Minimum Support Distance?  

Page 1 / 2
  RSS
AbeFM
(@abefm)
Reputable Member

I've been trying to take advantage of the fine stepping of the Prusa to allow me to fine tune my support distance in pursuit of printing small (~3 mm dia) spheres.

I have done several long prints to compare support distance, and find little to no difference even with significant changes.

 

z-contact-distance XY dist % of 0.2 Support Object
As printed Distance
0 50 1.54 1.69 0
0.001 50 1.24 1.69 0.3
0.01 50 1.23 1.69 0.31
0.05 50 1.19 1.69 0.35
0.1 50 1.14 1.69 0.4
0.2 50 1.04 1.69 0.5
0.5 50 0.74 1.69 0.8
         
z-contact-distance XY dist % of 0.2 Support Object
As printed Distance
0 5 1.54 1.69 0
0.05 5 1.19 1.69 0.35
0.1 5 1.14 1.69 0.4
0.2 5 1.04 1.69 0.5
0.5 5 0.74 1.69 0.8

It looks like the minimum support distance of 0.3 mm - significantly larger than the 0.2 claimed - could be causing my issues.

 

What can I do to restrict the support distance to be 1 (or even less) layers?

 

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 7:13 pm
Nikolai
(@nikolai)
Noble Member

If you're laying down an object on support (without any bridges) then the distance will always be the same as the gravity just push it against the support. But if you're supporting overhang or bridge, then the distance plays a bigger role. Extruded line cools down as it slowly moves toward the support. Depending on many factors the extruded line will touch more or less the support. So you will actually never be able to measure 0.1/0.2 distance what you've set in the slicer. If you could, then you don't need support 🙂

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 11/07/2019 7:22 pm
bobstro liked
AbeFM
(@abefm)
Reputable Member

Repeat data from above, plus more

 

It seems independent of:

First layer thickness.

Repeated Slicing

X-Y Distance

Contact Z Distance

 

It seems DEpendant on:

layer height.

 

 

z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error  
0 0.08 50 1.54 1.62 0 0  
0.001 0.08 50 1.22 1.62 0.32 0.319  
0.1 0.08 50 1.12 1.62 0.42 0.32  
0.2 0.08 50 1.02 1.62 0.52 0.32  
0.5 0.08 50 0.72 1.62 0.82 0.32  
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error  
0 0.1 50 1.52 1.63 0.01 0.01  
0.001 0.1 50 1.22 1.63 0.31 0.309  
0.1 0.1 50 1.13 1.63 0.4 0.3  
0.2 0.1 50 1.02 1.63 0.51 0.31  
0.5 0.1 50 0.73 1.63 0.8 0.3  
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error  
0 0.15 50 1.54 1.69 0 0  
0.001 0.15 50 1.24 1.69 0.3 0.299  
0.01 0.15 50 1.23 1.69 0.31 0.3  
0.05 0.15 50 1.19 1.69 0.35 0.3  
0.1 0.15 50 1.14 1.69 0.4 0.3  
0.2 0.15 50 1.04 1.69 0.5 0.3  
0.5 0.15 50 0.74 1.69 0.8 0.3  
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error  
0 0.15 5 1.54 1.69 0 0  
0.05 0.15 5 1.19 1.69 0.35 0.3  
0.1 0.15 5 1.14 1.69 0.4 0.3  
0.2 0.15 5 1.04 1.69 0.5 0.3  
0.5 0.15 5 0.74 1.69 0.8 0.3  
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error  
0 0.2 50 1.45 1.69 0.04 0.04  
0.001 0.2 50 1.25 1.69 0.24 0.239  
0.1 0.2 50 1.15 1.69 0.34 0.24  
0.2 0.2 50 1.05 1.69 0.44 0.24  
0.5 0.2 50 0.75 1.69 0.74 0.24  
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error First Layer
0 0.3 50 1.58 1.88 0 0 125%
0.001 0.3 50 1.48 1.88 0.1 0.099 125%
0.1 0.3 50 1.38 1.88 0.2 0.1 125%
0.2 0.3 50 1.28 1.88 0.3 0.1 125%
0.5 0.3 50 0.99 1.88 0.59 0.09 125%
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error First Layer
0 0.3 50 1.5 1.8 0 0 100%
0.001 0.3 50 1.41 1.8 0.09 0.089 100%
0.1 0.3 50 1.31 1.8 0.19 0.09 100%
0.2 0.3 50 1.21 1.8 0.29 0.09 100%
0.5 0.3 50 0.91 1.8 0.59 0.09 100%
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 7:43 pm
AbeFM
(@abefm)
Reputable Member

One last thing - after some reading on GitHUB - being reminded of how I switched to CURA over a year ago... then switched back for the MMU support.... This issue has been around FOR EVER.

I discovered turning off 'detect bridging perimeters' fixes the issue:

z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error Notes
0 0.15 50 1.54 1.69 0 0 No DetectBridge
0.001 0.15 50 1.54 1.69 0 -0.001 No DetectBridge
0.1 0.15 50 1.44 1.69 0.1 0 No DetectBridge
0.2 0.15 50 1.34 1.69 0.2 0 No DetectBridge
0.5 0.15 50 1.04 1.69 0.5 0 No DetectBridge
               
z-contact-distance Layer height XY dist % of 0.2 Support Object As printed Distance Error Notes
0 0.2 50 1.45 1.65 0 0 No DetectBridge
0.001 0.2 50 1.45 1.65 0 -0.001 No DetectBridge
0.1 0.2 50 1.35 1.65 0.1 0 No DetectBridge
0.2 0.2 50 1.25 1.65 0.2 0 No DetectBridge
0.5 0.2 50 0.95 1.65 0.5 0 No DetectBridge

 

So... I'd suggest Prusa add tooltips to explain how the features don't do what they say - and I would strongly suggest they MAKE them do what they say.  I'll see how it prints now.

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 8:20 pm
AbeFM
(@abefm)
Reputable Member

Perhaps "Loose Bridge Support" should be an option which retains the current behavior, and "Tight Bridge Support" would be the version where the supports keep the same distance above bridges and non-bridges (hence leaving out weird factors of nozzle diameter, etc).

You could change either "Loose" or "Tight" to "Normal" if you prefer the term.

 

Posted by: nikolai.r

If you're laying down an object on support (without any bridges) then the distance will always be the same as the gravity just push it against the support. But if you're supporting overhang or bridge, then the distance plays a bigger role. Extruded line cools down as it slowly moves toward the support. Depending on many factors the extruded line will touch more or less the support. So you will actually never be able to measure 0.1/0.2 distance what you've set in the slicer. If you could, then you don't need support 🙂

Those are distances generated by the reported printing height of the layer, the "as printed" actually means reported heights less one layer height.

This post was modified 2 years ago by AbeFM
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 9:39 pm
Nikolai
(@nikolai)
Noble Member

Sorry, I can't really follow you and based on the feedback, looks like I'm not the only one.

The included image in the first post (300-microns.png) is showing an object laying completely on the support. Any real print measurements would give  you full contact.

Most of the entries in the tables are not really self-explanatory. It would be great if you could explain them in more detail. What is your reference object, how do you measure all the values and where do you get them from.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 11/07/2019 10:02 pm
bobstro liked
AbeFM
(@abefm)
Reputable Member

Pardon a poorly thought out post.

I was trying to correct it, when I hit upon the issue (a software... "decision" in the slicer), and hadn't gotten back to making corrections. I will edit the top post ASAP. Thanks for pointing it out.

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 10:08 pm
--
 --
(@-2)
Illustrious Member

Be sure you set the support and infill to use variable layers ... that is, don't connect them every 1 layer.

It is always wise to get more than one opinion... as for trusting Prusa? No way man....
Posted : 11/07/2019 10:13 pm
AbeFM
(@abefm)
Reputable Member

Pardon a poorly thought out post.

I was trying to correct it, when I hit upon the issue (a software... "decision" in the slicer), and hadn't gotten back to making corrections. I will edit the top post ASAP. Thanks for pointing it out.

-----------

Guess you can't go back. 😛

 

First one correction: In the image, I am only showing the top layer of support interface (dark green), the intervening layers, and stopping on the first layer with the object (orange) in it. Every layer shown excepting the top and bottom are empty layers between the supports and the printed object.

 

Now the little key I was trying to write up when I got torn away:

Here's the variables, explained - 

The literal setting used to slice:
z contact distance, Layer height, X-Y Dist %

And reported distances in slicer - moving the slider until the object first appears and recording the z-height in mm where the slicer reports it.
Support, Object

Calculated Values:
As Printed Distance = (object) - (Support) - (layer height)
Error = (As Printed Distance) - (z contact distance)

TLDR:  the software makes a decision to use a significantly larger z-gap than you ask for. If you need close supports, turn off Detect Bridging Perimeters, or consult the github issue #102

In the meantime, since the developers have thought this behavior reasonable for literally years running, I thought instead of convincing them to change their mind, maybe they can just make it an option, see github issue #2620

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 10:24 pm
AbeFM
(@abefm)
Reputable Member
Posted by: Tim

Be sure you set the support and infill to use variable layers ... that is, don't connect them every 1 layer.

I am checking that out.... Synchronize with object layers is off, grayed out, actually.

I looked because that sounded familiar, but I'm not sure what settings you're specifically referring to. Can you elaborate?

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 11/07/2019 10:31 pm
Nikolai
(@nikolai)
Noble Member

Thanks for the links and explanation. Now I understand the issue. Once you activate the bridge detection for support, the bridging layer/flow logic is being applied to all supports. In this case you will get different z-gap distance what you're actually set in the slicer. Depending on the model it might be good or bad.

That's why bubnikv kept it like it is. You can't make everybody satisfied. As you mentioned, if you really need close support, just deactivate the bridging detection.

Another useful hint and a tool in my tool box!

 

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 11/07/2019 10:57 pm
--
 --
(@-2)
Illustrious Member

The Enable Variable [x] and the Sync Object [ ]...

It is always wise to get more than one opinion... as for trusting Prusa? No way man....
Posted : 12/07/2019 1:58 am
AbeFM
(@abefm)
Reputable Member

If you uncheck "Sync Object" the wipe tower turns off - and I still need purge on color changes. So it may well be there is no choice here.

 

I suppose a purge bucket would work.

 

Anyway, I did some printing with the supports closer - it's better, but it still remains challenging to print small spheres.

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 12/07/2019 5:59 pm
Nikolai
(@nikolai)
Noble Member
Posted by: abraham.m

Anyway, I did some printing with the supports closer - it's better, but it still remains challenging to print small spheres.

Time to get PVA filament? 😉

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram...
Posted : 12/07/2019 6:24 pm
AbeFM
(@abefm)
Reputable Member

I have some, and I'm trying it. I seem to need a shocking amount of purge (~250 um^3) so have any layer adhesion on my TPU/PLA.

 

I've used it on a dual nozzle printer with some success. It's reasonable PVA will end up being the answer (I've used PETg also, it doesn't really stick, either). But it saves time and money if I can just print with TPU supporting TPU - I would just like to do the best job possible, and in this case that means being able to set the support to a value and have it get there.

 

Having a variable called "z-contact distance" that doesn't set the z-distance is like an emergency fire escape door that leads to a closet with a "ha ha" sign.

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 12/07/2019 6:32 pm
AbeFM
(@abefm)
Reputable Member

Bump!

 

Can someone, PR related or otherwise, tell me how to set the Z-distance to a specific value while leaving "detect bridging" on?

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 30/07/2019 12:54 am
--
 --
(@-2)
Illustrious Member

I just did some tests on an unrelated part and found using 0.01 mm was successful, as well as 0.05 and 0.10 mm distances. And all provided a detectable change in contact distance, with 0.01 being almost full contact, as expected.

It is always wise to get more than one opinion... as for trusting Prusa? No way man....
Posted : 30/07/2019 1:51 am
AbeFM
(@abefm)
Reputable Member

That's great news! Hopefully it's a settings issue.

 

Do you have detect bridging perimeters set ON? Can you share, or even export, your settings?

I attached a picture showing how I got the distance.

 

I tired to attach a 3mf, not sure if it'll work.

 

This post was modified 2 years ago 2 times by AbeFM
I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 30/07/2019 2:44 am
--
 --
(@-2)
Illustrious Member

test

It is always wise to get more than one opinion... as for trusting Prusa? No way man....
Posted : 30/07/2019 4:21 am
AbeFM
(@abefm)
Reputable Member

If I slice it, AS IS, with only the manual support enforcer, I get some layers that have the right spacing.

 

I put the object on its side (loop at the top), set supports to "everywhere", and both set ups have a 0.41 mm spacing.

 

I would suggest trying it with Support Interfaces on, it makes it easier to see, but doesn't change result.

 

Have you loaded my attachment to see how it slices?

I maintain an informal list of San Diego, CA 3D printing enthusiasts. PM me for details. If you include a contact email and I can add you to the informal mailing list....
Posted : 30/07/2019 10:16 pm
Page 1 / 2
Share:

Please Login or Register