The Beta 1.5 Discussion Continued

[funBart]

The last part of the ''m not dead" discussion copied to be continued, as the thread got way to long.

Please post only Beta related issues here, and open a new topic for other (new) issues. Browse and search first if there already posts about your question.

There are as well some links mentioned in the welcome message above. 

Bart

lonesock (Jonathan)

Sep 9

One day I will learn to not change code right before compiling and uploading a build.  But obviously not today!

I fixed the bug, and another where changing between parallel and sequential mode didn't trigger a re-compilation of the paths.  Thanks for the quick feedback!

thanks,

Jonathan

Attachments (2)

KISSlicer_Win64_1_4_6_2.rename2zip

888 KB   View   Download

KISSlicer_Win32_1_4_6_2.rename2zip

632 KB   View   Download

Mark as complete

Andrea Pirazzini

Sep 9

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

Attachments (1)

Immagine.png

15 KB   View   Download

Mark as complete

PenskeGuy

Sep 9

It should be Wall or Skirt, yes? Wall is just a multi-level Skirt. When you slice with Skirt, and then reselect Wall, without re-slicing, the Skirt just jumps up to the max height of the part; but it remains in the same place on the bed. How are you getting both?

On Tuesday, September 9, 2014 1:07:31 AM UTC-7, Andrea Pirazzini wrote:

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

This post has been edited

Mark as complete

lonesock (Jonathan)

Sep 9

I think I know what happened.  Skirt is now disabled and invisible in the Raft menu, but if it was already selected in the profile, it was retained as being enabled.  I will go fix that now.

thanks,

Jonathan

Mark as complete

lonesock (Jonathan)

Sep 9

Oh, one more quick question on the Prime Skirt / Wall: I can think of 3 more modes for the Wall, 

• The Wall goes up until all extruders in use have been primed at least once
• sort of an extended Prime Skirt so that all extruders get primed
• this will probably be taller than a single layer Skirt, but shorter than the Full Wall
• The Wall goes up until there is only 1 extruder left 
• so it can be used as a prime / wipe for changing extruders
• This will be taller than the previous mode, but often not the full height of the model
• User Specified height
• model dependent, so harder to fit in the UI
• would need a default value for when the Prime mode is set to this, but the model's prime height hasn't been set.
• I don't really like this one, IFF the other 2-4 modes would do everything required.

So, do these seem useful to anyone?  They are relatively easy to put in place (well, not the 3rd mode), and if they are at all helpful, I would like them in.  *However*, if they are hard to explain well and cause more confusion than help I may not want them in.

Opinions?

thanks,

Jonathan

Mark as complete

me (funBart change)

Sep 9 (19 hours ago)

Option 2 should do the trick for me, sir, as it doubles as wipe and prime wall, which will result in ultimate beautiful multi-colored prints in my humble opinion.

Option 1 is very limited and option 3 is a little vainglorious and swanky.

LearningNewWordsBart

- show quoted text -

Mark as complete

plexuss

01:20 (18 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

me (funBart change)

01:44 (17 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Of course it's Jonathan's call, but I got the impression he's still focusing on the beta. If you don't mind to put it on hold a little while longer until that is finished, he can make his mind up.

Thanks!

Bart

On Wednesday, 10 September 2014 01:20:53 UTC+2, plexuss wrote:

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

Michael Skupien

02:19 (17 hours ago)

I really like the option 2 as it should help with multi colour.  One thing that i noticed though with the prime pillar is that sometimes, the ooze from the other extruders would build up and cause the non used extruders to knock the pillar down, a major issue. It would be nice to be able to add a brim the the prim pillar too, or in this case, the prime skirt.  This would eliminate potential skirt delamination.

In addition, im not sure if this is possible by playing with the <nextx>/y commands, but it would be extremely nice if the print head could inteligently move fully (all extruders outside the prime skirt) away and back at the same level as the skirt to brush off any ooze from the nozzles.  Currently, durring an extruder change, i cool the unused extruder below oozing temp, but there is still a small (2mm long) sting on the cooled extruder that could be brushed off. As well as a similar ammount on the extuder that just started, but this is knocked off durring priming.  

I have been forced to run with cooler temps so that the extruders cool faster and only ooze a couple mm, but it is starting to cause layer adhesion issues.

Thanks for all your efforts on such an awesome software, keep it up,

Michael

Mark as complete

Jesper Lindeberg

09:46 (9 hours ago)

Option 2 is fine i think. 

Did you look more at the new seam option? 

With the new seam settings it makes more sense to have seams on straight lines, than on corners. Could you make a check box to select corner/straight line?

Thanks

[lonesock (Jonathan)]

Good idea to split this thread, Bart, thanks!

So, here's an update: (WARNING: LARGE-ish BUG, only try this one if you want to specifically verify the changes listed below)

• Seam can happen in the middle of a straight line now (I may make this optional: "anywhere" or "prefer corners")
• Even when the slice plane is exactly on a surface, the perimeter is not blocky
• BUG: Sometimes there are some artifacts *outside the part* when the slice plane is right on a surface.
• I am starting to really dislike this bug!
• definitely check all paths before printing with this version

Various other things:

• regarding the Prime Skirt / Wall
• I like the idea of a possible brim on the Prime Wall...unfortunately that means I need to add another option
• I will add the second mode (go as high as the last extruder change)
• any success stories with it yet?
• regarding the forum
• I am fine with another location, but there were some really good posts / pictures, etc. on the old forum.
• I can probably get a backup of the old database.  I know nothing about this...is porting to a new forum even possible?
• Will it be possible to populate whichever forum we choose with the "best of" threads from here?
• regarding sleep
• I need some

That's about all my brain can process at the moment.  [8^)

thanks, everyone,

Jonathan

EDIT: removed the attachments...a newer one is a few posts down.

[Andrea Pirazzini]

Found a bug with dual extrusion. Look at the video attached (is better than photos to understand). V1.4.6.3

[lonesock (Jonathan)]

Here's a new one with which to play.  [8^)

v1.4.6.4 Changes:

• Fixed the problem with blocky paths when the slicing plane is right on a model surface.
• Changed some code to be much more robust with good meshes, and hopefully w/ some bad ones too. (bit slower though)
• The "Inflate Raft" control sets the Prime Skirt / Wall gap too. (so can't use the prime with a raft yet)
• Fixed the bug Andrea found where an object which didn't use the same extruders as other objects on the same Z plane would have the Prime wall paths out of order (that was a fun one to find!!)
• Added the Prime "Short Wall" mode, where the Prime Wall will go to the layer of the last extruder change plus 1.

Please test this version, I think it is fairly robust, possibly even useful.  [8^)

thanks!

[Andrea Pirazzini]

Ok perfect now is working! Thanks!

[lonesock (Jonathan)]

Great!

Here is an updated version: v1.4.6.5.  The only change is the starting point of the prime wall will rotate around, so there will not be a weak seam point in the wall.  Also, multiple extruders on the same slice will each have an offset starting point, also for a stronger wall.

Unless someone (Bart? [8^) finds a big bug in this version, I will try to get Mac and Linux builds up for this version.

enjoy!

Jonathan

EDIT: An updated version is down a few posts.

[lonesock (Jonathan)]

Well, I found 2 bugs, one big and one small.  The big one is, multiple instances of the same object with a prime wall could very well crash, and the small bug is the packer doesn't take the prime gap (using the raft inflate value) into account.

I will be working on these and post an update ASAP.

thanks,

Jonathan

[funBart]

OOOOOOOoooooooooo!!!

<new tokens!>

I love new tokens!

Especially about <seconds>!

BUT!

you still have to explain to me where I can use the <ACCEL> token.

As it's part of the printer profile, you only can use it once....

When it was a part of the material profile you could use it to speed up the support extruder and slow down the object extruder.

Even better: when there was an option to apply it *ONLY* to the perimeter path and another to infill and support paths..., than we would talk business....

Desperately searching for Bugs! Can't find one! AAAAAARG!
Bart

[funBart]

Any change to add a setting for extra suck and prime at extruder change? Or a <NEXTE> token?

As my stupid Makerbot chockes on to many G92E0 at extruder change....

So

G92 E0

G1 E -2

G92 E0

is resulting in no extruder change (position).

Question: what is happening when putting something like this at extruder change with Makerbot firmware (no G92 E0):

G1 E <E+1> 

I couldn't figure it out with the Gcode. It seems to add about E10 or something.

Bart

 

[lonesock (Jonathan)]

Here is an update, fixing the multiple-instances-of-multiple-objects-with-Prime-Wall bug.  Packing again works as expected.  [8^)  Thanks to Bart's reminder, I fixed the time tokens, but...

Bart, I changed them internally.  Instead of just displaying seconds, which was annoying since there was no division, I am actually displaying time in HH:MM:ss.  They are based on a running total of the estimated print time, and compared against a (slightly different) estimate of the total time.  The tokens are now: <TTOTAL>, <TELAPSED>, and <TREMAIN>.  I was mostly thinking of using <TREMAIN> and probably <%> to populate the display every layers.

I am not exactly sure what the <E> token is doing...the code looks like I would expect, but I am getting weird values as well, let me look at that some more.  It might work when using the firmware mode "Abs E without Reset"...untested.

thanks,

Jonathan

EDIT: removed outdated builds

[funBart]

EDIT:

Is it intended that the new Prime/Skirt wall is switching each layer per extruder? I mean, one layer, the outside wall is extruder 1, and the second layer the outside wall is extruder 2. It's not a problem when printing with the same material, but when using PLA and ABS, there is probably a bounding problem.

I think I found the E token bug.

The E token is now taking the E value of the same extruder, but one layer down.

so:

 Absolute E mode 

T0 layer 1

ends with E100

    toolchange / G92 E0

T1 layer 1

T1 layer 2

    toolchange / G92 E0

T0 layer 2

Deselect extruder takes the E value from T0 layer 1 (so 100)

Select extruder takes the E value from T0 layer 1 (so 100)

G92 E0

continues printing

Makerbot mode

Makerbot mode has the same behavior, and is partly correct. I think there is a mixup between Makerbot and absolute E mode. Maybe it's easy for you to find by knowing this (if I'm correct).

To be complete I think Makerbot mode should work like this.

To be more clear: T0 takes A values and T1 B values:

T0 layer 1

A0

A100

     toolchange

Deselect gcode takes A100

Select Gcode takes B0

T1 layer 1 B0

T1 layer 2

B80

     toolchange

Deselct Gcode takes B80

Select Gcode takes A100

T0 layer  2 A100

T0 layer 3  etc...

Bart

[Davide Ardizzoia]

Argh...some time ago i pointer out about [e] token delayed behaviour...
By the way, Bart, how did you fine out the new tokens ?
Brgds
Davide Ardizzoia

[Bart]

Davide, I thought I was smart I found that E behavior ;-) So you already found that bug a while ago? Why is that Lonesock not listening to you? Because you are Italian? Can't be that..... ;-)

I found out about that tokens simply by trying the previous Beta: they were mentioned in the Printer//Gcode fields under the [*] N layers input field.

Bart

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.
To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/853b3037-00e3-4fab-b9db-cd33a526b3be%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[lonesock (Jonathan)]

Davide, sorry I missed that report!  Or at least it didn't make it on my ToDo list, which given my memory is functionally the same thing.  [8^)

Thanks to both of you.  Could you please try out this version 1.4.6.7?  The problem may be fixed.

thanks,

Jonathan

EDIT: newer ones are posted below.

[funBart]

It seems to work well in MB mode. Only thing is: it rounds to 2 decimals, in stead of the usual 4 decimals. I don't know if that's a problem though.

Bart

[funBart]

About the time tokens: would it be possible to force them always containing the same amount of characters? 

To force them in a 4x20 display on different rows, it would be handy that you know on forehand how many characters the time display has:

So 5 minutes would be 00:05:00, always 6 digits, and not 5:00 as now.

Bart

[lonesock (Jonathan)]

Good idea...should I do the same for the %?  (Probably use spaces instead of a leading '0', so for e.g. it would be "  9.5")

Jonathan

[lonesock (Jonathan)]

This version matches the tokens' precision to the rest of the G-code, and the time values are now fixed width (00:00:00).

thanks,

Jonathan

[Isaac Thompson]

i've missed so much stuff over the summer! now i have to catch up....i should find the latest download and try it out. i have sliced a few things lately and the version i've got is slicing great, but i wanna try out the new stuff! i'll post some pics pretty soon

[frozen...@yahoo.com]

I'm still seeing what i consider a minor bug, on a layer change, KS will print on the same island/section that was last on the previous layer causing too much heat build up on that section, instead of jumping to another section/island which gives enough time for the last island to cool before the next layer.

Say on a standing figure when printing the legs, correct example would be left leg/ right leg/ layer change/ left leg/ right leg, etc. But KS often does left leg/ right leg/ layer change/ right leg/ left leg, printing the next layer starting with the last section of the previous layer.

[funBart]

Time Tokens in action....

M70 P5 (KISSlicer ROCKS!    Total <TTOTAL>      Elapsed <TELAPSED>    Remain <TREMAIN>)

For Marlin M117 is used, but I don't know if it takes P5, I think it has to be M117 P5000 or M117 S5 to display the text for 5 seconds.

Bart

[Andrea Pirazzini]

Hi Funbart

Good! have you set them on N [*] LAYERS  Tab ?

It will be possible to insert those codes on marlin Firmware and have them on a sub-menu?

Il giorno domenica 14 settembre 2014 17:40:19 UTC+2, funBart ha scritto:

Time Tokens in action....

M70 P5 (KISSlicer ROCKS!    Total <TTOTAL>      Elapsed <TELAPSED>    Remain <TREMAIN>)

For Marlin M117 is used, but I don't know if it takes P5, I think it has to be M117 P5000 or M117 S5 to display the text for 5 seconds.

Bart

On Sunday, 14 September 2014 17:18:51 UTC+2, frozen...@yahoo.com wrote:

[Bart]

Indeed at the N layers tab, and in the startGcode / extruder change as well. 

I only know basics about Marlin, so I can't help you there. It would be nice to have a submenu though, as now you have to wait for a layer change to display the message. To have access all the time would be nice indeed.

Bart

--

You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/56cd746c-85e9-4be0-9703-bd4b23e1dbe0%40googlegroups.com.

[EdK]

It makes sense to stay on the one island and do it first for the next layer, when only considering speed.

But for the best quality, it is better, as you say, to do everything is precisely the same order each layer.

Perhaps this could be a Style choice, or change depending on the Fast/Precise setting (at a given percentage that could be selected?). It would be OK if it only stayed on the same island when Fast/Precise was set to 0%, or less than 25%, for example.

Ed Koffeman

On 9/14/2014 10:18 AM, frozen.rabbit via Kisslicer Refugee Camp wrote:

I'm still seeing what i consider a minor bug, on a layer change, KS will print on the same island/section that was last on the previous layer causing too much heat build up on that section, instead of jumping to another section/island which gives enough time for the last island to cool before the next layer.

Say on a standing figure when printing the legs, correct example would be left leg/ right leg/ layer change/ left leg/ right leg, etc. But KS often does left leg/ right leg/ layer change/ right leg/ left leg, printing the next layer starting with the last section of the previous layer.

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/f372ad45-1a84-49b2-bcec-4e096fb97e98%40googlegroups.com.

[Joris [van Tubergen]]

In marlin you can set display text with the M117 gcode... maybe that helps...

[i think it is only the bottom line]

cheers \ joris

www.rooiejoris.nl

www.facebook.com/europerminutedesign

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/5415C8EC.5060109%40kinetric.com.

[Franz Scherz]

I used M117 to display one status line on my Repetier SW (similar to Merlin).

It displays only one line for me in which I would like to have the current layer number, the total layer number and the time left. My postprocessor just calculated the % of the pront and the height in mm as you see the picture below.

Is there a variable which shows the number of total layers? This would allow me to do what I want without a post-processor!

[garr...@gmail.com]

I would like to see the behavior described below. 

[Michael Skupien]

I am currently printing a part with the new skit to last ext change setting, and it doesnt seem like there is a retraction or a wipe when transitioning off of the skirt and onto the part

[Davide Ardizzoia]

Why not keep Prime Pillar ?
Especially with very large prints (ie.: 600x300mm), the prime skirt/wall add AGES....

Best Regards
Davide Ardizzoia

[Bart]

Well, the new wall skirt is nice for dual extrusion to wipe the ooze of the inactive extruder. But that only works for small objects, so that inactive extruder actually crosses frequently that wiper-wall.

So I agree that the original Prime Pillar has it's use as well, especially for larger dual -or more- colored objects.

Why not keep both?

Bart

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/72598d65-f20c-400e-9a4c-739be76ae09c%40googlegroups.com.

[lonesock (Jonathan)]

That should be straight-forward to do...and maybe the extra gap could be implemented as additional pillar radius.  Does it really add that much time?  I know it adds quite a bit of plastic, but I thought I had it running at solid speeds....I will need to check that.

Jonathan

[Bart]

That extra time is how you see it: compared to a fairly solid print it's ok, but when printing a more fragile structure, it's indeed time consuming.

something like this is resulting in 30% time to print the object the rest to print support and walls:

I have a similar bracelet printed with the new wall feature, combining ABS with PLA, both white, so difficult to see  (PLA to dissolve with caustic soda):

That took long compared to the actual print. I had a too high temperature, so excessive oozing here ;-)

See the problem of sticking PLA/ABS on the wiper-wall, because of thermal stress torn apart, but all kept together enough to function as expected.

But for a dual test object the wiper-wall turned out nice (0.3mm layer, very fast printed, so not great looking):

See the Ooze NOT at the print but at the wiper wall.

BTW: I don't get why the extruders don't have their own wall (like it was with the prime pillar). Now they switch each layer: not nice for a combo of PLA and ABS, barely attaching to each other.

Bart

2014-09-17 23:51 GMT+02:00 lonesock (Jonathan) <lone...@gmail.com>:

That should be straight-forward to do...and maybe the extra gap could be implemented as additional pillar radius.  Does it really add that much time?  I know it adds quite a bit of plastic, but I thought I had it running at solid speeds....I will need to check that.

Jonathan

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.

To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee-camp+unsub...@googlegroups.com.
To post to this group, send email to kisslicer-refugee-camp@googlegroups.com.

Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/8cb812ef-91ed-41d4-b911-995873ccd342%40googlegroups.com.

[lonesock (Jonathan)]

I didn't use one wall per extruder because it takes a lot of extra material to keep a full N walls in the case where one extruder only gets used infrequently (e.g. only at the top of a print).  I can use the same behavior as the prime pillar used to, if that is deemed worthwhile.  Or, Yet Another Option... [8^/

Thoughts?

Jonathan

[Bart]

Ok, but it seems on the Path view to be two walls, one extrusion width from each other.

No further ideas. But as reference: Makerware is using a zig-zag wall for T1 at the left side of the object and a zig-zag for T0 at the right. But that's not usefull for 3/4 extruders maybe , as they all have another config probably (2x2, or 1x4).

--

You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee-camp+unsub...@googlegroups.com.
To post to this group, send email to kisslicer-refugee-camp@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/90bfa503-0606-4897-9d00-e3434dda0984%40googlegroups.com.

[lonesock (Jonathan)]

There are 2 walls for any layer with 2 extruders active, 3 for 3, 1 for 1, etc.  And they are done inside to out, that way if the outer loop's material droops (with no loop beneath it), it will catch on the outside of an existing wall, instead of dragging onto the part.

Jonathan

[Bart]

ok, didn't notice that, as I mostly print objects with the other extruder printing the support touching the build plate.

I suppose your concept will work most of the time, especially when printing with the same type of material..

Maybe it's a little more difficult when the wiper wall is really stacked, resulting in 10mm PLA single wall, above that 10mm ABS single wall etc, resulting in minimum attaching and pushing the ABS wall off.

I suppose it's best to keep all how it's now, maybe adding the earlier Prime Pillar, and wait for experiences of other users.

Bart

 

2014-09-18 0:53 GMT+02:00 lonesock (Jonathan) <lone...@gmail.com>:

There are 2 walls for any layer with 2 extruders active, 3 for 3, 1 for 1, etc.  And they are done inside to out, that way if the outer loop's material droops (with no loop beneath it), it will catch on the outside of an existing wall, instead of dragging onto the part.

Jonathan

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.

To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.

Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/386247af-28b3-4378-894d-e5eae729207a%40googlegroups.com.

[PenskeGuy]

I would seriously recommend not doing away with the Prime Pillar. Its uses and advantages are many. Doing so would limit a lot of users to using old versions. Being able to control the size of it and its placement would be a good enhancement. Sometimes, I would prefer placing it in the center of a toroidal part.

Why is the wall idea even being considered to replace the Pillar? The Pillar uses enough material as it is. That's why I'd like to adjust its size. The wall will use a ton of material. I would only use it on problem parts.

Question: The tooltip says "good for cooling small parts". How? Blocking the airflow doesn't seem like it would cool better. Further, I understood that the wall idea was to keep ABS parts warm to limit warping.

Color me confused.

@Bart
How did you get the holdoff from the part? When I try a slice, the wall is right up against the part—far too close for my liking; especially after looking at the third image in your 3:06 PM post, above. Yikes!

[Matej Rozman]

I have to agree with Penske on this one. Wall is just waste of material, it doesent do any good, if you want to print somewhat bigger, wall will peel/deform from the plate, if you print small you really dont need wall.
It would be okay though if you could make wall with birm so it would at least have some stick area to the plate, but Id still choose pillar over wall any time.

Again, supporting Penske`s wish to make prime pillar adjustable in size /diameter and user placeable. I have this on my wish list since the first time printing with pre1.5.

Best regards,

Matej

Dne četrtek, 18. september 2014 06:53:41 UTC+2 je oseba PenskeGuy napisala:

[lakyrc]

Hi all,
hopefully this is the place where you are collecting the problems with 1.5 beta. Here's one I found and maybe not known yet:
- Using 1.5 Beta Pro, 'Default Extruder Mapping' is not taken into account when generating the G-Code. Example:

- All four parameters are set to Ext 2 (T1)
- The generated G-Code still starts with:
  G54
  M108 T0

where it should generate
  G54
  M108 T1

Thanks.

[Tom Tumilty]

HELP please not sure why my prints look like this any suggestions, there is so much information here its hard to follow maybe this has been brought up before sorry if it has but I can't find it 

On Wednesday, 10 September 2014 13:40:20 UTC-4, funBart wrote:

The last part of the ''m not deadmy" discussion copied to be continued, as the thread got way to long.

Please post only Beta related issues here, and open a new topic for other (new) issues. Browse and search first if there already posts about your question.

There are as well some links mentioned in the welcome message above. 

Bart

lonesock (Jonathan)

Sep 9

One day I will learn to not change code right before compiling and uploading a build.  But obviously not today!

I fixed the bug, and another where changing between parallel and sequential mode didn't trigger a re-compilation of the paths.  Thanks for the quick feedback!

thanks,

Jonathan

Attachments (2)

KISSlicer_Win64_1_4_6_2.rename2zip

888 KB   View   Download

KISSlicer_Win32_1_4_6_2.rename2zip

632 KB   View   Download

Mark as complete

Andrea Pirazzini

Sep 9

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

Attachments (1)

Immagine.png

15 KB   View   Download

Mark as complete

PenskeGuy

Sep 9

It should be Wall or Skirt, yes? Wall is just a multi-level Skirt. When you slice with Skirt, and then reselect Wall, without re-slicing, the Skirt just jumps up to the max height of the part; but it remains in the same place on the bed. How are you getting both?

On Tuesday, September 9, 2014 1:07:31 AM UTC-7, Andrea Pirazzini wrote:

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

This post has been edited

Mark as complete

lonesock (Jonathan)

Sep 9

I think I know what happened.  Skirt is now disabled and invisible in the Raft menu, but if it was already selected in the profile, it was retained as being enabled.  I will go fix that now.

thanks,

Jonathan

Mark as complete

lonesock (Jonathan)

Sep 9

Oh, one more quick question on the Prime Skirt / Wall: I can think of 3 more modes for the Wall, 

• The Wall goes up until all extruders in use have been primed at least once
• sort of an extended Prime Skirt so that all extruders get primed
• this will probably be taller than a single layer Skirt, but shorter than the Full Wall
• The Wall goes up until there is only 1 extruder left 
• so it can be used as a prime / wipe for changing extruders
• This will be taller than the previous mode, but often not the full height of the model
• User Specified height
• model dependent, so harder to fit in the UI
• would need a default value for when the Prime mode is set to this, but the model's prime height hasn't been set.
• I don't really like this one, IFF the other 2-4 modes would do everything required.

So, do these seem useful to anyone?  They are relatively easy to put in place (well, not the 3rd mode), and if they are at all helpful, I would like them in.  *However*, if they are hard to explain well and cause more confusion than help I may not want them in.

Opinions?

thanks,

Jonathan

Mark as complete

me (funBart change)

Sep 9 (19 hours ago)

Option 2 should do the trick for me, sir, as it doubles as wipe and prime wall, which will result in ultimate beautiful multi-colored prints in my humble opinion.

Option 1 is very limited and option 3 is a little vainglorious and swanky.

LearningNewWordsBart

- show quoted text -

Mark as complete

plexuss

01:20 (18 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

me (funBart change)

01:44 (17 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Of course it's Jonathan's call, but I got the impression he's still focusing on the beta. If you don't mind to put it on hold a little while longer until that is finished, he can make his mind up.

Thanks!

Bart

On Wednesday, 10 September 2014 01:20:53 UTC+2, plexuss wrote:

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

Michael Skupien

02:19 (17 hours ago)

I really like the option 2 as it should help with multi colour.  One thing that i noticed though with the prime pillar is that sometimes, the ooze from the other extruders would build up and cause the non used extruders to knock the pillar down, a major issue. It would be nice to be able to add a brim the the prim pillar too, or in this case, the prime skirt.  This would eliminate potential skirt delamination.

In addition, im not sure if this is possible by playing with the <nextx>/y commands, but it would be extremely nice if the print head could inteligently move fully (all extruders outside the prime skirt) away and back at the same level as the skirt to brush off any ooze from the nozzles.  Currently, durring an extruder change, i cool the unused extruder below oozing temp, but there is still a small (2mm long) sting on the cooled extruder that could be brushed off. As well as a similar ammount on the extuder that just started, but this is knocked off durring priming.  

I have been forced to run with cooler temps so that the extruders cool faster and only ooze a couple mm, but it is starting to cause layer adhesion issues.

Thanks for all your efforts on such an awesome software, keep it up,

Michael

Mark as complete

Jesper Lindeberg

09:46 (9 hours ago)

Option 2 is fine i think. 

Did you look more at the new seam option? 

With the new seam settings it makes more sense to have seams on straight lines, than on corners. Could you make a check box to select corner/straight line?

Thanks

[lonesock (Jonathan)]

The Prime Pillar will return.  

The reason for the Wall instead of the Pillar was for sequential prints.  When printing sequentially, having a single pillar obviously wouldn't help much, so I went to the Wall to have a 1-prime-per-extruder-per-layer-per-model.  Once that functionality was in, I was hoping this would be the 1-size-fits all solution, letting me get rid of the Pillar and any associated options.  Since it seems to *not* be a solution for all the world's problems, I will add an option for the pillar again.  [8^)

Regarding using the wall for part cooling vs keeping warm, it's kind of both.  "Cooling" is too strong a word, but it does allow time where the head is not touching a small plastic island, letting it gel a bit, much like the prime pillar would.  However, the wall as a whole should help the part's overall temperature stay a bit higher.  IMO the ideal temperature for the entire printed part would be somewhere just under melting, so the wall seems like a win here.  Of course, the placement of any fans will affect all this.

See, this is why I ask for feedback...I know full well my decisions are reasonably rational, but based entirely on my (limited) experiences.  You guys are like a world-class think tank, focused directly on 3D printing problems and solutions, with the experience to back up your suggestions!  [8^)

Jonathan

[lonesock (Jonathan)]

OK, try this version.  

You can now select a Prime Skirt / Pillar / Short Wall / Wall.  The "Inflate Raft / Prime Gap" also affects the Prime Pillar's radius (and the base diameter is a little smaller than it used to be).  The Prime Pillar paths are not quite as thick as they used to be (they had been 1.5 * extrusion width, now they are 1.0 * extrusion width).  

I'm guessing someone will ask for both the skirt and the prime pillar at the same time (but I am hoping not).  [8^)

Jonathan

[Davide Ardizzoia]

One step closer to perfection!

BUT

IMHO having the wall sitting on the outer edge of the raft is not extremely useful. IMHO I would think of a minimal distance between object and wall.

Best Regards
Davide Ardizzoia

[lonesock (Jonathan)]

Yeah, the raft and skirt/wall don't play well together yet.  The raft is generated around the bed contact surfaces and any support base.  For example, if you had a cone, point downward, the raft would only be around the point (if support wasn't needed).  Now, the skirt / wall paths are computed around the entire object (based on what I call the solidified shadow of the object), that way the wall can grow to the height of the entire model without hitting anything.  So, to put the wall on the raft, the raft will need to learn about the actual wall shape.  

There are 2 ways to do it...the easiest being to use the exact same full part shadow.  The downside is, since I solidify the shadow for the wall (that way there aren't multiple interior walls, or missing exterior walls, like if your model consisted of 4 individual pillars), this potentially makes the raft area much larger.  The other technique would be to add the shadow of the wall to the raft computation.  This has the downside of potentially having a disjointed raft, which may not even be a downside.  I am leaning toward method #2 unless it proves very difficult to program.

thanks,

Jonathan

[Boris Camelo]

Jonathan 

Great Prime Pillar back!. Can you include a new feature over prime pillar? this feature likes a fill pillar. For each actual extruder, do the circle and fill the circle. Additional mix wall with prime pillar it's good idea, because with the prime pillar stabilizes the filament and external wall protect of  ooze from the other extruders.

Thank you!

Great job for  the best slicer software.

[frozen...@yahoo.com]

I'm seeing a slight inflate on the second layer of a 2 layer skirt with square sided models. 3rd and 4th layers of a thicker skirt match the 2nd layer.

[Dreide]

I think that layer adhesion within the wall/pillar (and adhesion of the wall/pillar to the bed or raft) is a problem which should be paid more attention to. Not only might one like to print with two different materials which do not necessarily stick well together on a one-loop basis, also calibration of the relative extruder positions is an issue here as the extrusion positions might not be perfectly aligned and on the same spot. Considering this, I don't think that narrowing the extrusion width for the wall/pillar was such a great idea. Which brings me back to an old request - having a parameter for the number of loops in the wall/pillar. Increasing the extrusion width for the prime/pillar might also be an option but comes at the risk of getting the extruder in a state which is sub-optimal for printing the part. Satisfying all the different requirements and preferences regarding dual extrusion will not be possible without adding more parameters to the user interface, I'm afraid.

[PenskeGuy]

On Thursday, September 18, 2014 8:45:40 AM UTC-7, lonesock (Jonathan) wrote:

The Prime Pillar will return.  

Great!
 

The reason for the Wall instead of the Pillar was for sequential prints.  When printing sequentially, having a single pillar obviously wouldn't help much, so I went to the Wall to have a 1-prime-per-extruder-per-layer-per-model.  Once that functionality was in, I was hoping this would be the 1-size-fits all solution, letting me get rid of the Pillar and any associated options.  Since it seems to *not* be a solution for all the world's problems, I will add an option for the pillar again.  [8^)

Could mini-pillars be generated in that case? A lot of parts, each having their own surrounding wall, would be a lot of material.
 

Regarding using the wall for part cooling vs keeping warm, it's kind of both.  "Cooling" is too strong a word, but it does allow time where the head is not touching a small plastic island, letting it gel a bit, much like the prime pillar would.  However, the wall as a whole should help the part's overall temperature stay a bit higher.  IMO the ideal temperature for the entire printed part would be somewhere just under melting, so the wall seems like a win here.  Of course, the placement of any fans will affect all this.

Ah, so its Contact Cooling that is what is cited. Got it. Might add that and Part Temperature Stabilization to the tooltip.
 

See, this is why I ask for feedback...I know full well my decisions are reasonably rational, but based entirely on my (limited) experiences.  You guys are like a world-class think tank, focused directly on 3D printing problems and solutions, with the experience to back up your suggestions!  [8^)

And pains in the @$$, too. :)

[Davide Ardizzoia]

Jonathan, while you are working in the support area, what about more control also on grid raft size/spacing ?
The pull on large parts is quite strong: having more control on basement (ie: large and thick but spaced first raft layer) would be great.

Best Regards
Davide Ardizzoia

[giovanni.v]

Yes, when the solid support is enabled I would like tho have it printed the same way the solid path are.

Having each layer crossed with the previous, instead of having all the layers printed with the same direction.

This will give a better quality of the support's surface and this will result on better quality of the part's surfaces printed on supports.

[funBart]

  The other technique would be to add the shadow of the wall to the raft computation.  This has the downside of potentially having a disjointed raft, which may not even be a downside.  I am leaning toward method #2 unless it proves very difficult to program.

To me it seems that method 2 would be sufficient.

Thank you to implement the two  different prime possibilities. I think together with the sequential print and different types of needs and prints, everybody has enough possibilities to choose from now.

I tried the method you described: it was working very well with two PLA extruders: the layers were two extrusion were placed were attaching very well, despite there were no extrusions under the outer wall.

I think I have found a little bug (of course...).

The Short wall seems only responsive to the support structure generated. When turned off (wit a multi-colored object) it's only one layer, in stead of generated to the last extruder change.

See pics.

The test stl's are attached.

[PenskeGuy]

Having some UI anomalies in 1.4.6.9.

When switching between Skirt, Pillar, Short Wall, Wall, sometimes the Wall won't come back. Other times it does, but once it goes AWOL, it's gone forever and a re-slice is necessary. Spurious. It's a great feature to not trigger a re-slice between choices unless you adjust the Inflate.

When slicing with Raft on and Pillar selected, the Raft is generated beneath the Pillar. Switching to Skirt, Short Wall or Wall and back to Pillar, the Raft under the Pillar disappears. This leaves the Pillar floating 2 layers up.

On the UI drop down, I'd suggest switching the order of Skirt and Pillar. Skirt, Short Wall and Wall are all the same path, just different heights. Pillar is a different path.

Further, why are these betas, supposedly of 1.5, numbered 1.4.6.x?

[Ewald Ikemann]

Am Samstag, 20. September 2014 22:09:31 UTC+2 schrieb PenskeGuy:

... Further, why are these betas, supposedly of 1.5, numbered 1.4.6.x?

This is usually made this way.
In this case "1.5" is the announcement of the new version number after the beta phase is closed.
For this is still beta and there is still work on it, the version numbers are 1.4.m.n to keep track of the development.

You can get further information here (Wiki) and here (Wiki).

Eddy

[PenskeGuy]

I'm not new to this. All of the programs I've ever been involved in, and it's been quite a few, were numbered 1.5b.x.y.z.

[lonesock (Jonathan)]

Hi, all.

It's been a busy week.  I will look at the skirt / pillar / wall issues.  Regarding the order for printing islands for each layer, I did get a regular "sweep" direction going, so each layer will try to basically go from one end to the other (only not in a stupid way), so you won't get 2 back-to-back layers on the same island of plastic when there are other options.  It's a tiny bit slower, but on the order of 1 minute per couple-hours print (unless your "rapids" are in fact "slows" [8^).  I will try to post that one in a bit.

Regarding the versioning, I am using a Code::Blocks plugin for auto-versioning, and it didn't quite behave as I expected (very likely my own fault).  So instead of spending time on that, I chose to use the "asymptotically approaching 1.5 numerically" scheme, and spend time on KISSlicer development instead.  [8^)

thanks,

Jonathan

[PenskeGuy]

Hear that; regarding what's important ! :) Thanks for the confirm. Just ends up being potentially confusing to new users when they go to the Downloads page and it's labelled "1.5 Beta 1" but the filename is referencing 1.4: "KISSlicer_Win32_1_4_5_10.zip". The possibilities are that there is an error in the page code or the new file hasn't been uploaded.

This even caught me when a user was asking a question, saying that they were using 1.5b. I DL all of the releases, as I happen to find them in these threads (not going to the Downloads page), and try them out. When the reference was to 1.5b, I went looking through the threads for a release that I missed; thinking that the 1.4.6.xx stream had been graduated to 1.5x status, is now on the way to RC and I missed its posting here. No prob. Got it now.

[DDME-Marc]

Hi Jonathan,

I have noticed with the release 1.4.6.9 that we are now getting blank lines when saving to gcode, not seen with releases just prior. Fortunately blank lines do not appear to cause any grief with our particular printer.

And since you are always striving for perfection I noticed a typo in the G code section which you may want to fix.

Cheers,

Marc

[lonesock (Jonathan)]

Here's the latest:

• Added some code to try to make the printer not start the next layer on the same island it just finished (more of a sweep now)
• Fixed the case where the short-wall didn't go high enough (when there was only a single extruder per layer, it didn't trigger the short-wall test).
• Fixed the blank lines in the G-code when comments are suppressed (needed to leave in newlines for when the comment came after a valid command, only in the G-code includes).

thanks,

Jonathan

[PenskeGuy]

Nice. On the Wall tooltip, I'd suggest: "...good for cooling small islands, keeping the overall part a more uniform temperature, priming..."

[Arjen]

Hi Jonathan,

One more bug i think. When using the wall together with the grid option the wall is starting at the second or third layer and thus printing in mid air. I dont know what the idea was offcourse, maybe printing the wall onto the grid? or printing the wall right next to the grid at layer 0 ?

Op woensdag 24 september 2014 00:36:56 UTC+2 schreef lonesock (Jonathan):

[funBart]

Just a question: the new Brim feature is printing @ perimeter speed, isn't it?  Would it make sense to print all Brim extrusions (as well when it's more layers thick) @ first layer speed? As it's now relatively slow when defining a large Brim diameter. 

Bart

[lonesock (Jonathan)]

The new prime paths should all be at the solid speed, which on my setup is usually greater than the perimeter speed.

EDIT: if it isn't critical, I could just make it the maximum speed (of perimeter, solid, sparse).  Thoughts?

Jonathan

[funBart]

Are you sure? Or what am I doing wrong? Just looked at the Gcode: as for the first layer, brim is printed  at 20mm/s (which is my perimeter speed). 

Solid is set @ 80mm/s / sparse at 120 / first layer at 60mm/s

That Brim loops are called "loop paths"in the Gcode.

Solid speed should be fine, as it probably is responding as well to the first layer speed and the Limit Increase?

Version 1.4.6.10

BTW: the new island-hop is looking good!

Bart

[lonesock (Jonathan)]

It should be labeled "Prime Pillar Path", since I hijacked that for all Prime paths now.

Jonathan

[funBart]

Ok, but why do I get other results for Brim? See screens:

[PenskeGuy]

Wall prints on top of the Grid for me, so it isn't "in mid-air". So does Skirt, which pretty much defeats the purpose of Skirt, so you just wouldn't choose that combination.

[Arjen]

Maybe only when the grid is rectangular? my grid has some corners in it. lets see if i can make a screenshot

EDIT: attached pics, you see the problem? look at the layer heights

Op donderdag 25 september 2014 03:07:42 UTC+2 schreef PenskeGuy:

[lonesock (Jonathan)]

@Arjen: I know which cases cause that, I think.  I will be working on explicitly adding raft support for under the Prime paths.  For shapes with vertical walls it will often work as-is, but there are many cases where it won't.

@Bart: I am a total idiot.  My mind was stuck on the prime paths, but you said "Brim", not "Prime".  [8^/   I could do the solid speed for all brim paths, or do fast outer ones, then slow down to perimeter speed close to the perimeter, just like the mirror of the regular loops.  Or do all brim paths at the average of brim and solid speeds.  I just want it to stick well.  I guess I would default to just using the solid speed, and users can do a first layer speed limit if they desire slower.  Thoughts?

Jonathan

[Bart]

@Bart: I am a total idiot. 

At least you have had a taste of life we ordinary people experience every day again...

 

I could do the solid speed for all brim paths, or do fast outer ones, then slow down to perimeter speed close to the perimeter, just like the mirror of the regular loops.  Or do all brim paths at the average of brim and solid speeds.  I just want it to stick well.  I guess I would default to just using the solid speed, and users can do a first layer speed limit if they desire slower.  Thoughts?

For me it seems enough it's the solid speed, tweak-able by the first layer speed and and limit increase.

The inverse interpolated perimeter speed for brim is maybe fancy, but OTOH, with a brim of 10mm it's on average still pretty slow. 

As said earlier: I'm very happy with the new Island Hop sequence of printing. Although it's hypothetical giving more stringing, the effect of more cooling of Islands in between is giving way more positive results.

 

To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.
To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/87a5ba83-41a1-4326-8c45-08857dd79e6b%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Arjen]

Jonathan, one more thing about the wall.

see attached pictures. when dual extruding it seems like the wall layer materials are constantly switching from position. this creates a very unstable wall. especially when using PLA in combination with PVA or something like that that doesnt attach that wel to each other.

EDIT, just for the record, this was'nt the case with a pillar. 

Op donderdag 25 september 2014 17:19:03 UTC+2 schreef lonesock (Jonathan):

@Arjen: I know which cases cause that, I think.  I will be working on explicitly adding raft support for under the Prime paths.  For shapes with vertical walls it will often work as-is, but there are many cases where it won't.

@Bart: I am a total idiot.  My mind was stuck on the prime paths, but you said "Brim", not "Prime".  [8^/   I could do the solid speed for all brim paths, or do fast outer ones, then slow down to perimeter speed close to the perimeter, just like the mirror of the regular loops.  Or do all brim paths at the average of brim and solid speeds.  I just want it to stick well.  I guess I would default to just using the solid speed, and users can do a first layer speed limit if they desire slower.  Thoughts?

Jonathan

[PenskeGuy]

[Tom Tumilty]

quick question what's the purpose of the wall

On Wednesday, 10 September 2014 13:40:20 UTC-4, funBart wrote:

The last part of the ''m not dead" discussion copied to be continued, as the thread got way to long.

Please post only Beta related issues here, and open a new topic for other (new) issues. Browse and search first if there already posts about your question.

There are as well some links mentioned in the welcome message above. 

Bart

lonesock (Jonathan)

Sep 9

One day I will learn to not change code right before compiling and uploading a build.  But obviously not today!

I fixed the bug, and another where changing between parallel and sequential mode didn't trigger a re-compilation of the paths.  Thanks for the quick feedback!

thanks,

Jonathan

Attachments (2)

KISSlicer_Win64_1_4_6_2.rename2zip

888 KB   View   Download

KISSlicer_Win32_1_4_6_2.rename2zip

632 KB   View   Download

Mark as complete

Andrea Pirazzini

Sep 9

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

Attachments (1)

Immagine.png

15 KB   View   Download

Mark as complete

PenskeGuy

Sep 9

It should be Wall or Skirt, yes? Wall is just a multi-level Skirt. When you slice with Skirt, and then reselect Wall, without re-slicing, the Skirt just jumps up to the max height of the part; but it remains in the same place on the bed. How are you getting both?

On Tuesday, September 9, 2014 1:07:31 AM UTC-7, Andrea Pirazzini wrote:

Hi Jonathan

Little Bug with Wall and Skirt enabled. The path are crossing. But i think that if i use the wall why i have to use the skirt to purge the nozzle right?

This post has been edited

Mark as complete

lonesock (Jonathan)

Sep 9

I think I know what happened.  Skirt is now disabled and invisible in the Raft menu, but if it was already selected in the profile, it was retained as being enabled.  I will go fix that now.

thanks,

Jonathan

Mark as complete

lonesock (Jonathan)

Sep 9

Oh, one more quick question on the Prime Skirt / Wall: I can think of 3 more modes for the Wall, 

• The Wall goes up until all extruders in use have been primed at least once
• sort of an extended Prime Skirt so that all extruders get primed
• this will probably be taller than a single layer Skirt, but shorter than the Full Wall
• The Wall goes up until there is only 1 extruder left 
• so it can be used as a prime / wipe for changing extruders
• This will be taller than the previous mode, but often not the full height of the model
• User Specified height
• model dependent, so harder to fit in the UI
• would need a default value for when the Prime mode is set to this, but the model's prime height hasn't been set.
• I don't really like this one, IFF the other 2-4 modes would do everything required.

So, do these seem useful to anyone?  They are relatively easy to put in place (well, not the 3rd mode), and if they are at all helpful, I would like them in.  *However*, if they are hard to explain well and cause more confusion than help I may not want them in.

Opinions?

thanks,

Jonathan

Mark as complete

me (funBart change)

Sep 9 (19 hours ago)

Option 2 should do the trick for me, sir, as it doubles as wipe and prime wall, which will result in ultimate beautiful multi-colored prints in my humble opinion.

Option 1 is very limited and option 3 is a little vainglorious and swanky.

LearningNewWordsBart

- show quoted text -

Mark as complete

plexuss

01:20 (18 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

me (funBart change)

01:44 (17 hours ago)

Re: [kisslicer-refugee-camp] Re: KISSlicer 1.5 preview (and I'm not dead!)

Of course it's Jonathan's call, but I got the impression he's still focusing on the beta. If you don't mind to put it on hold a little while longer until that is finished, he can make his mind up.

Thanks!

Bart

On Wednesday, 10 September 2014 01:20:53 UTC+2, plexuss wrote:

Hi there. so, just curious, do you want to start using kisslicertalk.com or just keep adding to this incredibly long thread? I think a forum would do Kisslicer better justice and lead to more users when they see there is a well organized community resource available to them. It's there, set up and ready to go! Also of course I can re-organize it to better suit our collective needs. Or, if you decide you don't want to use it let me know so I can take it down. 

Plexus/brett

Mark as complete

Michael Skupien

02:19 (17 hours ago)

I really like the option 2 as it should help with multi colour.  One thing that i noticed though with the prime pillar is that sometimes, the ooze from the other extruders would build up and cause the non used extruders to knock the pillar down, a major issue. It would be nice to be able to add a brim the the prim pillar too, or in this case, the prime skirt.  This would eliminate potential skirt delamination.

In addition, im not sure if this is possible by playing with the <nextx>/y commands, but it would be extremely nice if the print head could inteligently move fully (all extruders outside the prime skirt) away and back at the same level as the skirt to brush off any ooze from the nozzles.  Currently, durring an extruder change, i cool the unused extruder below oozing temp, but there is still a small (2mm long) sting on the cooled extruder that could be brushed off. As well as a similar ammount on the extuder that just started, but this is knocked off durring priming.  

I have been forced to run with cooler temps so that the extruders cool faster and only ooze a couple mm, but it is starting to cause layer adhesion issues.

Thanks for all your efforts on such an awesome software, keep it up,

Michael

Mark as complete

Jesper Lindeberg

09:46 (9 hours ago)

Option 2 is fine i think. 

Did you look more at the new seam option? 

With the new seam settings it makes more sense to have seams on straight lines, than on corners. Could you make a check box to select corner/straight line?

Thanks

[PenskeGuy]

To both keep the part a more uniform temperature by limiting air flow around the part and to also allow for cooling of small cross section islands that suffer from contact heating when the head is otherwise not able to move to another area. Check the tooltip.

Please endeavor to delete all but the minimal amount of quoted previous messages when replying; if any are included at all. Your post is less than a one-liner, yet contained 15 or more previous posts in their entirety. I lost count.

Thank you for your understanding.

[funBart]

On Thursday, 25 September 2014 20:30:53 UTC+2, Arjen wrote:

Jonathan, one more thing about the wall.

see attached pictures. when dual extruding it seems like the wall layer materials are constantly switching from position. this creates a very unstable wall. especially when using PLA in combination with PVA or something like that that doesnt attach that wel to each other.

We discussed it, as I got the same comment, see responses earlier in this thread:

funBart BTW: I don't get why the extruders don't have their own wall (like it was with the prime pillar). Now they switch each layer: not nice for a combo of PLA and ABS, barely attaching to each other.

lonesock (Jonathan)

I didn't use one wall per extruder because it takes a lot of extra material to keep a full N walls in the case where one extruder only gets used infrequently (e.g. only at the top of a print).  I can use the same behavior as the prime pillar used to, if that is deemed worthwhile.  Or, Yet Another Option... [8^/ 

https://groups.google.com/d/msg/kisslicer-refugee-camp/UBI5gPxyc0o/v33XYB-mVPYJ

https://groups.google.com/d/msg/kisslicer-refugee-camp/UBI5gPxyc0o/58DuzEPnYe4J

 

[Tom Tumilty]

Thanks PenskeGuy sorry for all the other crap 

--
You received this message because you are subscribed to a topic in the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/kisslicer-refugee-camp/UBI5gPxyc0o/unsubscribe.
To unsubscribe from this group and all its topics, send an email to kisslicer-refugee...@googlegroups.com.

To post to this group, send email to kisslicer-r...@googlegroups.com.
Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/44c5eafe-0beb-4168-ba67-1601128bb2ff%40googlegroups.com.

[frozen...@yahoo.com]

Crowning path overlaps look a little buggy, haven't seen it overlap before like this.

[Arjen]

Ah! my bad, didn't see that. Then i would go for 2 options! ;)

Op vrijdag 26 september 2014 13:42:55 UTC+2 schreef funBart:

[PenskeGuy]

Just checked 1.4.6.10 and Inflate Support still won't accept a negative value. By default, Support always exceeds the part X/Y dimension. Rather than negative values, can it just match the part dimension and let the user expand it?

Regarding the Wall, Short Wall, Pillar, in the case of single material + Support, I can see where not having the Support extruder contributing to the Wall or the Pillar would be beneficial. For the majority, I care about accuracy only on the part. Support is done at a faster speed, accuracy really doesn't matter and it can be Stacked. Having the ability to exclude that extruder from these external elements would save time and materials.

Where I noticed this was on a part that had a long stalk up the center and didn't go to full X/Y dimension until a long ways above the bed. I had Short Wall selected, but a full 2-material Wall was generated instead. This is because the Last Extruder Change was all the way up at the point where the overall dimensions were realized and Support was no longer needed. This is just a waste of material and the time to deposit it. To get around that, Skirt can be chosen; but then you don't have the warming function of the Wall if you want it.

[frozen...@yahoo.com]

I'm getting repeating non-random start points for loops. Printing with translucent materials, I'm getting a heavy visible seam. Would be less visible if random, or moved to a corner, but the start points are in the center of the loops. This is with Jitter set to 360.

This is a box lid. 3rd image is output from KS 1.2.0.1.

[PenskeGuy]

Is this with 1.4.6.10? Not seeing anything unexpected. Jitter == 0, I get a seam wherever the Seam Angle dial points. Move the dial to a corner, that's where the seam goes. As aed previously, a readout or an input field on the Seam Angle would make placing it so much easier. Took me ten minutes to hit the corner of a cube, once the dial had moved from the 45° default position.

Jitter == 360, I get starts distributed all over, as expected.

[Bart]

I replicated the behavior Steve got (with a rectangular box with walls as he had and had the same "not random enough" result (but different concentrated paths?). Tried it with different values for Jitter and rotating the box a little as well.  I think to remember that Jonathan said there was something with straight walls with Jitter.

2014-09-28 4:41 GMT+02:00 PenskeGuy <bj-g...@imperialearth.com>:

Is this with 1.4.6.10? Not seeing anything unexpected. Jitter == 0, I get a seam wherever the angle dial points. Move the dial close to a corner, that's where the seam goes. Jitter == 360, I get starts all over, as expected.

On Saturday, September 27, 2014 4:07:04 PM UTC-7, frozen...@yahoo.com wrote:

I'm getting repeating non-random start points for loops. Printing with translucent materials, I'm getting a heavy visible seam. Would be less visible if random, or moved to a corner, but the start points are in the center of the loops. This is with Jitter set to 360.

This is a box lid. 3rd image is output from KS 1.2.0.1.

--
You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee-camp+unsub...@googlegroups.com.
To post to this group, send email to kisslicer-refugee-camp@googlegroups.com.

Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/e11b4071-0646-4b48-9e84-abfc74a2a9c3%40googlegroups.com.

[frozen...@yahoo.com]

That is from 1.4.6.10. Jitter correctly randomizing start points for perimeter, but not so much with loops. There is a seam down the middle of each side of the printed box lid. 1.2.0.1. looks to correctly randomize loop start points.

[Franz Scherz]

As mentioned already elsewhere, I would love to have a 'nearest corner' option for Jitter! 

If I use random I do not really get nice prints.

Therefore I try to set the angle usually into a corner.

[cakeller98]

I always found it tedious to go looking for the upload posts

I found a slick way to do  assuming you get individual posts via email, and use gmail.

You have to be subscribed such that you get each posts to the group as a separate email (otherwise you won't get them in your email, and this search doesnt seem to work in google groups.

In order to search for a specific type, search for:

for Mac version search:

    from:lonesock filename:Mac

for Windows

    from:lonesock filename:Win64

or

    from:lonesock filename:Win32

for Linux

    from:lonesock filename:Linux32

or

    from:lonesock filename:Linux64  

If you just want to find all posts from Jonathan that contain Kisslicer

from:lonesock filename:KISSlicer

EDIT(again): I really wish google groups worked like that!!! 

On Tuesday, September 23, 2014 1:01:45 PM UTC-7, PenskeGuy wrote:

Hear that; regarding what's important ! :) Thanks for the confirm. Just ends up being potentially confusing to new users when they go to the Downloads page and it's labelled "1.5 Beta 1" but the filename is referencing 1.4: "KISSlicer_Win32_1_4_5_10.zip". The possibilities are that there is an error in the page code or the new file hasn't been uploaded.

This even caught me when a user was asking a question, saying that they were using 1.5b. I DL all of the releases, as I happen to find them in these threads (not going to the Downloads page), and try them out. When the reference was to 1.5b, I went looking through the threads for a release that I missed; thinking that the 1.4.6.xx stream had been graduated to 1.5x status, is now on the way to RC and I missed its posting here. No prob. Got it now.

On Tuesday, September 23, 2014 9:37:19 AM UTC-7, lonesock (Jonathan) wrote:

Regarding the versioning, I am using a Code::Blocks plugin for auto-versioning, and it didn't quite behave as I expected (very likely my own fault).  So instead of spending time on that, I chose to use the "asymptotically approaching 1.5 numerically" scheme, and spend time on KISSlicer development instead.  [8^)

[Bart]

Hi Chris! Nice that you are back!

2014-09-30 22:05 GMT+02:00 cakeller98 <cakel...@gmail.com>:

I always got frustrated looking through the posts for the betas... then I realized if you search your gmail (assumign you have gmail, and are subscribed to all posts) you can search your gmail for 

  from:lonesock has:attachment

and you'll find all the posts he put out.  I may be subscribed to Jonathan tho... point is - it's a clean way to search.  (though there may be another clean way as well, but this works well for me)

On Tuesday, September 23, 2014 1:01:45 PM UTC-7, PenskeGuy wrote:

Hear that; regarding what's important ! :) Thanks for the confirm. Just ends up being potentially confusing to new users when they go to the Downloads page and it's labelled "1.5 Beta 1" but the filename is referencing 1.4: "KISSlicer_Win32_1_4_5_10.zip". The possibilities are that there is an error in the page code or the new file hasn't been uploaded.

This even caught me when a user was asking a question, saying that they were using 1.5b. I DL all of the releases, as I happen to find them in these threads (not going to the Downloads page), and try them out. When the reference was to 1.5b, I went looking through the threads for a release that I missed; thinking that the 1.4.6.xx stream had been graduated to 1.5x status, is now on the way to RC and I missed its posting here. No prob. Got it now.

On Tuesday, September 23, 2014 9:37:19 AM UTC-7, lonesock (Jonathan) wrote:

Regarding the versioning, I am using a Code::Blocks plugin for auto-versioning, and it didn't quite behave as I expected (very likely my own fault).  So instead of spending time on that, I chose to use the "asymptotically approaching 1.5 numerically" scheme, and spend time on KISSlicer development instead.  [8^)

--

You received this message because you are subscribed to the Google Groups "Kisslicer Refugee Camp" group.

To unsubscribe from this group and stop receiving emails from it, send an email to kisslicer-refugee...@googlegroups.com.
To post to this group, send email to kisslicer-r...@googlegroups.com.

Visit this group at http://groups.google.com/group/kisslicer-refugee-camp.

To view this discussion on the web visit https://groups.google.com/d/msgid/kisslicer-refugee-camp/80f05f53-ad66-451b-893d-9953d855fa1e%40googlegroups.com.

[Chris Keller]

Thanks Bart! 

I've been swamped! I'm still swamped, but a little less.

  --- Chris

[funBart]

Possible bug regarding the line-end code in the .ini file, when not  having the "comments" checked at the Printer // Firmware tab. As result the User / Prefix code is on one line in the Gcode output (windows / latest Beta)

The ini file is in the post:
https://groups.google.com/forum/#!msg/kisslicer-refugee-camp/-mW8RzwpXAg/npxL020WjTAJ 

Bart / Cordvision

[PenskeGuy]

Not seeing this on 64-bit.

[lonesock (Jonathan)]

Hey, everyone.

Just a quick update: I am working on finding a stupid bug.  Once I get it, I will sweep through this thread and try to find any bugs mentioned here, then an updated build post.  

Sorry for the silence, I will be noisier soon.  [8^)

Jonathan

[Davide Ardizzoia]

Frankly, I can live with current beta... just badly miss more user controls on grid (raft) controls, especially being able to get a wider spacing on the very first layer of grid and more control on other layers of the grid.

Bugs are never stupid, rather the opposite, if it takes time to chase them  :-)

Best Regards
Davide Ardizzoia

[DDME-Marc]

Hi Jonathan,

One more item for your sweep list please. 

When generating G codes with the latest release (1.4.6.10) I see the program ends with all commands listed in the 'deselect extruder' tab, which in my case the keep warm temperature assigned to the extruder, filament retract etc. and then proceeds with the postfix commands

If I'm not mistaken the deselect extruder tab used to be omitted when generating the program end (going by some older saved programs)?

Many thanks and cheers,

Marc 

[Tarjei Knapstad]

I tried the "Support Z-roof" option today while waiting for a print. I can see the dividing plane, but after slicing I get no supports at all...

I also found some other support weirdness. When slicing an object which has a hollow cylinder, I get the same support density on the outside no matter if I choose Coarse or Ultra. On the inside of the cylinder, the support is Coarse in the bottom half, but then suddenly switches to the same density as the outer support in the upper half. If I choose Ultra, the support is more uniform on both the inside and outside. Screenshots and STL attached.

EDIT: also uploaded my KISS settings in case it matters :)

Regards,

--

Tarjei

[Aggresive]

Hey Jonathan,

I'd like to request what is hopefully an easy fix/addition.

When the 'skirt' has finished printing, the extruder is not retracted, which means that the hotend oozes a little during the travel to the first perimeter line, and starts 'printing' without it having a good anchorage to the build plate, which means it just drags the extruded filament around the plate for a couple of millimetres.

The latest print was a 62mm travel between the end of the skirt and the start of the perimeter line, my 'trigger [mm]' is set to 10mm. So it should have retracted/wiped, but didn't.

I'm using the latest build (1.4.6.10)

excerpt from gcode;

; 'Prime Pillar Path', 0.8 [feed mm/s], 45.0 [head mm/s]

[code before this point removed]

G1 X5.33 Y141.35 E12.3837

G1 X5.33 Y127.85 E12.6094

;

; 'Perimeter Path', 0.8 [feed mm/s], 45.0 [head mm/s]

G1 X66.32 Y165.69 E12.6094 F10500

I've tried the model with a Pillar, and it destrings/wipes as expected.

; 'Prime Pillar Path', 0.8 [feed mm/s], 45.0 [head mm/s]

[code before this point removed]

G1 X149.87 Y194.21 E4.1793

G1 X150 Y196 E4.2093

;

; 'Destring/Wipe/Jump Path', 0.0 [feed mm/s], 45.0 [head mm/s]

G1 X150 Y196 E4.2093 F10500

G1 E1.2593 F1800

G1 X150 Y196 Z0.466 E1.2593 F1500

;

; 'Perimeter Path', 0.8 [feed mm/s], 45.0 [head mm/s]

G1 X109.29 Y146.82 E1.2593 F10500

G1 X109.29 Y146.82 Z0.25 E1.2593 F1500

G1 E4.2093 F1800

G1 X109.52 Y147.27 E4.2177 F2700

[lonesock (Jonathan)]

So, found my bug.  There isn't as much done as I'd like:

• Fixed blank lines in G-code when removing comments
• Fixed typo ("adter => "after")
• Brim is now at solid speed
• Prime paths are a tiny bit thicker (25% wider than extrusion width)
• Trying to place a raft under Wall, if necessary (mostly works)
• Added version and OS to the window title

I will be combing through this thread and the feature requests over the weekend.  I will try to let everyone know what I'm going to work on.  If this version works well, I will try to get the Mac and Linux builds up soon.

thanks,

Jonathan

[Davide Ardizzoia]

Started using it....

Thank you!


Best regards
Davide Ardizzoia

[frozen...@yahoo.com]

Travel moves seem a bit more cross print in this one, still repeating start points for loops.

[lonesock (Jonathan)]

I am not seeing that travel behavior.  (Is this maybe run under WINE on Linux?  I wonder if the rand() implementation is different.)  You probably did this already, but could you zip up the STL, INI files, and resulting G-code and email it to me?  I will take a look.

thanks,

Jonathan

[frozen...@yahoo.com]

I'm running XP-64bit. Not sure when this behavior started. It does it with both these .stl files. Tested .25, .35, and .5 mm extrusion widths, with same results.

Nothing off or tweaked from my normal slice settings, 0.1mm layer, 0.35 extrusion width, 360 jitter, Oversample 0.05, Crowning -1.

Just looks to me that loops are not randomizing start points. In the wall_test.stl, start points are also not very jittered for the perimeters either when looked at layer by layer, in Repetier.

Results are from latest beta 1.14, same with last beta, but is a clean slice with 1.2.0.1. Test these .stl's, I can send gcode and ini's later if you don't get the same results.

Edit: Did an ini file check by creating a new KS folder with no ini files, only changed jitter to 360 in sample style settings and I get the same center heavy start points. Start points are not exactly in center but more like jitter is only set to 30 or 45.