Issues printing ColorFabb PA CF with Hardcore

[V3DPrinting 11]

Hi

 

I am trying to print Carbon Filled PA from ColorFabb on my Ultimaker 3 and Ultimaker S5, but I am having some jamming in the print core.

 

I am using 3d Solex Hardcore with an EverLast 0.6 nozzle.

 

The heat builds up in the hardcore, so the filament softens well after the heat break and jams the hardcore up to the bowden tube if the print is left unattended.

 

It is my second attempt, on the first one the Hardcore was new and after some tests was not able to print even with PLA due to a defect in the heatbreak, Thanks to Carl from 3d Solex, the issue had been fixed very quickly and efficiently.

 

This second attempt, I took an Hardcore which was performing well with a 0,6 brass nozzle, swapped the nozzle by a 0,6 Everlast nozzle and started to print.

 

Anyone had the same issues ?

 

Anyone printing ColorFabb PA CF on Ultimaker ?

 

I've ended printing the customer order on a printer with E3D V6 and hardened steel 0,6 nozzle, printing like a charm.

[JohnInOttawa 104]

Are you able to share your print settings?  temps, layer heights, speeds?

 

On the UM3, what is the extruder configuration?  Standard? bondtech? something else?

 

Thanks

John

[V3DPrinting 11]

On UM3 I have Bondtech QR 3.0 extruders. On UM S5, my last attempt it's stock.

 

Layer height is 150 microns, bed 50°C nozzle 270°C. I've used the standard speeds for Nylon settings in both cases.

[JohnInOttawa 104]

Hmmm.  That sounds like a frustrating challenge.

 

So is the print core suffering excessive heat creep up the heat break, or is the filament loitering in the print core too long, or is the nozzle itself experiencing some sort of local clogging at the tip that is causing things to back up? 

 

I see Colorfabb recommends a minimum layer height of 0.2 which is a bit higher than you are using.  I know when I printed Colorfabb Copperfill, I created a similar problem for myself when I went to a layer height below 0.2, of interest I was also using a 3D Solex print core, but with a steel nozzle.  I was able to print for about 90 minutes before things clogged.  I know there is a warning about copperfill conducting heat so one needs to keep it moving at a pretty fair clip.  Maybe the minimum layer height and width for the 0.6 nozzle is intended to mitigate the heat migration by keeping a fresh supply of relatively cool material flowing into the heat break?

 

Just curious, I imagine you are using an enclosure on the front.  Anything on top?  I wonder if there is any chance the bowden tube is pre-heating - not to nozzle temp of course, but even a delta of 10C would put it that much closer to its softening regime upon entry into the print core.

 

Final thought, if this is straight up heat creep in the print core, any chance the print core cooling fan is encountering issues or blockage? 

 

Wish I could provide you something more definitive.  Hopefully someone more knowledgable can wade in shortly.

 

John

 

 

 

 

[V3DPrinting 11]

10 hours ago, JohnInOttawa said:

Hmmm.  That sounds like a frustrating challenge.

 

So is the print core suffering excessive heat creep up the heat break, or is the filament loitering in the print core too long, or is the nozzle itself experiencing some sort of local clogging at the tip that is causing things to back up? 

 

As the issue appears in the very first layers and the nozzle itself is not clogged at all, the print core is not cooling properly for that special filament.

Or more precisely, the filament thermal conductivity is much higher so the heat is building up in the filament more easily.

 

I printed the same part on UM S5, same settings or so, same print core but standard 0.6 brass nozzle, with ColorFabb PA Neat without any issues.

 

 

10 hours ago, JohnInOttawa said:

I see Colorfabb recommends a minimum layer height of 0.2 which is a bit higher than you are using.  I know when I printed Colorfabb Copperfill, I created a similar problem for myself when I went to a layer height below 0.2, of interest I was also using a 3D Solex print core, but with a steel nozzle.  I was able to print for about 90 minutes before things clogged.  I know there is a warning about copperfill conducting heat so one needs to keep it moving at a pretty fair clip.  Maybe the minimum layer height and width for the 0.6 nozzle is intended to mitigate the heat migration by keeping a fresh supply of relatively cool material flowing into the heat break?

 

Interesting.

I have managed to print a bigger part with PA CF at 0.1 layer height with an E3D v6 and 0.6 nozzle without any clogging or jamming issue.

The only point was that some carbon fiber was more visible on the surface. 

 

Same for the part I have tried yesterday at 0.15 layer height. Less carbon fiber visible though.

 

When ColorFabb released initially the material, recommended layer height was 0.15 but nozzle was not specified.

As the main problem with filled materials is the filling clogging the nozzle, using an higher bore nozzle helps a lot. 

 

10 hours ago, JohnInOttawa said:

 

Just curious, I imagine you are using an enclosure on the front.  Anything on top?  I wonder if there is any chance the bowden tube is pre-hearing - not to nozzle temp of course, but even a delta of 10C would put it that much closer to its softening regime upon entry into the print core.

 

Yes my printers are enclosed on the front and top, but the top is left open, so the heat is not accumulating. 

 

10 hours ago, JohnInOttawa said:

 

Final thought, if this is straight up heat creep in the print core, any chance the print core cooling fan is encountering issues or blockage? 

 

As as the first attempt on UM3 was already having these issues, I have checked and cleaned the printcore cooling fan before starting the second attempt on UM S5

 

My though is the cooling on the hardcore is not sufficient for that special filament, as there are less fins on the cooling part and maybe the print head fan is not that efficient for the two print cores to be cooled.

 

I am curious to see the Printcore CC and reviews printing ColorFabb PA CF with it.

[JohnInOttawa 104]

That's an excellent analysis.  I had not noticed the different cooling fin configuration.  Are you heating both print cores for this work?

 

I don't know of any way to get a temperature reading on the print core fins.  A thermal imager would be best as it would not interfere with airflow like a probe would, but with the fan in place and lateral airflow affecting boundary conditions that's impossible.  Indirect measurement like exhaust air temperature would be misleading if the problem is that the fins can't shed heat fast enough.  A direct contact measurement would carry all sorts of risks, not least of which would be detachment and collision with one of the fans.

 

I don't know of a practical way to feed colder air into that fan.

 

It sounds like 3D Solex has been actively engaged, I know @gr5 would have a wealth of technical knowledge on these cores, hopefully there is a path forward from their help.

 

Thanks again for sharing - it will ultimately help everyone.

 

John

Edited December 29, 2018 by JohnInOttawa

[V3DPrinting 11]

22 minutes ago, JohnInOttawa said:

That's an excellent analysis.  I had not noticed the different cooling fin configuration.  Are you heating both print cores for this work?

 

No it's a single extrusion job, so only one print core involved.

 

Here's a picture of the old version 2017 on the left and the new one 2018 on the right.

Less fins on the new version. And less than the original Printcore.

 

22 minutes ago, JohnInOttawa said:

 

I don't know of any way to get a temperature reading on the print core fins.  A thermal imager would be best as it would not interfere with airflow like a probe would, but with the fan in place and lateral airflow affecting boundary conditions that's impossible.  Indirect measurement like exhaust air temperature would be misleading if the problem is that the fins can't shed heat fast enough.  A direct contact measurement would carry all sorts of risks, not least of which would be detachment and collision with one of the fans.

 

Agreed. Maybe gluing thermistors on the dissipator would give some indications, but that means testing equipment and time.

Being a service provider, I don't have the time for that.

 

22 minutes ago, JohnInOttawa said:

I don't know of a practical way to feed colder air into that fan.

 

It sounds like 3D Solex has been actively engaged, I know @gr5 would have a wealth of technical knowledge on these cores, hopefully there is a path forward from their help.

 

Thanks again for sharing - it will ultimately help everyone.

 

John

 

Yes maybe Carl @swordriff or @gr5 have an idea on how to solve the problem, linked to that particular filament.

 

I haven't tried yet Glass Fiber filled nylon, but I may as I'm working on some replacement parts with a customer that requires XSTRAND GF30 PA.

I know it has been qualified on the UM S5 and some material settings are available, so it works with the PrintCore CC at least.

It would be good if it also works on the Hardcore with Everlast 0,6 nozzle, as I already have it.

[JohnInOttawa 104]

While this filament makes for a pretty expensive 'try and see' test, running with 0.6 for a while to see how long it takes for trouble to recur would likely confirm your suspicions.  If you are able to move volume at a rate that keeps up with heat creep and the heat absorption characteristics of the filament upstream of the nozzle, then it would seem certain that it's a cooling problem within the core and filament combination. 

 

I'm trying to think of other filaments that might print at that temperature but don't have as good heat conduction.  maybe polycarb or high temp nylon without additives.  I'm still trying to figure out the heat path - whether it is just along the metal path from nozzle through heat break then cooling tower and heating the filament the whole way, or if the primary contributor is within the filament itself.  The rapid onset of problems suggests it's likely the former.

 

I wonder why the number of cooling fins was reduced.  Hmmmm.

 

John

[V3DPrinting 11]

John

 

It is not easy to create the right conditions with that material as print speed is pretty low (30 - 40 mm/s) and temp pretty high 280°C for a 0,2 mm layer height.

Increasing the nozzle diameter will lead to a temp increase so increase the problem.

 

I print some grades of ABS at high temp : 270°C with the 3D Solex hardcore. No issues at all. Of course it's with a standard 0,4 brass nozzle, but it doesn't have any impact on the current problem.

I have also printed ColorFabb HT on UM3 with the Hardcore without issues. Printing Temp is also 270°C

 

To me it is the material itself who creates the problem first.

Then maybe the thermal dissipation of Hardcore + fans is not sufficient and the filament above the heatbreak reaches the glass transition temperature, softens and jams the print core.

 

So I am curious to see in details the Printcore CC to understand how they overcome that challenge.

[V3DPrinting 11]

I am now facing the same jamming issues on my printer with E3D V6, but after 1 hour printing approx.

Parts which printed well on wednesday are now jamming the hotend (0.2 mm layer height, 280°C, Bed 50°C, speed 40mm/s)

Same for parts printed at 0,15mm layer height, 270°C, speed 40 mm/s

 

I have lowered the temp to 265°C for 0,15mm layer height keeping speed at 40 mm/s in order to maintain the flow. I have also opened the front of the printer to decrease the temp of the air cooling the hotend.

Still jamming the hotend after 1 hour printing.

 

That confirms the fact that the filament itself is drawing up the heat past the heatbreak, as the E3D V6 are very good hotends.

 

The spool has been opened in may '18 and kept in a dry box, even when printing.

There might be some slight moisture in the filament as the surface finish is not as good as the very first print done in May.

 

I am wondering if there is a relationship between the softening temperature and moisture level in the filament.

 

I will dry the spool tomorrow 

 

Any help appreciated.

Edited December 30, 2018 by V3DPrinting

[JohnInOttawa 104]

That's discouraging to hear that the alternate print head is encountering issues.  But at the same time, your testing methods seem robust to the point you can isolate it to the filament.  It does sound like shelf life or humidity could be factors.

 

I have not seen anyone try to chill filament and don't know what would happen if one tried.  Certainly a refridgerator could be a low humidity environment and keeping the filament as cold as possible during print might delay softening just long enough to get the material through the 'danger zone'.  Or it could just fail.  I also don't know how rapidly the filament would warm to room temperature along the bowden tube.

 

I have one of those electric coolers that should hold a spool.  I may have to give it a try.

 

Maybe colorfabb can lend some expertise....

 

John

[V3DPrinting 11]

I have dried the filament for 6 hours at 70°C on Monday.

I have made a perfect 2h print with it on my printer with E3D V6.

 

So there is a relationship between moisture and softening temperature.

Filament needs to be very dry in order not to soften too much.

 

I still need to do a test on UM with the Hardcore. I expect to do it this afternoon 

[V3DPrinting 11]

I have finally managed to make the test. I am successfully printing ColorFabb PA CF with an 3d Solex Hardcore + 0.6 Everlast nozzle for now more than 1h without any flaws on an UM3.

 

The part is a gear printed at 150 microns layer height, 270°C printing temperature, 50°C Bed, 70 mm/s print speed with infill speed at 40 mm/s, fan 20%.

So with the same parameters as in the failed tests.

 

The only difference is the spool had been dried for 6 hours at 70°C early this week and kept in a sealed bag or a dry box since.

 

BTW, the print surface is very good now, which wasn't the case previously.

 

There is a direct relationship between the level of moisture in the filament and the softening of the filament. This led to print core jamming in the previous tests with that particular filament.

 

So ensure that the ColorFabb PA CF is very dry before starting any print with the Hardcore.

 

[V3DPrinting 11]

Good news!

 

The print completed in 2h20 without any issues, printing with the Hardcore + 0.6 Everlast nozzle

 

So the problem was too much moisture in the filament.

[JohnInOttawa 104]

This thread should be a sticky - it's a great reference for that filament.

 

Thank you for sharing.

Cheers

John