hp-65

I had a lot of fun with 13.06.x during the last, past days. Until now. Until printer number three sunk its head directly into an object :-/ It looks like the new project planner ignores or misinterprets the given head sizes in the preference menu. Play with it and watch the grey zones around the models! Additionally, the lack of specifying the order of prints makes things even worse because the algorithm seems to work only towards a positive y direction. If the head size towards y-max is only 10mm, but has 80mm to y-min, one loses 70mm of space because there's no way of telling Cura it better starts with the object at y-max. And because Cura 13.06.4 does not care about the y-min size, I lost an extraordinary beautiful fan shroud in a plastics vs. plastics fight :(

An infill value of 50% matches the perimeter wall, a 100% value shoots over and leaves the object: I guess that's not suppose to happen, right?

I tested it with a 160MB file. It took less than 30s for a complete slice, but all of this happened in the background and I did not experience the slightest performance issues. Rotating, moving or scaling the object, changing input field values or opening the menus did work as usual. It rocks 8)

Oh yes, you are right! The new slicer isn't that forgiving any more. The old Cura behaviour was to simply skip the inner walls and start with the infill rightaway. 13.06.3 behaves different. But even after reducing the 0.7mm down to 0.35, the infill algorithm doesn't work as it should. This is the old behaviour, with an infill setting of 50%: This is the same in 13.06.3. As you can see, a 50% overlap setting gives a 100% match with the perimeter walls: I also tested 100%. In this case, the infill crosses the perimeter walls and "leaves the object". Someone forgot a divide by two here :wink: That's right. With Cura 13.06.3 I am going to divide this by 2 :wink: No, seriously. I spent a lot of time tuning all these machines and 60% gives perfectly strong and rigid results.

Oh no, I did not swap them. After some more experiments, it seems that the new overlap algorithm doesn't work as expected. The walls have a thickness of 1.0mm The nozzle width is set to 0.35mm, which leaves exactly 0.3mm in the middle. With an overlap of 60%, the nozzle should travel 0.7mm * 0.6 = 0.42mm. Enough to fit a 0.35mm tip. The new slicer fills the gaps up to an infill value of 42%. One percent more (43%) fails. Mhhh, let's see: 0.42 * 0.7 = 0.294mm 0.43 * 0.7 = 0.301mm This exactly matches the gap distance between the walls. Looks like the new slicer forbids the nozzle center to pass an already layed down string. I guess that's not how it should work. About to test some more, may be there are some more dependencies...

I did not test the wizard of 13.06.x because all previous versions did not work. They either moved the extruder far too slowly (few clicks in several seconds) or much too fast. Never digged deeper, but I assume it was caused by an improper M92 E... command. What about just having a window (or an import function) where one can enter all the good old, proven settings? Thanks for clarifying that! Still stuck in v13.03, but now, I can see the bed. Sufficient for home usage :wink: BTW.: A little other problem, that already happened in 13.03 too: Sometimes, depending on what other applications are running, I can not enter any values in the input fields or file dialogs! The cursor always jumps back (or forth) to a specific position in the field. This behaviour seems to correlate to OpenGL software (Had it with Rhino 3D and OpenSCAD).

Discard my vote above! I just began to like the buttonless real-time feature: http://www.youtube.com/watch?v=JMAJYALHQcI

After manually adding machine_type = ultimaker to "preferences.ini", I managed to regain control over orthogonality ;-) At least now, the Ultimaker bed appears...

For your entertainment... A few years ago, I bought a 3kg spool of a medium-rigid filament from eBay. The best stuff I ever had: Though it already got very soft above ~50°C, this filament had absolutely no warp. I used it up and received a 2nd spool a few days ago. After manually feeding the material through the extruder, I already noticed that this filament was somehow different. A lot of bubbles came out, even at completely ridiculous low temperatures. I hoped the best and printed some solid parts. Because I already used 13.06.3 for some other test prints, I did not pay special attention to the newer Cura version. Nice! Not that much difference compared to the old, completely bubble free material and 13.06.3 seemed to work great. And then came this little case: At first, I printed the shocking 13.06.3 version, the "old-13.03" print was done afterwards. Same file, same material, same settings and same machine... Cura 13.03 left and 13.06.3 on the right side: After a while, I noticed that one part of the 13.06.3 print looked great too, see lower part of the rear wall on the right. And here's the reason. The ugly top part, had this rapid move going through and the oozing material from the nozzle sat right between the two walls and bent them apart. Closeup (boosted dynamic contrast a little, doesn't look that glassy in real life): The lower part only had the two walls: Basically, two questions arise: 1) Ok, beside all these filament caused problems, obviously because of a lot of moisture in it, why didn't 13.06.3 fill the walls? Wall thickness: 1.0mm Nozzle size: 0.35mm Infill overlap: 60% Does it need the extra 0.05mm (3 x 0.35mm = 1.05mm) Cura 13.03 does this: 2) I printed about ~100kg of different PLA sorts, but I never had a problem like this. Does it make sense to dry this filament? According to this PDF (unfortunately in German only, but graphics on page 3 should be self-explainatory (moisture vs. time)), it would be required before each print: http://atr-solutions.de/app/download/5785460138/I-PDF+Mann_Hummel_KU110937_KU4_12.pdf Sorry for the subject, I just couldn't resist. I really don't blame it on 13.06.3 (yet :wink:, just needed to share this. Never experienced anything like this before.

+1 for a button

The "all objects at once" algorithm sunk the head into finished prints several times, but I never had any problems with the sequential printing option (except mixing very high and very small objects the same time). The old project planner settings moved to the main preferences, but all values are set to "0mm". Suggested values appear if you hover the mouse over the input fields. You'll need to enter them manually...

... were these intentionally left out? - retraction minimum travel - select wether to print all objects at once or sequencially - an isometric top view for the "new planner" - there's no coordinate system any more; where am I ? :wink: I fully respect your ambitions to create an UI as easy as possible, but at least to my opinion, missing any or all of these options really, really hurts...

Three short questions to the UM staff: 1) If I buy an UM today, are all of the upgrades built-in? - V2 hot end - extruder drive upgrade - knurled bolt, V3 2) Do you still offer a completely assembled machine or was this service a one-time promotion offer? 3) I read about a dual extruder upgrade. Any release date/plans yet? Thanks HP

A quick solution: - browse to your Cura directory - enter "Cura/gui" - open the file "projectPlanner.py" with an editor - look around line 158 and change self.alwaysAutoPlace = True to self.alwaysAutoPlace = False Just printed five objects "around my screw"... HP-65

Yes, but what is so great about printing at the center of the machine and nowhere else? I don't want to exchange the tape after every little print. I have plenty of unused space. Maybe I am using the print head to knock an object off the platform. Possibly I only want to specify an offset because my machine is equipped with several extruders. Maybe there's a screw in the middle of the table (*1*). What about experiments while building other, new machines and the bed is not in the middle of the printing area? And, and, and... Up to Cura 12.11, there were two easy ways of choosing the position of an object: a) machine center definitions in the "advanced config tab" b) the project planner Now, both of them are gone. The missing variables resulted in a gcode like G01 X?machine_center_x? ... and killed a print head just a few minutes ago, because it, well, hit a screw (see (*1*) again). Cura gives us the possibility to rotate or scale an object, flip it across any axis, checking the gcode output graphically, it now even can display overhangs, but I can't simply move the part? BTW.: I never used Cura for controlling any machine, because it (yet) lacks many features, I only used my modified Marlin/Display/SD-Card code or Printrun. But this time, well, is there actually code behind the "PRINT" button? It does absolutely nothing... Now, back to repairing the machine and browsing some sources 8) Have a nice Christmas HP-65 === (*1*) I am not going to discuss that ;-))

I second that. And there are much more reasons to keep this...

Regarding https://github.com/daid/Marlin Is "Marlin_v1" exactly the branch that currently ships with Cura 12.08?

A little late, but maybe this is of interest for others, who have comparable issues (though this is somehow unlikely ;-) ) I wrote some test programs and discovered some defective SRAM cells. I replaced the MCU and made a tiny step forward: At least I was able to connect to the machine, control X, Y, Z, the heater and everything else, except for the extruder. I hooked up a scope and found out that the extruder was operated at a ridiculous low speed, only a few ticks per second (on fastest extrusion settings). After browsing the source code, I found the magic and undocumented modal code M92. BTW, stuff like this is really brilliant >8-( if(value < 20.0) { float factor = axis_steps_per_unit[i] / value; // increase e constants if M92 E14 is given for netfab. max_e_jerk *= factor; max_feedrate[i] *= factor; axis_steps_per_sqr_second[i] *= factor; } Adding a M92 E865.888 which I read in a post somewhere else, feels like the extruder is running with the right speed, but at this point a lot of other problems popped up. (E865.888 is a ridiculous value, at least regarding the physical tolerances of an extruder). Until now, I wasn't able to print with with 0040 or Cura. Maybe only a few or more steps forward... ;-)

I disassembled it and found: - Mega 2560 - UM PCB rev. 1.5.4 As far as I understood, this should work with Marlin, but it doesn't... I really would appreciate any help or hints on this! What do they report? Minor issues like bad printing results or real blockers, like "nothing is working", like here? But as I already wrote, this is not (only) related to Cura. 0035 (with Marlin) does exactly the same: Absolutely nothing...

If this is true, my board is a Mega2560, because I used Cura to update the FW to Marlin, but it refused to program it back to 5D. I had to use 0025... I will open the machine and take a look inside, when I'm back home... Are there any additional, external HW changes? Temperature sensor, inverted end switches or anything else that could interfere here? I am pretty sure my board is 1.5.4... I found schematics of 1.5.3 and 1.5.4. The only differences are the fan connectors and minor layout changes around the screw terminals. Where can I find the docs for 1.5.6 (or 1.5.5)?

Hi all, I bought my UM around 9/2011. Until now, I never used anything else but 0025 with 5D FW, and I never had any problems. Recent tries to upgrade to 0035 or Cura failed due to some, yet unknown, incompatibilities with the Marlin firmware. As soon as Marlin is loaded, nothing further happens... Where can I find a description or history of the UM hardware? Do I need any new electronics or can I build a Marlin compatible version for my old stuff? thx