This topic contains 48 replies, has 21 voices, and was last updated by chanyufei 4 months, 1 week ago.
-
Posts
-
Adding a default scaling to material profiles would do it I should think, then abs profiles could add the 1.7/2% and any other materials that needed similar treatment could have it too, without the machine/software trying to work it out on the fly.
CEL!!!!! WHY U NO These basic tools!!! I use this functionality all the time.
Rotate X axis 45/90 degrees?
Rotate Y axis 45/90 degrees?
Rotate Z axis 45/90 degrees?
Unlock proportions and..
Scale X ?
Scale Y?
Scale Z?
+1 exactly
@mistsoul It must my fault as I was reading those posts in different way - as if there was accuracy problem rather just to address ABS shrinkage. If it is for ABS shrinkage, then if they end in firmware printing with other materials where there’s no shrinkage or there’s different percentage of shrinkage would be grossly incorrect. If I was right, then whole discussion was to ‘fine tune’ real movement of the head and not to address amount ABS shrinks. If it ends in firmware - wouldn’t Robox need to know about filament’s shrinking properties so it can then automatically re-adjust those settings? Not sure that it would be best way forward. AM to automatically (or semi-automatically) scale object based on what material it is going to be printed with makes far more sense… Especially if it is an option only.
@maxromantschuk I have observed similar problems in past and started daydreaming of solutions to it. One would be to post-process gcode and detect path ends and do something about them. One thing could be to deliberately lift head for a layer or two, slow down first part of the movement (away from end of extrusion point) and then speed up. That would allow better break of filament from end of extrusion path and less pull before it speeds up to next position. And I saw it not only on the first (fragile) layer but others layers, too.
Because of that I posted all feature requests which I think I’ll just move to tickets as there’s no other way to talk to CEL (and I was really trying hard to not submit many tickets giving them breathing space)…
+1 on the Variable Print Speed. Slowing it down during the important first layer helps it sticking to the bed then I can speed it up again on subsequent layers. This is especially true with Normal and Fine modes.
Prints are definitely better. I do miss some of the tools too like nozzle height adjustment tool without going through calibration.
Finnish translation is wrong in nozzle flow/opening calibration, last question to more presice.
Ended up answering wrong
Im @ work now and dont remember exactly, but how the question was made it make u answer wrong.
Other than that, its nice update indeed
@robyneil But there is a speed of the first layer in profile. You can see mm/s to something different than 25.
But I am not sure it would help. That speed is for extrusion, problem I’ve seen (not sure if it is same as yours) is when you get to end of the extrusion path, head moves at maximum speed to next start of extrusion path. That yank causes last part of the extruded path to be pulled up…
@andylion that definitively would make sense - each material to have ‘predefined’ scale in its profile. Question, then, would be - do you see it being applied to the object or not, how you deal with object layout regarding scale, what happens if you change object - do you automatically see all object swell/shrink on the plate preview, etc… Still - it would be nice for at least some kind of warning or something to prevent you forget to add 102% (or whatever) for ABS (and any other material that needs different scale).
But that would define overall speed - not the speed of first layer nor speed of head moving with full speed between path ends and new path beginnings are those that cause issue.
When I get Robox working again I’ll make a little video of it. BTW Pete - since you have revealed your presence (
) - what should I do: make a tickets ofhttp://www.cel-robox.com/forums/topic/plugin-framework-for-am/
or
http://www.cel-robox.com/forums/topic/ability-to-prepare-print-even-if-printer-is-not-attached/
or you (CEL) is aware of those?
First one would definitively allow far more tinkering and fixing of issues like this one in some post-post-processing scripts…
@clicky I presume it would multiply the base size by the material scale, then the scaling applied by the user to the objects would multiply. e.g:
10cm gets 102% to turn it into 10.2cm, user sets 50% on the model, so it gets shrunk to 5.1cm, which should print at 5cm.
@andylion but when it is applied? When you add object to the AM? If so - what happens if you change filament before printing?
Or later when you press print button? If so - how AM is going to scale objects - especially if they are already tightly squeezed on the print bed? You can scale them to overlap…
@clicky
I’d have it so anytime the software needs to check the size it just checks the baseSize * materialScale * printScale or for that matter any time it needed the scale it just checked materialScale * printScale to get it. That way if the filament or profile changes so does the scale.
It would depend on how they have already implemented how values are stored and retrieved and when such things are checked of course. An extra check or two on profile changes might be needed.
I’m not experienced at all in regular software though, I’m a mere student but while I’m studying software development I aim to do game development, so I always tend towards thinking “How would I apply that in a game architecture” which can be different, not least because there is the expectation that things can/will change every frame and so the system needs to automatically handle that sort of thing :D.
You must be logged in to reply to this topic.
