N2 Single Extruder 2016 - Part Cooling

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Wed Mar 18, 2020 5:57 pm

So, I have never changed the stick part cooling fan on my N2. Unfortunately im seeing really bad results with thin/small items like this "rod" that prints with other items at the bottom but alone towards the top. I want to get a better part cooling fan. Is this thingiverse cooling part from 2016 the best option? https://www.thingiverse.com/thing:1960676
anyone have others? Anyone want to share the best part cooling for the N2 Single Extruder? :)

Screenshot 2020-03-18 13.53.58.png


Screenshot 2020-03-18 13.54.05.png


Screenshot 2020-03-18 13.54.11.png

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Thu Mar 19, 2020 1:06 pm

While you can change the cooler, let's not kid ourselves on this specific print.
What you are fighting is going to be near impossible for a fan cooler to fix.

Now that I've blasted you out of the water, let's understand why I'm saying that and what the real fix is. Just look at your object. The nozzle is barely moving on a per layer basic, lifting, and then printing more. Between radiant heat, the slow rate of movement and extrusion, the previous layer NEVER has a chance to cool. Further, to think some fan also blowing at around the nozzle if going to fix this is just a sad thing. Think about it, you either end up cooling the nozzle or the air is heated by the nozzle negating the effect. Worse, even with insane amounts of air, air cooling that tiny of a space simply does not work.

The other thing going on is your slicing creating this gcode. One of the things is that slicers long ago implemented some bad ideas. One example is minimum layer time and a concept of slowdown. So if you only print this tiny vertical object and that's if for that layer, then because the short distance, the slicer then slows the speed of movement and extrusion down to take a minimum time. Well on some objects that might work, but once we get to a certain minimum size, now we dwell the nozzle and all that radiant and conductive heat down, just cooking that layer. Again, no fan on the planet is going to solve bad gcode. Even then, let's say we speed it up. Well, given the small size, even that has limits right?

What you need to do is identify these types of critical prints, place either more objects or sacrificial objects in the slicing path. This way, the nozzle moves away from the layer thus allowing fan cooling to actually work, and fixes that layer minimum time because the nozzle has more distance on a per layer and more moves. Again, something as simple as saying I need one, let me print 4 takes it from junk and a waste of time to success and extra parts with minimal waste. And no insane modifications required that still don't solve the problem.

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Thu Mar 19, 2020 1:09 pm

And, the best cooler for this type of print if you do go down the modification path is the Berd air cooler, and that's because it provides high pressure airflow directly surrounding the nozzle with minimum nozzle cooling. But to implement the berd on a N2 is a non trivial exercise and could require custom firmware and wiring- beyond the mounting of the mod.
Last edited by Jetguy on Thu Mar 19, 2020 1:23 pm, edited 1 time in total.

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Thu Mar 19, 2020 1:13 pm

And, based on this, you might need to keep some form of stock fan + the Berd cooler https://forum.raise3d.com/viewtopic.php ... erd#p40598

I found my post and pictures on my install https://forum.raise3d.com/viewtopic.php ... erd#p33477

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Thu Mar 19, 2020 3:44 pm

Again, the root cause of this specific print failure of something this small and tall with no other adjacent objects is the slowdown and dwell time caused by minimum layer time. Even if you turn this off, it might help, but printing more copies of the object is actually the best fix in most cases.

To understand what the value are and what they mean is pretty simple.
When enabled, the 15 sec minimum is a threshold for this rule. You can change that to whatever value then it invoke the rule more or less often given the variable nature and shape of prints.

So, when the minimum layer time isn't met, then the slicer invokes the minimum speed setting for those segments.
Example is 10mm/s. You could raise that value, but again, kind of defeats the point. What we have is a rule that is supposed to invoke in certain situations and by doing so, typically slow the print down to meet a 15 second minimum. What breaks this is when you have some tiny tube or rod or say a point on a cone shape, and no other adjacent objects or layer fills, then when slowdown invokes the nozzle just sits and cooks the layer and remelts the previous layers from conducted and radiated heat.

The best bet is probably to turn this off and you'll notice it's on by default in every profile I checked.
Raise3D slowdown.jpg

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Thu Mar 19, 2020 3:49 pm

FWIW, the cooler you linked I believe is also a good candidate. Blowers generate more static pressure airflow and thus flow higher rates through the constrictive printed ductwork. A such, this kind of does what the Berd cooler does, provide a ring of airflow around the nozzle.

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

Re: N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Thu Mar 19, 2020 8:25 pm

Jetguy wrote:And, based on this, you might need to keep some form of stock fan + the Berd cooler https://forum.raise3d.com/viewtopic.php ... erd#p40598

I found my post and pictures on my install https://forum.raise3d.com/viewtopic.php ... erd#p33477


As always, I appreciate you and your input and guidance, VB. I do have a major concern about all of this thought. I have evidence that the exact model I printed was printed on an Ender 3 flawlessly. The piece I shared above is the long "rope" in this model below. If theirs is flawless, when isnt a $2,500 Raise3D N2 able to do the same as a $300 Ender3?

Screenshot 2020-03-19 16.22.18.png

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

Re: N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Thu Mar 19, 2020 8:38 pm

Jetguy wrote:FWIW, the cooler you linked I believe is also a good candidate. Blowers generate more static pressure airflow and thus flow higher rates through the constrictive printed ductwork. A such, this kind of does what the Berd cooler does, provide a ring of airflow around the nozzle.


Again thank you (i didnt hit refresh before my last reply, so i didnt see ALL of your awesome info)

Ok, so that's good news, because I have one of those blowers right here, and the part already printed in case I got good feedback.

YOu did blow my mind a little saying turn off the slow-down in S2D, as someone told me to raise the time from 15s to 25s.. however, their slicer moves the nozzle to home to pause, where as both S3D and IdeaMaker just hover over the print cooking it like you said. Wish there was a way to not do that @Vicky :)

User avatar
Vicky@Raise3D
Posts: 6495
Joined: Fri Mar 25, 2016 3:54 am

Re: N2 Single Extruder 2016 - Part Cooling

Postby Vicky@Raise3D » Tue Mar 24, 2020 5:51 pm

Do you mean that you don't like that the Minimal Layer Time affects a whole layer? If so, you can try Small Features Detection.
It can detect Small Features on your model and adjust printing settings for those parts.
https://support.raise3d.com/ideaMaker/5 ... 5-436.html

Jetguy
Posts: 3015
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: N2 Single Extruder 2016 - Part Cooling

Postby Jetguy » Tue Mar 24, 2020 7:31 pm

I could say yes or no to that question, it's still a broken concept.
Slowdown as a function in actual 3D printing at the nozzle actually printing plastic does more harm than good.

If you slow down, then the nozzle is not only radiating heat, but also conducting heat through the liquid plastic to the previous layer.
In addition, this slow movement prevents airflow from the adjacent fan from reaching the layer (it blows around where the nozzle is- not where the nozzle tip is).

Doesn't matter how you detect it. When the function invokes- that's the wrong thing to do.
The right thing is to move the nozzle AWAY from the small layer object, thus giving the layer time to cool without a nozzle adding more heat, and then return back for the next layer. This too has issues, because now you force a retract each layer, and then if the nozzle is not printing and just idling parked, that can lead to ooze and other drip, which is a double problem of now not being primed for the start of the next layer and potential booger on the nozzle that now gets dragged into the actual part.

Which is why I said the PROVEN advice.
Either print more than one copy of the small object, and that in turn forces the nozzle to move between multiple small objects but increases the minimum layer time to not require a slowdown function
OR
Add a sacrificial object tower the same height as the object, and that does the same thing if you don't or can't print multiples of the object.

Both of those methods do not invoke slowdown nor do they cause a paused retraction where the nozzle is not printing and cooking the filament

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

Re: N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Sat Apr 11, 2020 1:33 am

Jetguy wrote:I could say yes or no to that question, it's still a broken concept.
Slowdown as a function in actual 3D printing at the nozzle actually printing plastic does more harm than good.

If you slow down, then the nozzle is not only radiating heat, but also conducting heat through the liquid plastic to the previous layer.
In addition, this slow movement prevents airflow from the adjacent fan from reaching the layer (it blows around where the nozzle is- not where the nozzle tip is).

Doesn't matter how you detect it. When the function invokes- that's the wrong thing to do.
The right thing is to move the nozzle AWAY from the small layer object, thus giving the layer time to cool without a nozzle adding more heat, and then return back for the next layer. This too has issues, because now you force a retract each layer, and then if the nozzle is not printing and just idling parked, that can lead to ooze and other drip, which is a double problem of now not being primed for the start of the next layer and potential booger on the nozzle that now gets dragged into the actual part.

Which is why I said the PROVEN advice.
Either print more than one copy of the small object, and that in turn forces the nozzle to move between multiple small objects but increases the minimum layer time to not require a slowdown function
OR
Add a sacrificial object tower the same height as the object, and that does the same thing if you don't or can't print multiples of the object.

Both of those methods do not invoke slowdown nor do they cause a paused retraction where the nozzle is not printing and cooking the filament


I totally understand. I was just thinking that rather than wasting plastic, if the head could move to home, wait 20 seconds, print, move, etc it would be good. But a sacrificial tower is likely a good idea.

I did print the part cooling piece, and it was working fine until i had a poor layer that lifted a bit and it pulled the piece off because it's so thin. I think a function of better cooling than the stock N2 fan (which has a GIANT opening and doesnt blow directly on the part) and the tower will be my solution.

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

Re: N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Thu Apr 16, 2020 7:58 pm

Jetguy wrote:While you can change the cooler, let's not kid ourselves on this specific print.
What you are fighting is going to be near impossible for a fan cooler to fix.

...
What you need to do is identify these types of critical prints, place either more objects or sacrificial objects in the slicing path. This way, the nozzle moves away from the layer thus allowing fan cooling to actually work, and fixes that layer minimum time because the nozzle has more distance on a per layer and more moves. Again, something as simple as saying I need one, let me print 4 takes it from junk and a waste of time to success and extra parts with minimal waste. And no insane modifications required that still don't solve the problem.


As always, you're right :)

I added a 15x15x40mm "tower", also slowed the printer down from 50mm/s to 35mm/s (painful..) and the result on a different print but with an even more delicate part was much better. Not perfect, but more than acceptable given how small it is....

Screenshot 2020-04-16 15.57.43.png


Screenshot 2020-04-16 15.57.50.png

NewbPilot
Posts: 287
Joined: Tue Jul 26, 2016 2:16 am
Location: West Palm Beach, FL
Contact:

Re: N2 Single Extruder 2016 - Part Cooling

Postby NewbPilot » Thu Apr 16, 2020 8:00 pm

Oh, the cooling fan worked, but it hung about 1/8th of a mm lower than the nozzle, and on a print that curled got ripped off the hotend. I need to reprint it - or just accept it is what it is and im never going to have perfect prints


Return to “N Series”

Who is online

Users browsing this forum: No registered users and 1 guest