(Soluble) Support for PETG - Full interface layer  

RE: (Soluble) Support for PETG - Full interface layer

have you tried printing with PETG yet? I just got my mmu2s last week, and I'm still trying to iron out a few issues, but I really like printing with PETG, and I too would like to find a support material. I have HIPS, and I will probably try it with PETG once I get PLA to print without issues. I will share my results if I get around to it, but I would really like to see your settings for PETG if you have tried it yet.

RE: (Soluble) Support for PETG - Full interface layer

Have you looked at Verbatim BVOH?

• https://www.verbatim-europe.co.uk/en/prod/verbatim-bvoh-filament-175-mm--natural-white-55901/?con=GB
• https://shop.prusa3d.com/en/special/161-verbatim-bvoh-soluble-support-05kg.html

It's a high-temp soluble material that should be PETG compatible. I haven't tried myself yet, my MMU2s bis to be delivered in about two weeks, hopefully. In the US, 3DXTech makes their Aquatek BVOH and Filament Innovations also produce a BVOH filament. In the Czech Republic, Fillamentum is about to release their version. 

RE: (Soluble) Support for PETG - Full interface layer

Hi Guys,

In general:
I had to go through a LOT of learnings and there is still a lot of work to do. MMU2, for me, feels overall like in a beta-status. 
It may be an adequate solution for multi-colour PLA prints, but everything else is challenging.

The Prusa support-chat told my straight that it is not yet tuned for prints like PETG with soluble material.

Some hints to get you started:
1.) Using Slic3r for Prints with support interfaces: The material you use has to be defined as soluble material. Even if it is not. (Filament settings). Otherwise Slic3r will alternate the materials for each layer in the support structure (only visible after exporting the gcode! "Slice now" works correctly).

2.) According to the Prusa manual and the Prusa support-chat, printing with different temperatures is not recommended.
This is where things get ugly: You need to adjust your temperature to avoid stringing on the tip of the filament when retracted.

The extruder adjusts the print temperature very quickly. I still do not fully understand why, but I do face a lot of stringing using different temperatures, which clogs the way to the extruder on a regular basis. This is why I'm currently looking into filament which can be printed at the same temperature as PETG. (Scaffold Snap, which is discontinued, High-T-Lay and Taulman Dow USM are options).
When the stringing issue is resolved, I may have some chance to run this printer unsupervised again. 

3) Make sure that your heatbed temperature is equal for all filaments you use (this is part of the filament settings!). If it is not, the extruder will wait above the purge tower and keep oozing material. Since it waits always in the same spot, material will accumulate. First it will bend your entire X-axis while the nozzle tries to melt through it. When it's high enough, the PINDA will connect and cause continuous crash detection.

4) Layer adhesion / PLA as support:
First of all: PLA makes an astonishing support material for PETG. It sticks good enough, but is easy to remove (full interface, zero distance on the Z axis with intentional, slight overextrusion of the PLA). The surfaces are just wunderful. I was able to fill an 1mm high and 10mm deep gap in the PETG structure with a full interface and just pull it out. It kind of lets you forget about soluble filaments. It is soooo perfect.... BUT:

4.1) The temperatures are quite different. I do have a lot of stringing, especially with the PLA (using Prusa PLA and Prusa PETG with the original Prusa settings). I did not run a study to directly point on the temperature difference as the root cause for the stringing. But I think this is what happens. Meaning I have to be around the printer and take active care about it constantly.

4.2) During my last print with PETG (and PLA support) it lost any layer adhesion to itself. No settings changed. The entire print falls apart under light loads. It looks beautiful, but is completely useless.
I am still investigating the root cause. Maybe something was overall wrong. But a friend gave me an interesting idea:
The standard purge tower is not exactly big. It is more than enough to have color accuracy. But it may be the case that the nozzle is not entirely clean yet.
If you think about it: Where will PLA impurities be? They would form around the outside of the filament extruded. And therefore all PLA impurity would be between the PETG layers - not inside the PETG. My understanding is, that PETG can form something like a molecular bond with itself. If this is interrupted by PLA impurities you got your answer..... maybe its wrong, but the idea fits very well.

I will retry this at some point and increase the purge volume.
By the way: Prusa recommends to adjust the pruge tower depending on the material used. Maybe they had this in mind. 
(Fun fact: Recommended purge volume for PVA is twice the standard setting - if I recall it correctly)

@Vojtech:
Thank you for the recommendation. But the conversation with the Prusa support was not exactly reassuring in this regard.
Correct my if I'm wrong, but BVOH seems to be quite similar to PVA. The Verbatim material should be better then the Primaselect, but I do not want to spend >75€ on another material which is not designed for the usecase and has to be printed at a different temperature.
During the last days I checked all the major brands. There are high-temp PVAs which do not work with PETG, low temp PVAs which would work with PETG, but cause the described nozzle temperature difference... This seems to come down to the specific design of the individual blend.

In general the market does not offer much for PETG. Lots of materials for PLA and ABS/ASA. But this is not what I am interested in.
After some readings about HIPS I also do not want to go this route. I am a chemist, but I don't need the messy stuff in my home.

I will try Scaffold Snap and High-T-Lay now. They sound like the best chance for success and I get both spools (500g/250g) for a similar price as the Verbatim spool.

I'll let you guys know what I find out.
In the meantime, if you listen carefully, you can hear me swearing. 😉

Regards,

Michael

RE: (Soluble) Support for PETG - Full interface layer

Posted by: michael.s106

I will try Scaffold Snap and High-T-Lay now. They sound like the best chance for success and I get both spools (500g/250g) for a similar price as the Verbatim spool.

 

I'll let you guys know what I find out.
In the meantime, if you listen carefully, you can hear me swearing. 😉

Regards,

Michael

Awesome, thank you for doing the test!

RE: (Soluble) Support for PETG - Full interface layer

Hey Guys,

two more learnings from today's prints. All about purging the nozzle:

1) It is surpirising how much filament is required to purge the nozzle clean. I printed back and orange Prusa PETG today. Black impurities are very well visible in the orange material.
The following picture shows a print with 35 perimeters (for appearance). The perimeters are printed first, then the inside is filled out rectilinear. It shows how long it takes until the bright orange is really clean.

This is printed with the standard purge tower settings (70). I increased it to 400 for this color combination.
Truth to be told, using black & orange PETG seems to be a worst-case scenario. Other colours work better. I'm just printing white/blue, and the white looks clean way faster.
However, for technical prints with soluble material, this is still important to know. For robust prints you do not want soluble impurities between your layers.

2) The purge volume can be adjusted in Slic3r for each individual extruder. This seemed to be handy, since black looks black pretty fast.
Of course the purge tower has to keep its width for the complete hight. Therefore Slic3r creates a hollow structure when less material has to be purged. The next layer just bridges the gaps between the printed lines below. This works well with PLA.... but not with PETG. The default distance between the gaps is 10mm. I'm not sure what speed is used. Maybe the bridging speed of the print settings. However, standard settings do not work for PETG:

This creates a blob around your nozzle, since the hot nozzle pics up the strings standing upwards.

Workarounds:
a) Use equal  purge settings for all materials. This does not avoid all hollow structures, but decreases the area.
b) Go to print settings -> Wipe tower -> "Maximal bridging distance" and decrease the distance (not tested, but I assume it helps).

Update concerning stringing: Be aware that equal temperatures are good, but you will still face stringing using PETG. Further tweaking of the settings will be required. When I find time I'll look into Nicolai's ramming settings and play more with the temperatures.

RE: (Soluble) Support for PETG - Full interface layer

Hey Guys,

one more hint for the fight against impurities:
During the first layer I cleaned the nozzle and the print bed manually. But I still ended up with cross-contaminations.

So, we all know that PETG has a tendency to accumulate on the nozzle. For single material prints this is not a big deal. Ususally it just disappears inside the print and you will never notice it. But it becomes a problem if you go for mutli-material. You may end up with a hole in your print (soluble). During filament changes, purge steps etc. you can basically be sure that your nozzle accumulates material. It's just a property of PETG. Other material may be mixed in, and the fun begins.

The solution is a silicon sock for your hotend. Do NOT use the normal, open model. That's what I used and it does not help. Go for the "pro" model.
They are quite durable, super cheap and worth every cent. (You can even geht them on Aliexpress. But at that pricepoint I prefer the original stuff from E3D.)
I repeated the print without manual cleanup and had no issue at all.

RE: (Soluble) Support for PETG - Full interface layer

I believe the technical term for the property of PETG accumulating on the outside of the nozzle is that it is 'boogery'. 😉 The sock does help turning the boogers into small balls that don't interact with the print much. But I've also found that different PETG brands display different levels of boogerity. Prusa (Filament-PM) PETG black is producing a huge amount, but when I print with Fiberlogy PETG Onyx, I get almost none. I assume the various other PETG-based polyesters (called CPE, nGen, etc, etc) will also vary in this property. So changing to a different black PETG may be a good option, tooo.

RE: (Soluble) Support for PETG - Full interface layer

Thanks Vojtech! I learned something 😉

Short update:
I tried to increase the temperature difference for a PETG/PLA print.
I noticed that the stringing of PLA in these prints is way worse than the stringing of PETG. This is contradictory to the material properties. However, it makes sense when PLA gets heated up too much before it is retracted. This time, I printed the PLA-support structure at 190°C. Unfortunately without noticeable improvements.

However, I watched the printer during the filament changes:
The retraction is a complex process (filament is retracted, pushed back in and then retracted again). In the timeframe between the start of the retraction process and the time the filament is actually pulled out of the extruder, the temperature did already increase by 20°C (to 210°C).

I didn't have time to investigate further yet. So I'm not sure whether this is a Slic3r or Firmware issue. But I'll check the Slic3r settings for the next runs. Maybe I find something to delay the heatup process.

Best Regards,
Michael

RE: (Soluble) Support for PETG - Full interface layer

Hey guys,

next update: I tried different settings with Prusa PETG.
- Nicolai's settings @230°C
- Nicolai's settings @250°C
- Prusa's settings (230°C, 1 cooling move)
- Prusa's settings modified: (4 cooling moves)

I see no improvement.... It all looks bad. I can go on messing blind with some more settings.... but right now I don't have really good ideas any more.

The new support filaments did not arrive yet. I'll let you know how they work. But as long as PETG looks as bad as it does, unsupervised prints are not really a good idea.

Regards,
Michael

RE: (Soluble) Support for PETG - Full interface layer

Hi Michael

Do you know this thread?

https://forum.prusa3d.com/forum/original-prusa-i3-mmu2s-mmu2-general-discussion-announcements-and-releases/mmu2-real-multi-material/

Unfortunately I haven't got my MMU so I can't try myself. Printing support structures with different materials is one of the main reasons why I've ordered a MMU

BR Fabian

RE: (Soluble) Support for PETG - Full interface layer

Hi Fabian

The thread is interesting, but not really necessary for the "PLA as support" scenario:
We may see inpurities in the PLA, since it purges at lower temperature. But PLA will only be a support structure. The layer adhesion is good enough for that purpose.

When we want to use PETG as support for PLA, it is a different story of course. Then Nicolai's work comes really into play.
But there are a lot of support materials which are desigend for PLA. Specifically the PVA+, which started the topic of "different temperatures" for me. So this is a task for another day.

General Update:
I was in contanct with the Prusa Support. They gave me a recommendation for the speed settings.
These had an actual input. Stringing was not gone, but due to the higher speeds the strings became thinner on average.

Then I combined the ramming settings of Nicolai with the speed-mod from Prusa and had some great results. Hardly any stringing! And if there was any, it was only super-thin strings.

Unfortunately, when I printed the same Gcode again today, stringing was back. Still only these thin hairs... but not the clean tips I hoped for.
I also tested 235°C and 240°C. But results became worse.

I cannot yet evaluate what this means for reliability. But overall I would call it progress.
With some luck these thin hairs will have only a minor impact.
Also, there are some more things to do. I still need to heat the filaments in an oven for a couple of hours to get all moisture out. This may help as well a little.
I also ordered a spool of PETG from "Das Filament". It is a German brand with a very good reputation. I've heard that their PETG has a reduced tendency for stringing.

These are small steps, but I still think I'm getting closer.
I'll focus on this a little longer, and then go on with the new support materials Scaffold Snap and High-T-Lay.

These are the settings I used:

RE: (Soluble) Support for PETG - Full interface layer

Hello Michael!
For several days I have been struggling with the same problem, I print 90% of Prusament PETG Jet Black, I installed the MMU only to support my prints and I met with the wall ...

I tried to use the HIPS Fiberlogs as a support - the prints were falling apart ... I increased the wipe tower to its maximum size - 600mm ^ 3 PETG cleaning before layer printing. The prints were 'a little' better, while they were still not as flexible as before and broke (like PLA).

Another test is to support PLA and exactly the same problem.

I tried to combine 2 PETG materials with each other - black and white, strength as it should be.

I come to the conclusion that using one extruder material support like PETG is impossible at the moment, unless a dedicated material is created (?) ...

I have a question whether you managed to solve the problem of delamination of PET prints with support material, maybe I have not tried something yet? 🙂

Best wishes!

RE: (Soluble) Support for PETG - Full interface layer

Hi Michael

Thank you very much for sharing your experience. I have the MMU unit now 1 month and tried the firs PET print with PLA support. Your settings worked well for me. For PLA I used the standard Prusa settings.

But I have a similar problem as wozny: After the first try with 300 mm^3 purge volume after PLA my part was very instable at the layer where the change from PLA back to PET took place. Increasing the purge volume to 350 mm^3 helped for my part. But also here layer adhesion is not as good as between the other layers.

The Multi Material Unit should be called Multi Color Unit that’s more the truth.

BR Fabian

RE: (Soluble) Support for PETG - Full interface layer

I came across this thread just now. I want to mention that 3DMaker PVA (which I bought recently) comes with a spec sheet that has recommended settings for temperature, speed, retraction, fan, jerk, acceleration, etc. This sheet was very useful for setting up this material's properties in PrusaSlicer.

It also has a temperature range of 190-230C. That upper end may work with PETG.

I've never used PVA before. I've printed one part so far (which failed, not due to the PVA). The MK3S+MMU2S handled the PVA well. The filament is smooth, flexible, slippery, and never jammed, although the material exhibited a lot of stringing, with the first retraction pulling back a string that tricked the selector sensor and had to be cleared. There were no problems after that. About 50 tool changes later the MMU2S simply stopped working (all lights flashing at once) so I had to terminate the print.

Thist PVA is extremely sticky when melted so it should have no problem adhering to PETG.

After soaking my failed part in water for 6 hours, it's now a soft gummy mass stuck to my part. Is that normal, or should I expect it to dissolve completely?

RE: (Soluble) Support for PETG - Full interface layer

Posted by: @anachronist

After soaking my failed part in water for 6 hours, it's now a soft gummy mass stuck to my part. Is that normal, or should I expect it to dissolve completely?

3DMaker responded to my query about this and said it typically needs soaking overnight in still water, or about an hour in an ultrasonic cleaner, and the length of time needed to dissolve also gives the material a consistency that is easier to print. It's been a couple hours since my last post and the PVA has dissolved further, to the point where my interlocking pieces can almost be separated. I'll give it more time.

RE: (Soluble) Support for PETG - Full interface layer

For people returning to this topic in the future, I did a bunch of tests with using PLA & PETG in same print and found that using different materials has a huge impact on model strength. For a typical purge volume of 250mm³ you can expect layer adhesion strength to be 1/5 of what it should be. Sadly only way to get around this is to purge crazy amounts of plastic ~2500mm³, or getting a 3D printer with at least two hotends

https://forum.prusa3d.com/forum/original-prusa-i3-mmu2s-mmu2-general-discussion-announcements-and-releases/just-how-multi-material-is-the-mmu2s/

RE: (Soluble) Support for PETG - Full interface layer

Posted by: @vojtech

I believe the technical term for the property of PETG accumulating on the outside of the nozzle is that it is 'boogery'. 😉 The sock does help turning the boogers into small balls that don't interact with the print much. But I've also found that different PETG brands display different levels of boogerity. Prusa (Filament-PM) PETG black is producing a huge amount, but when I print with Fiberlogy PETG Onyx, I get almost none. I assume the various other PETG-based polyesters (called CPE, nGen, etc, etc) will also vary in this property. So changing to a different black PETG may be a good option, tooo.

Just a quick PSA on a Zombie thread:

I had lots of problems with "boogery" when printing with PETG, until I realised that I was over-extruding. Turning down the extrusion multiplier just a little (without under-extruding) has massively reduced my PETG booger problems. So, something to try?

RE: (Soluble) Support for PETG - Full interface layer

PETG with BOVH can work, but I find that the problem comes from adhesion issues on the purge tower.  I have gotten it to work, but it doesn't work well.  I reached out to prusa about it some time ago and the responded back that solubles with the mmu are really only recommended with PLA.  I'm sure I have the email I got from them somewhere.   As far as printing with PETG, I agree, an extrusion multiplier of around .96-.97 usually eliminates the blobs.  I have found if you're printing something with a lot of supports you also should raise the temp from the default 230-240 to something like 250.  Otherwise, in my experience, they tend to get dragged when the nozzle moves.

RE: (Soluble) Support for PETG - Full interface layer

A very good support material that is pretty much designed for PETG is VXL 70 or VXL 90 from Xioneer Systems.

https://www.xioneer.com/support-materials/

This is also very easy to clean.