  Those alone represent over $100 in add-ons for one of their lower tier offerings.  There are other niceties, like Klipper out-of-the-box and the slide rails as well, although the value of those things depends on your usage profile.  Klipper can improve the appearance of parts printed at high speeds, but if appearance is important, slower speeds almost always yield a better result, especially where materials other than PLA are concerned.  And the value of the rails is also a bit controversial.  They add weight, but also rigidity to the beam and are less likely to wear out over long and hard use.  That stated, the regular wheels are very accurate, durable (when not overtigthened) and are dirt cheap and easy to replace if ever necessary.

Normally, I would shy away from units with special nozzles.  However, since this is a Creality machine, I suspect that they will be supported for a reasonable time to come, and it helps that there are several models which use the same nozzle design.  Just be aware that they are a consumable and that they cost more than the typical nozzles, so you'll likely be tempted to try to fix one (often times unsuccessfully) rather than tossing it and using a new one.  Just factor in the cost of the replacements and have plenty on the shelf so you won't need to in a pinch.

But the consensus seems to be that they are decent machines, and Creality has a reasonably good track record compared to other brands. So for the price, and more importantly what you get for it, I think it's something I would buy as a first 3D printer as someone who has already been exposed to them and intends to use it productively.

[Ond]

Thanks for the info Randy, I ended buying the Creality V3 KE. My discount to myself was generous 

Reprinted a few parts for my project with some improvements.  I'm happy with the results:

[RandyT]

Thanks for the info Randy, I ended buying the Creality V3 KE. My discount to myself was generous 

Reprinted a few parts for my project with some improvements.  I'm happy with the results:

Spiffy!  So, on the webs, did you tweak some things to deposit like that or were there more manual artistic shenanigans used?

[Ond]

Thanks for the info Randy, I ended buying the Creality V3 KE. My discount to myself was generous 

Reprinted a few parts for my project with some improvements.  I'm happy with the results:

Spiffy!  So, on the webs, did you tweak some things to deposit like that or were there more manual artistic shenanigans used?

Thanks, that web print is as per the mesh that I designed:

Basically started with circles and line type objects and injected several noise modifiers to get the look.

Something interesting is extracting game meshes, smoothing them into high-res models, then outputting to STL for printing. A complex mesh would best be divided up into several parts to reduce supporting structures and printing complexity. Here's an example I extracted just for fun (legalities aside  - Armor frame from Fallout 4:

[lilshawn]

i build wiff squars and sirkuls an stuff in tinkercad.

real cad is 2 hard

[RandyT]

Thanks, that web print is as per the mesh that I designed:

I considered that they might be fully modeled, but my brain recoiled at the possibility. It does look nicely organic!

i build wiff squars and sirkuls an stuff in tinkercad.

real cad is 2 hard

I was fortunate enough to be exposed to parametric modeling at my last 9 to 5 (over 20 years ago...oof) and it was a slog.  When it was being explained to me and while watching someone use it, I could literally feel my synapses re-routing and it was painful.  But one day, as if by some sort of biological sorcery, it just "clicked" and it was downhill from there.  Those learning curves can sometimes feel like a straight vertical line, but they are surmountable.  The tools have also become quite a bit better since then.

That stated, you might want to take a look at Plasticity.  It's CAD, but meant for those who might feel more at home using something like Blender or those who like modifying primitives (squars and sirkuls an stuff..lol.)  There's still a learning curve that my brain rejects due to the parametric modeling programming I received, but it looks very capable in the hands of someone more accustomed to more artistic ways of designing things.

But on the free side for local installs, there are basically only a couple of decent options: FreeCAD (and it's forks) and Blender.  FreeCAD has it's own ways of doing things, which parametric modelers who are already trained seem to despise, but if you are going in cold, there's no reason not to learn it.  Blender, as you probably know, is an insanely capable beast and there is a plugin in the works for doing CAD functions within it.  With all that capability, the learning curve is particularly steep. It's a 3D modeling application, so there is a ton of stuff not particularly aimed at functional design work.  So unless your interest is in the artistic over the functional or you already have experience in that realm, I wouldn't recommend that direction to someone who just wants to design their own parts.

[Ond]

I considered that they might be fully modeled, but my brain recoiled at the possibility. It does look nicely organic!

I was tempted to make them even more bumpy and irregular but settled for the result you see. 3D printers are such a natural progression from 3D design. I guess it's taken me this long to acquire one because I wasn't that impressed with them in the early days. I have a few build tasks coming up which will need printed parts. I'll post the results here as well.  I'm also doing a larger articulated mech spider with more detail, just for fun! Happy to share the STL when its ready.   

[nitrogen_widget]

Dipping my toes into klipper.
on my free ender3pro.
it's been an experience getting everything loaded and setup and actually printing.
understanding print.config, macros, and tuning the options for better prints.

not expecting any type of speed machine from this stock e3p.
this was just to learn it and get a printer going again.

and there is a lot to learn.
but once i have it down i have a number of dead printers i can resurrect and upgrade to print with again.

[BadMouth]

Dipping my toes into klipper.

If you are the type to tinker and change things there is no comparison with Marlin.
Just change a config file and reboot (actually just restart klipper, not reboot the pi) vs compiling and reflashing firmware.
Much easier to understand than RepRap, at least for me.

The only complaint I have is that every once in a great while an update to Klipper or Linux will break things and I'll have to make changes to Linux/Moonraker or revert to an older install to get the printer working again.  It isn't that bad if you keep a backup to restore.
If you get it set up and never update it, you're golden. 
Pretty sure one of their goals for this year was to improve such situations, but I probably haven't updated in a year.

[nitrogen_widget]

i'm liking klipper.
it was a real pain at first wrapping my head around everything.
now when i hit start my printer homes, does a probe just on the parts of the bed something is printing on then spits out a purge line and gets to work.
i'm able to get 60mm/s external perimeters, 80mm/s internal, and 100ms for infill and supports with much better quality than originally.
i let klipper handle accelerations.
also using stock screen which is nice. nothing graphical and it's knob controlled instead of touch but it's does what i need.

what i found is any faster than current and the poor machine starts shaking itself to death.
so i didn't even bother with input shaping.
running a .6 nozzle and quality is pretty good for miniature terrain.

i have prusaslicer uploading directly to the ender and i use an old chrome book for control.

next is my longer lk5pro with the bad board.
i have a btt pico board with an rpi zero 2 and the rpi touch screen to put into it.
printed new screen case and need to design new board mounts for inside the printer.
new cooling printed up and adding bltouch.
again not looking for speed just want another working printer, though like the ender i expect to get some speedier prints out of it and plan to use input shaping on this one. it's a little newer and i think it can handle at least 100mms external perimeters.



i have a few more printers dead i plan to put klipper on also.

i don't believe klipper is the answer just for the sake of klipper but when your printer has a fault and blows a mainboard for the stupidest reasons its time to rip that stuff out and go your own route and now that i have some macros and understanding on configuring why wouldn't i go klipper?

[BadMouth]

The rpi2 can handle running klipper and input shaping, but asking it to also run a camera might be too much for it.
I am using one on a 150mm Tiny-M without a camera or screen.
It runs something like 28k accelerations, but it outruns the hotend past 150mm/s.
I don't use it much because it isn't enclosed, but have never had an issue with the rpi zero 2.

Once you get comfortable with Klipper, give a CAN toolhead board a try.
If you have printers that break toolhead wiring over time, it's only 4 wires to deal with instead of a big bundle.
There are some newer USB solutions as well.

[nitrogen_widget]

didn't know a can breakout board existed. thanks!

so found my rpi screen won't work on pizero2 because no spi connector.
going with a btt pico and btt pi 1.2 plust tft 3.5 for my longer when i get a chance to drag it out of the basement to look at mounts so i can print up brackets for the board.

i really need to go corexy i think for significant speed.
friend of mine's kid picked up a K1s and it's damn fast.

[BadMouth]

i really need to go corexy i think for significant speed.
friend of mine's kid picked up a K1s and it's damn fast.

They are great up to a 300mm build plate.  Much bigger and the length/stretching of 6mm belts prevents you from cranking up the speed and acceleration.  I only print 65mm/s perimeters at 4.6k accels on my 350mm V2.  I can push it to 90, but the prints don't look as nice.  9mm belts would improve things, but I haven't gone down that road yet.  With the price of the Bambu and K1, it doesn't make any sense to build your own at this point unless you already have 90% of the parts laying around.

[nitrogen_widget]

I tried getting more speed out of my stock ender3pro.
yeah. i can do 60mm/s external perimeters but is shakes the printer like crazy and i get clogs on solid infill for speeds over 40 and over 30 extruder clicks like crazy.
poor little guy. started stringing on me also. lol.

i'm gonna dial him back down and just let him print on his own time.
i was going to upgrade the hotend but it was a free box of parts that looked like it was tossed down the stairs in anger. the fact it produces nice prints at all is a miracle.

[RandyT]

I tried getting more speed out of my stock ender3pro.
yeah. i can do 60mm/s external perimeters but is shakes the printer like crazy and i get clogs on solid infill for speeds over 40 and over 30 extruder clicks like crazy.
poor little guy. started stringing on me also. lol.

Yeah, I gave up trying to rush 3D prints a long time ago.  It's hard enough to get repeatable results even when running at slower speeds.

My opinion on speed is probably not a popular one, but if you need lots of parts quickly, the better answer is probably to buy more printers.

[nitrogen_widget]

I tried getting more speed out of my stock ender3pro.
yeah. i can do 60mm/s external perimeters but is shakes the printer like crazy and i get clogs on solid infill for speeds over 40 and over 30 extruder clicks like crazy.
poor little guy. started stringing on me also. lol.

Yeah, I gave up trying to rush 3D prints a long time ago.  It's hard enough to get repeatable results even when running at slower speeds.

My opinion on speed is probably not a popular one, but if you need lots of parts quickly, the better answer is probably to buy more printers.

they print as fast as they print. i'm printing large structures that take up a good portion of the bed.
it's not going to be done quickly.

however, i really do get good prints consistently from my printers with .2 layer heights.

[lilshawn]

I tried getting more speed out of my stock ender3pro.
yeah. i can do 60mm/s external perimeters but is shakes the printer like crazy and i get clogs on solid infill for speeds over 40 and over 30 extruder clicks like crazy.
poor little guy. started stringing on me also. lol.

i'm doing 120mm/s infills, 60 on outer walls and 80 on inner walls. (i do a lot of slight overhangs and slower speeds keeps them from getting curled up.) I was getting the very start of extruder skips at 120infill/100outer/100inner. random clunks every 5 or 10 minutes... so RIGHT on the edge.

if you still have the stock extruder, I recommend removing the screws attaching the hotend to the heatsink... all it does is suck heat out. as long as you don't crash the head doing something dumb like trying to print without a bed level sensor of some kind, you'll be fine with just the support of the heatbreak tube. doing this really help stabilise the head temperatures... especially when you really start pushing the filament in when i start getting impatient for a rough test part and i'm dialing up the the print speed to 150% over my 120/60/80mm speeds.

as for your stringing. you might have to tweak your retraction distance up 0.1 or 0.2mm. or possibly lowering your print head temp a couple degrees. (try -3 degrees to start)

printing out a couple of the available stringing test prints (ones that do the automatic variable retraction amounts and one with varying head temps already in the gcode) will help you narrow down the issue and what you need to tweak.

I switched to almost buying exclusively PLA+ when i need new material and been running it at 210c no issues with extruder slip/stringing. PLA+ I have, says it's good 205 to 230c and i just picked 210 since i also run straight PLA and the PLA seems quite happy with that temp and the same settings i've tweaked to... so i don't have to bother with remembering to change settings when i change materials. So I feel I COULD push it much faster, but honestly... it's already like 8 hours for a print... what do i care if i could get it in 7 hours 45 minutes. i'll be sleeping when it finishes anyway.

[nitrogen_widget]

the extruder motor on my ender3 is failing.
it's jumping around and clicking and it isn't a clogged hot end either because it will keep printing making noises.
i've double checked e-steps, reduced flow, turned off pressure advance, and slowed solid infill speed to a snails pace and no love.
If i raise the hotend temp past 220 i get heat creep and a clog.

print quality went downhill also.

so since i need a new extruder motor and better heatsink cooling I ordered a sprite pro direct-drive hotend.
I like direct drive.
i have a few printers with it and they give me less problems when it comes to clogs.

[RandyT]

the extruder motor on my ender3 is failing.
it's jumping around and clicking and it isn't a clogged hot end either because it will keep printing making noises.

The nozzle is most likely clogged.  This is exactly what mine does when I need to replace it.  Keeps printing, but clicking and poor quality.  A new nozzle fixes it every time.

What happens is the cooked polymer builds up around the inside edges of the filament melt zone, leaving an increasingly narrower tunnel down the middle.  The longer a plastic cooks, the harder it gets and the less it can flow.  Eventually, it will reduce flow and heat reaching the filament to the point that the extruder can't push enough down the tunnel to do it without stalling the motor.  Unless your new extruder uses different nozzles, it will still happen eventually.  At least based on what I have seen with my quasi-direct extruder setup.

[nitrogen_widget]

the extruder motor on my ender3 is failing.
it's jumping around and clicking and it isn't a clogged hot end either because it will keep printing making noises.

The nozzle is most likely clogged.  This is exactly what mine does when I need to replace it.  Keeps printing, but clicking and poor quality.  A new nozzle fixes it every time.

What happens is the cooked polymer builds up around the inside edges of the filament melt zone, leaving an increasingly narrower tunnel down the middle.  The longer a plastic cooks, the harder it gets and the less it can flow.  Eventually, it will reduce flow and heat reaching the filament to the point that the extruder can't push enough down the tunnel to do it without stalling the motor.  Unless your new extruder uses different nozzles, it will still happen eventually.  At least based on what I have seen with my quasi-direct extruder setup.

i'd agree if that was the case. i've replaced the nozzle, taken the heat block off, checked the inside of the heatsink was clean, verified ptfe end was not restricted, ect. and it still skips on the solid infill and only the solid infill. also if i print pet-g then i def. get a clog. if i print pla around 220 i def. get a clog. could be the extruder, maybe the motor, poor cooling of the hotend? running a minimus OG with a 4010 fan. it's blowing at the heat sink. it's been progressively getting worse the last few days no matter how many nozzles i swap in and i'm running .6mm nozzles so so clogs should be minimal. but now it just starts clicking and creaking right from the start.


i'm tired of screwing with it so ordered the direct drive sprite pro. it's a bolt on.

[RandyT]

i'd agree if that was the case. i've replaced the nozzle, taken the heat block off, checked the inside of the heatsink was clean, verified ptfe end was not restricted, ect. and it still skips on the solid infill and only the solid infill. also if i print pet-g then i def. get a clog. if i print pla around 220 i def. get a clog. could be the extruder, maybe the motor, poor cooling of the hotend? running a minimus OG with a 4010 fan. it's blowing at the heat sink. it's been progressively getting worse the last few days no matter how many nozzles i swap in and i'm running .6mm nozzles so so clogs should be minimal. but now it just starts clicking and creaking right from the start.

Ok, those things weren't mentioned. 

Could be a thermocouple or heat block issue as well.  I don't know why just pet-g would result in a clog, other than that it usually needs higher temperatures.  It almost sounds like the heater block isn't coming to temp, but the thermocouple is telling the controller that it is.  I.e. not a clog, just the heater isn't maintaining the melt temps.  Or the motor could be weak, but that isn't usually what happens.  The steppers usually either work or they don't, which makes sense, because a motors electrical parts are just windings which short or burn open, without a lot in the middle.  Also, it seems that those who have run into this issue seem to have always found other solutions to the problem which didn't involve a motor replacement.  A weak motor will also get pretty hot while in operation, as it's turning that energy into heat instead of force.  One individual who had this issue had a cold motor and found that it wasn't getting the correct power to drive it.  His solution was to disconnect and reconnect the motor cable to the PCB, as some light surface corrosion had caused a poor connection.  So a check of wires and connectors is a good idea as well.

But Bowden extruders just suck in general.  IMHO, they are a holdover from the DIY olden days which needs to be forgotten.  Replacing it with a direct extruder would be the best first step for any printer.  Worry about the other stuff afterward if you still have problems.

[nitrogen_widget]

i think i assembled it wrong. the nozzle and heatbreak are down further than usual.
everything is tight.
but it's down a few mm's more than usual and i didn't catch it because i always re-zero the hotend when changing nozzle.
could a few extra mm's make that much of a difference?

idk.
but my direct drive came in so i'm putting that on.

[RandyT]

i think i assembled it wrong. the nozzle and heatbreak are down further than usual.
everything is tight. but it's down a few mm's more than usual and i didn't catch it because i always re-zero the hotend when changing nozzle.  could a few extra mm's make that much of a difference?

A few mm is a lot to be off.  If that is creating  a cool zone prior to the nozzle (due to the fan blowing on an exposed area or something) it certainly could result in extrusion difficulties.  Typically, the nozzle is butted up tight to the heat block and at least on mine, further insulated with a silicone boot right to the tip of the nozzle.  Heat will still be conducted to the nozzle, but probably not as much as you think, as the temps are measured directly from the heat block.

But I would have to see where the length gain is occurring to have an idea what's happening.

idk.
but my direct drive came in so i'm putting that on.

(where's the thumbsup emojii?    )

[nitrogen_widget]

because it was bugging me i looked closer and got out some cut strips of tissue paper and found that because of the slightly lower position of the tip of the nozzle the part cooling vents were pushing the strips of tissue onto the nozzle.
so i think i was cooling the nozzle.
not scientific but enough for me to say "that was probably it, now bring on the direct drive"
i'll try to finish it this weekend.