Bridge Line Width + Improve bridge density #1722

New Issue

MrUnknownDE · 2026-04-05T19:25:46+02:00

MrUnknownDE commented

2026-04-05 19:25:46 +02:00

Originally created by @ianalexis on 11/3/2025

Closes #11209, #11412 and #11954

The Issue

Currently, bridges calculate their path with a line width equal to “Internal Solid Infill” or, if this is not defined, to the general setting.
Most of the time, these have a value greater than the nozzle size, since when pressed against the previous layer they can generate a greater width and improve the bond between layers.

The problem here is that bridges have nothing to press against, as they are printed in mid-air.

The solution

Separate the line width value for bridges in order to correctly use 100% (line width) so that each bridge line is welded to the adjacent line without compromising printing time (if the overall line width were reduced).

Capped at 100% or equal nozzle size.

Flow and bond

Perform various tests considering the following points:

Connection of lines in the short section
Connection of lines in the long section
Quality of the bottom layer (how even it is, penalizes under- and over-extrusion)
Quality of the top layer (penalizes over-extrusion)

On average, the following conclusions can be drawn:

Width > nozzle (CURRENT) is a mistake and has the most negative impact. Compensating for the flow or density to counteract the effect is feasible but not precise. It varies greatly depending on the shape and length of the extrusion.
Width = nozzle (This PR default) is beneficial, but even more so if it is compensated for by either slightly increasing the density or increasing the flow.
Width < nozzle (This PR) also produces a good finish with or without compensation.
Increasing the flow can be an excellent way to compensate, but if you exceed it, you may get poor finishes.
Reducing the flow because it further reduces spacing can help, but you will not get the best results.
Increasing the density of internal and external bridges to 125% to increase bonding in smaller pieces and avoid ironing: Thanks @valerii-bokhan !!!

Bridge Flow Density

BRIDGE FLOW < 1 CHANGE THE SPACING.
BRIDGE FLOW > 1 CHANGE THE FLOW IN THE SAME PATTERN.

Previously, it was recommended to reduce the Bridge Flow or density to compensate the spacing between lines and generate a thinner first bridge layer with better cooling, which would then serve as a base for the next layers.

With this new setting, the idea is the opposite: to maximize contact and bonding between the lines, so increasing the Bridge Flow or using Thick Bridges further improves the finish surface.

Bridge Density

Due to the erratic behavior of flow compensation (depending on whether it is greater or less than one and whether thick bridges are enabled or not), increasing the density of the bridging pattern seems to be a more effective way to compensate.
REMOVED FROM THIS PR, @MakeSometh1ngWonderful implemented it here #11283

Thick Bridges

Clarification how it changes layer high to match nozzle diameter.

Tests

My personal test data is in the table here and discord chat.

Similar results where found by @RF47 tests, the best results are achieved with:
Line width = 100% of nozzle size
Bridge flow of 150%
Bridge speed of 10 mm/s

509235257-7a7c292a-1605-4243-9fc4-aff23c2ca3c4

Credits

@RF47 's width main idea and testing
Slic3r Bridge Flow Math
@MakeSometh1ngWonderful density idea and videos (The BEST settings for bridges and Absolutely PERFECT Unsupported 3D Printed Bridges).
@valerii-bokhan his ultra small bridge testing.
THIS BEAUTIFUL COMMUNITY!!! Thanks all for your participation.

close #10516
close #11412
close #11954

*Originally created by @ianalexis on 11/3/2025* Closes #11209, #11412 and #11954 # The Issue Currently, bridges calculate their path with a line width equal to “Internal Solid Infill” or, if this is not defined, to the general setting. Most of the time, these have a value greater than the nozzle size, since when pressed against the previous layer they can generate a greater width and improve the bond between layers. The problem here is that bridges have nothing to press against, as they are printed in mid-air. # The solution Separate the line width value for bridges in order to correctly use 100% (line width) so that each bridge line is welded to the adjacent line without compromising printing time (if the overall line width were reduced). Capped at 100% or equal nozzle size. ![bridge_line_width_1](https://github.com/user-attachments/assets/fb69c60a-dd1e-4329-891b-08a04bb3dbd7) ![bridge_line_width_2](https://github.com/user-attachments/assets/d0c4ce05-3062-444e-8551-f976b64f7f74) <img width="1584" height="813" alt="imagen" src="https://github.com/user-attachments/assets/7bd9e37e-0d2d-467c-bc34-294620bb44f8" /> ## Flow and bond Perform various tests considering the following points: - Connection of lines in the short section - Connection of lines in the long section - Quality of the bottom layer (how even it is, penalizes under- and over-extrusion) - Quality of the top layer (penalizes over-extrusion) On average, the following conclusions can be drawn: - Width > nozzle (CURRENT) is a mistake and has the most negative impact. Compensating for the flow or density to counteract the effect is feasible but not precise. It varies greatly depending on the shape and length of the extrusion. - Width = nozzle (This PR default) is beneficial, but even more so if it is compensated for by either slightly increasing the density or increasing the flow. - Width < nozzle (This PR) also produces a good finish with or without compensation. - Increasing the flow can be an excellent way to compensate, but if you exceed it, you may get poor finishes. - Reducing the flow because it further reduces spacing can help, but you will not get the best results. - Increasing the density of internal and external bridges to 125% to increase bonding in smaller pieces and avoid ironing: Thanks @valerii-bokhan !!! <img width="4032" height="3024" alt="imagen" src="https://github.com/user-attachments/assets/416d48d6-84b0-4f74-b838-79fa6ca7ea3d" /> <img width="885" height="401" alt="imagen" src="https://github.com/user-attachments/assets/432226d1-8fd9-43f6-827f-20855b4037ba" /> ### Bridge Flow Density BRIDGE FLOW < 1 CHANGE THE SPACING. BRIDGE FLOW > 1 CHANGE THE FLOW IN THE SAME PATTERN. Previously, it was recommended to reduce the Bridge Flow or density to compensate the spacing between lines and generate a thinner first bridge layer with better cooling, which would then serve as a base for the next layers. With this new setting, the idea is the opposite: to maximize contact and bonding between the lines, so increasing the Bridge Flow or using Thick Bridges further improves the finish surface. ### Bridge Density Due to the erratic behavior of flow compensation (depending on whether it is greater or less than one and whether thick bridges are enabled or not), increasing the density of the bridging pattern seems to be a more effective way to compensate. REMOVED FROM THIS PR, @MakeSometh1ngWonderful implemented it here #11283 ### Thick Bridges Clarification how it changes layer high to match nozzle diameter. #### Tests My personal test data is in the table here and discord chat. <img width="1354" height="849" alt="imagen" src="https://github.com/user-attachments/assets/30bad6bc-be6f-4adf-8474-b0c34ded19b5" /> Similar results where found by @RF47 tests, the best results are achieved with: Line width = 100% of nozzle size Bridge flow of 150% Bridge speed of 10 mm/s <img width="2992" height="2992" alt="509235257-7a7c292a-1605-4243-9fc4-aff23c2ca3c4" src="https://github.com/user-attachments/assets/06556e91-206f-4802-8e0c-ad0c27c47305" /> # Credits - @RF47 's width main idea and testing - [Slic3r Bridge Flow Math](https://manual.slic3r.org/advanced/flow-math) - @MakeSometh1ngWonderful density idea and videos ([The BEST settings for bridges](https://youtu.be/xQBLv3cPUbo?si=8dj4Rq8cKriL6k-e) and [Absolutely PERFECT Unsupported 3D Printed Bridges](https://youtu.be/eaasEkFULKE?si=SO-9Xr-_1t6p3sMy)). - @valerii-bokhan his ultra small bridge testing. - THIS BEAUTIFUL COMMUNITY!!! Thanks all for your participation. close #10516 close #11412 close #11954

Sign in to join this conversation.

No Label Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted Community testers wanted

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github/OrcaSlicer#1722