Incorrect raft contact z distance [unified issue tracker]

[nordurljosahvida]

This aims to be a single thread to track all raft contact z distance related issues, all of which don't seem to be gaining enough attention individually, nor are the suggestions given in any way conclusive.

Also, I haven't seen some conclusive research onto the topic that provides data as to what actually currently happens under different scenarios.

Version

<= 2.2.0-alpha2, testing in this post with 2.1 stable

Operating system type + version

several if not all

3D printer brand / version + firmware version (if known)

seemingly irrelevant

Behavior

The z distance / air gap that is left between the last raft layr and the first model layer is vastly incorrect, leading the object in several cases to be printed in mid-air, far above the last layer of the raft.

Previous threads mentioning the problem

• Distance from Raft as Air Gap like in cura #2823
• Erroneous "Contact Z Distance" #2766
• Unstable first object layer with raft #118
• Make support separation distance configurable to negative numbers to keep ABS from warping #102
• first layer of support for multi filament support oddly spaced #1445 [different but related]

Project File (.3MF) where problem occurs

raft_issue.zip

Analysis

First off, there needs to be clarity and documentation about what we're talking about.

I'm using MK3S printer [standard 0.4 nozzle] with generic PLA and 0.15 quality profile. 4 layers of raft and 0.1mm contact z distance. supports are enabled but don't seem to affect this, but are helpful to understand the thickness of the first model layer by visual comparison in front view mode [shortcut: 3 key]

1. In prints with raft enabled, does the first layer height value apply to the first raft layer or to the first model layer, or both? Apparent answer: first raft layer only

2. In my tests the first model layer above the raft is printed at 0.4mm layer height [with a 0.4mm nozzle by the way!], respecting the contact z distance setting, so after the last raft layer the toolhead moves 0.5mm up and prints a 0.4mm layer 0.1mm vertically off the raft. The setting seems to be: first model layer above raft height = nozzle diameter. This doesn't work!

3. Raft layer height [except first raft layer] is not customizable by the user, in my tests it's printed at 0.3mm layer height with 0.4 nozzle and 0.45mm layer height with 0.6 nozzle. The setting seems to be: non-first raft layer height = nozzle diameter *3/4. This is reasonable.

4. Number of raft layers doesn't seem to affect the issue.

5. Supports generated over parts of the raft [such as the edges often times] if supports are also enabled other than just the raft don't seem to affect the issue, they only appear to introduce [as expected] micro-layers of supports between the last full and proper raft layer and the first model layer, without affecting their positions or heights.

6. If you set first layer height to 0.4 in the above 3mf prusaslicer file:

• First layer / first raft layer: 0.4mm
• Second layer / second raft layer: 0.3mm
• Third layer / third raft layer: 0.3mm
• Fourth layer / fourth raft layer: 0.3mm [toolhead height 1.30mm]
• Fifth layer / first model layer: 0.4mm with 0.1mm z-airgap [toolhead height 1.80mm]

If instead you set first layer height to 0.2mm in the above 3mf prusaslicer file:

• First layer / first raft layer: 0.2mm
• Second layer / second raft layer: 0.3mm
• Third layer / third raft layer: 0.3mm
• Fourth layer / fourth raft layer: 0.3mm [toolhead height 1.10mm]
• Fifth layer / first model layer: 0.4mm with 0.1mm z-airgap [toolhead height 1.60mm]

Expected behaviour:

1. The first model layer printed at 1:1 nozzle size layer height is in my opinion way too high, and produces in almost all cases a random print in the middle of the air. Its value should be customizable, if not at least set to a reasonable fraction of the nozzle diameter.
2. The first layer height setting must specify that it's being applied to first raft layer and not first model layer above the raft in prints with a raft.
3. Perhaps the user should be allowed to customize the non-first raft layer height, otherwise it should be specified what it's going to be calculated at.
4. super-mega feature request: allow the raft contact z distance to be different than the support contact z distance. This will in so many cases need to be vastly different! And the support option of the height range modifier feature was removed! I've found myself splicing together 2 gcodes several times to achieve both a functional raft with a negative contact z value [which is in and of itself something that was requested and is now available but we absolutely don't know why it's necessary and if so even more to the point, why are we printing so high up off the raft?] and supports that didn't actually physically conflict with the model but were in fact removable.

Bonus issue

support_issue.zip

What on earth is going on with support [not raft!] contact z distance? In the 3mf below the last support inside the X letter of the calibcube is printed at 1.44mm height, and the ceiling is printed at 2.00mm height. you can tell that on the sides it's printing at 0.15mm height, correctly, because otherwise it would be printing inside the gyroid infill if it were printing a thicker layer, instead above the X void you can see from the front view [shortcut: 3 key] that it's printing a thicker layer. How thick? It would need to be 0.36mm thick to "start" at 1.66mm height and therefore respect the 0.2 contact z distance. Is it? Is it not? How do we know? Why is it doing this? How does it calculate it?

Thanks

[Shama71]

I think that need to make separate settings for the raft so that there are no misunderstandings and add more parameters, such as width, etc.

[sungdzda]

This issue deserves more attention. It's surprising to me that most users are not even alarmed by the raft/support quality of Prusa printers. One negative effect is the fact that Prusa printers are uniquely characterized by the "reverse elephant foot" phenomena which significantly impacts functionality of the printed model. Also see #2767. It is worth noting that the raft/support implementation of Slic3r/PrusaSlicer differs grossly from industry customs and standards. Furthermore, Prusa development seems to be driven by a few unhealthy opinions (promoting true raftless printing style and the use of polyesters which never have any warp issue, while by the large ignoring other use cases). While it is possible to switch to another slicer program, PrusaSlicer is the only one that offers integration with MMU. This creates unpleasantness as the MMU along with material capabilities (as manifested in design) are what really define ownership value of the Prusa.

[nordurljosahvida]

I really do not understand the statements on slic3r's website where they say that rafts were for pre-heated bed / pre-mesh bed levelling days. I print with PETG every day and not only do I print on the PEI sheet with PETGs rather than the textured one [except for small objects], it still warps off! So rafts are definitely still really important IMO, let's make them work.

[bubnikv]

I print with PETG every day and not only do I print on the PEI sheet with

PETGs rather than the textured one [except for small objects], it still warps off! So rafts are definitely still really important IMO, let's make them work. I don't understand how raft can help with warping off. If there is an excessive stress, something somewhere has to let go. The print will either detach from the print bed, from the raft, or the print layers will delaminate. čt 2. 1. 2020 v 15:44 odesílatel Nicolas North <[email protected]> napsal:

…

I really do not understand the statements on slic3r's website where they say that rafts were for pre-heated bed / pre-mesh bed levelling days. I print with PETG every day and not only do I print on the PEI sheet with PETGs rather than the textured one [except for small objects], it still warps off! So rafts are definitely still really important IMO, let's make them work. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#3425?email_source=notifications&email_token=ABMPSI7MSHULXZD64AJHQQLQ3X4U3A5CNFSM4KBFVAZKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEH6P2NI#issuecomment-570228021>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABMPSI4E24YTR3QIXKHGNDDQ3X4U3ANCNFSM4KBFVAZA> .

[nordurljosahvida]

Absolutely, but in very many cases a raft with sufficiently small contact z distance has saved my prints, i too would love to know why, or how to identify that stress you're mentioning.

[sungdzda]

Another scenario that ordains the use of rafts is when the model has a small contact area with the platform. Many new Prusa users would happily leave their printers alone and go to sleep only to find out 8 hours later a hotend/fan duct surrounded by plastic goo that is very frustrating and it is nearly impossible to undo the damage. Despite raftless printing (including any use of glue stick and surface treatment) having certain merits, in the generalized case a raft is the best option to ensure consistency, dimensional accuracy, and for preserving surface quality of the first layer.
Image source reddit

[bryanmtdt]

Please, this is an important setting, support and raft settings to be separated and first model layer height on other values rater than nozzle diameter, I've spend several hours trying to get a good balance but it seems to be pretty difficult right now.

[alfredone78]

Please, fix this issue: I'd like to have a first layer (above the raft) with an height different from the nozzle size! For example, I want that the first layer is 0.2 mm, just like the other layers (default height).
Instead, now as things stand, I can't!!!

[Zaniix]

I just ran into this issue as well. It does not seem to matter what settings I change, the first layer after the last raft layer is printed too high as if I was just printing in the air. I wasted a lot of filament before I realized it was the slicers fault and not my printer.

[str-dst]

same here.

layer height set to 0.2mm and contact-z distance set to 0.2mm as well I would expect the slicer to leave out exactly one layer between the last interface layer and the first object layer.

Instead I get three missing layers resulting in a gap of 0.6mm which results in mid-air printing and thus unusable print results.

In this state PrusaSlicer is pretty much unusable for any models that require supports for printing.

[Drachenkaetzchen]

I ran into the same issue. Please either fix this issue or at least make an in-slicer warning that this feature is currently broken. I spent hours debugging my hardware when in fact it was a slicer issue :(

[nemart69]

Just FYI: I submitted PR #6003 which addresses most of the issues mentioned here a couple of days ago. This PR seems to be processed by PrusaSlicer development team.

Some examples here (white: Polymaker PC-Max: nozzle 0.4mm, 280/110 C in enclosure, silver: Prusa PLA):

[bubnikv]

Implemented with #6003 and aee136c

Object over raft is no more printed with bridging flow, but with a normal flow.
Gap between the raft and 1st object layer is configurable independently from the support gap.