netfabb material defs: what would you like?

[jelleatprotospace 0]

The newly generated materials also seem to cause a negative extrusion to occur? Is that something that rings a bell to you?

[ddurant 0]

The newly generated materials also seem to cause a negative extrusion to occur? Is that something that rings a bell to you?

I'll look into that. I think it's just doing the default values for reversal at the moment..

For the start.gcode stuff, you'll need to copy/paste it from my program into netfabb. It's the button labeled "..." in Machine then click the Ultimaker Specific tab.

[ddurant 0]

Also, sorta again, one of the big goals here is to figure out what to feed back to the people at netfabb - I'll consider this a success if I can start telling people that they don't need my program (or some other program) because it's all built into netfabb..

If you have observations on what works well and what doesn't, please post 'em. Include as much detail as you'd like..

 

I've realised I know quite a little about the math behind the reprap (even after like 3+ years of doing it) :oops: and you've made me want to learn more to unlock more potential in the machine and enjoy it more with less frustration because of that.

Glad it's working for you!

There's very little actual math in the program! It's probably +95% windows forms stuff and xml reading/writing. All it's really doing is calculating the volumes in the print (thread length * width * height) and figuring out how much filament is needed to get that volume..

[spunkhart 0]

Still having problems with your Software... It is importing the saved profile now without tempering.

But it is still creating a weird toolpath.. I tried this on WinXp SP3 32bit and Win7 Ultimate64bit..

My Netfabb version is 4.8.0.

Here are some pictures:

creating toolpath with the shiped PLA-High Settings:

http://wirsch.com/Netfabb-standard.JPG

and here with a profile from your software (i changed no settings... just pressed "save")

http://wirsch.com/Netfabb-new.JPG

As you can see, the toolpath somehow only creates walls.. no solids...

[ddurant 0]

Still having problems with your Software...

:(

Can you post the .reprapconfig file (it may need to be .zip'ed first) here?

[spunkhart 0]

Still having problems with your Software...

Here you go (see attachment). Thanks for taking interest.

[ddurant 0]

Thanks.. It looks like your extruder speeds are off by a factor of about 15 million, which is a bit much and would explain why it's not happy. An extruder speed of 31251996 RPMs just isn't gonna work.

I suspect this is more problems with , vs . for decimal places. I'll see if I can sort out where it's getting confused - might not be until tomorrow night.

[ddurant 0]

Can you give this one a shot? If there's a netfabbconfig.xml file in the directory you keep this in, delete that before trying the new version..

Please let me know if this works!

edit: and the 'fix' here, which is a bit sleazy, is to force the program to run with English units; using a . character for a decimal place instead of a , character. You should probably keep this in mind and use . when entering numbers until I get some code that's a bit less sleazy..

edit: old version deleted - see first post in thread for the latest..

[spunkhart 0]

I haven't tried to open the files in netfabb, because i am at work now, but from the looks of the files it is much better now. Thanks.

http://wirsch.com/reprapconfig.JPG

(the left one is the new version)

I don't know how many decimal places are necessary (it seems that the software cuts off after 3) but i will give it a try when i am at home.

[ddurant 0]

I haven't tried to open the files in netfabb, because i am at work now, but from the looks of the files it is much better now. Thanks.

http://wirsch.com/reprapconfig.JPG

(the left one is the new version)

I don't know how many decimal places are necessary (it seems that the software cuts off after 3) but i will give it a try when i am at home.

Excellent. That looks much better!

[ddurant 0]

The newly generated materials also seem to cause a negative extrusion to occur? Is that something that rings a bell to you?

I'll look into that. I think it's just doing the default values for reversal at the moment..

Actually, I'm not doing default values - I'm explicitly setting them to 0. Well, I guess if the defaults are all 0, I am doing the default values. Zeros either way, though...

Do you have any more details on what you're seeing?

[joostvb 0]

Here's the program. Yes, it's windows-only (though likely works on mono) and there's no source yet.. If those are a problem, the offer to be a data entry monkey for people still stands - only takes a minute to type stuff in and get a profile.

 

Thanks, I will give it a try.

[spunkhart 0]

I haven't tried to open the files in netfabb, because i am at work now, but from the looks of the files it is much better now. Thanks.

http://wirsch.com/reprapconfig.JPG

(the left one is the new version)

I don't know how many decimal places are necessary (it seems that the software cuts off after 3) but i will give it a try when i am at home.

Excellent. That looks much better!

I am still testing it seems that the feed rate is pretty high, although I entered a fillament diameter which is bigger than my actual diameter..

It would also be nice to be able to enter the nozzle size.. when i print objects with only one layer thickness i measure around 0.37mm...

[ddurant 0]

I am still testing it seems that the feed rate is pretty high, although I entered a fillament diameter which is bigger than my actual diameter..

It would also be nice to be able to enter the nozzle size.. when i print objects with only one layer thickness i measure around 0.37mm...

Not sure I understand about the feed rate.. Are you saying just that it moves too fast? If so, tell my program to use slower feed rates!

Why did you enter a filament diameter that was incorrect?

Nozzle size doesn't matter too much for this stuff. It does define the min/max thread widths you can use but that's about it - it really doesn't have anything to do with layer height..

I think I'm confused.. Can you post more details on what you tried and what happened?

[joergen 2]

I just made a new profile for the latest netfabb (and the latest marlin):

Ultimaker Logos EPS.zip

The actual profile was using mostly fast&thing and fast&medium.

when I printed it, it was exactly the opposite from what I expected: the amount of PLA coming out was correct, the speed was grotesquely slow: something that was supposed to go at 170mm/s, was printing at something like 17mm/s, as if everything was 10x slower... the only thing that had the proper speed was the jump.

in addition, repg/merlin gave me warning when I cranked it up to 800%: too long extrusion prevented.

any pointers where to look for getting proper printing speed? (I know my printer can do 200mm/s, or is there a "hand brake" set in the latest marlin?)

[ddurant 0]

I just made a new profile for the latest netfabb (and the latest marlin):

[attachment=0]Screen shot 2012-01-08 at 1.26.56 AM.png[/attachment]

The actual profile was using mostly fast&thing and fast&medium.

when I printed it, it was exactly the opposite from what I expected: the amount of PLA coming out was correct, the speed was grotesquely slow: something that was supposed to go at 170mm/s, was printing at something like 17mm/s, as if everything was 10x slower... the only thing that had the proper speed was the jump.

in addition, repg/merlin gave me warning when I cranked it up to 800%: too long extrusion prevented.

any pointers where to look for getting proper printing speed? (I know my printer can do 200mm/s, or is there a "hand brake" set in the latest marlin?)

Can you pick a G1 command somewhere in the middle of the gcode file and look at the F component? That will be the feed rate, in mm/minute..

[joergen 2]

Can you pick a G1 command somewhere in the middle of the gcode file and look at the F component? That will be the feed rate, in mm/minute..

most of the F parameters are perfectly correct numbers:

 

M106M109 M92 E14G21G90G28G92 X0.0000 Y0.0000 Z0.0000 E0.0000G1 E0.0000M107T0.0000G1 X0.0000 Y200.0000 Z0.0000 F5400.0000 G1 X0.0000 Y200.0000 Z4.9500 F1800.0000 G1 X0.0000 Y200.0000 Z5.0000 F1800.0000 M109 S225.0000 G1 X0.0000 Y200.0000 Z9.9500 F1800.0000 G1 X0.0000 Y200.0000 Z10.0000 F1800.0000 (begin region 1)M106(begin layer 1 at 0.100)G1 X0.0000 Y200.0000 Z10.0600 F1800.0000 E0.0000 G1 X0.0000 Y200.0000 Z10.1000 F1800.0000 E0.0000 G1 X77.0500 Y100.4700 Z10.1000 F16800.0000 E0.0000 G1 X80.4779 Y96.0417 Z10.1000 F16800.0000 E0.0000 G1 X80.4779 Y96.0417 Z0.1500 F1800.0000 E0.0000 G1 X80.4779 Y96.0417 Z0.1000 F1800.0000 E0.0000 G1 X79.7900 Y96.7800 Z0.1000 F10128.0000 E2.9248 G1 X79.1500 Y97.8700 Z0.1000 F10128.0000 E6.5883 G1 X78.8200 Y98.8800 Z0.1000 F10128.0000 E9.6679 G1 X78.7000 Y99.8600 Z0.1000 F10128.0000 E12.5295 

 

I generated some gcode with slic3r, and all the straight lines had the right speed, but everytime there was some form of curve, it dropped drastically down to a crawl, maybe due to the fact that most of the G1 commands didn't have a F component:

 

; generated by Slic3r 0.5.7 on 2012-00-07 at 22:52:43; layer_height = 0.1; perimeters = 3; solid_layers = 3; fill_density = 0.4; nozzle_diameter = 0.5; filament_diameter = 2.91; perimeter_speed = 150; infill_speed = 150; travel_speed = 300; extrusion_width_ratio = 0; scale = 1M104 S225 ; set temperatureG28 ; home all axesM106 ; fan on; M109 S225 ; wait for temperature to be reachedG90 ; use absolute coordinatesG21 ; set units to millimetersG92 E0 ; reset extrusion distanceM82 ; use absolute distances for extrusionG1 Z0.100 F18000.000G1 X67.940 Y78.360G1 F1800.000 E1.00000G1 X132.060 Y78.360 F2700.000 E1.65417G1 X132.810 Y78.470 E1.66191G1 X133.310 Y78.600 E1.66718G1 X133.790 Y78.770 E1.67237G1 X134.250 Y78.980 E1.67753G1 X134.480 Y79.100 E1.68018G1 X134.700 Y79.230 E1.68279G1 X135.320 Y79.670 E1.69054G1 X135.700 Y80.010 E1.69574G1 X135.880 Y80.190 E1.69834

 

not sure if there is something strange with the latest merlin... I will go back to an older version and see if the same happens...

[joergen 2]

How about using this tool to make netfabb volumetric?

you have all the basic numbers to calculate the extrusion speed, so in theory it should be possible to go back to a M92 E8xx.xx value (which can be easily measured&calculated, instead of the random historic E14), and generate proper E values for all the combos you mentioned (75)... all that would change for netfabb are new extruder speed values, but it would be inherently volumetric...

any thoughts?

[ddurant 0]

any thoughts?

That was the original plan but since the netfabb start gcode is machine-specific, not material-specific, it would break the existing material definitions. You'd either have to use just my stuff or hand-edit gcode files if you switched back & forth..

edit: that and the firmware is slowing you down.. Those F values do indeed look correct to me.

edit2: When you connect repg or whatever to the printer, do you get some debug info from the firmware? Marlin will spit out min/max feed rates and such. If I had to bet, I'd bet you're tripping over one of those settings.

[joergen 2]

I did post it to Erik also, to see if it;s a marlin issue, or a user error on my end:

 

echo:maximum feedrates (mm/s):

echo: m203 x500.00 y500.00 z5.00 e45.00

echo:acceleration: s=acceleration, t=retract acceleration

echo: m204 s3000.00 t3000.00

 

other than the e45 value, nothing stands out as too low...

what I do not understand is how the firmware defines everything as mm/sec and the gcode feeds it as mm/min...

[ddurant 0]

what I do not understand is how the firmware defines everything as mm/sec and the gcode feeds it as mm/min...

Maybe there's a better reason for it but I suspect it's like that simply because it's always been like that..

[cyclone 0]

what I do not understand is how the firmware defines everything as mm/sec and the gcode feeds it as mm/min...

The couple of commercial machines I have, use gcode with feedrates in units/min American machines use inches per minute.

Not sure why. Probably as Dave says It was always done that way in the past.

-b

[joergen 2]

edit: that and the firmware is slowing you down.. Those F values do indeed look correct to me.

edit2: When you connect repg or whatever to the printer, do you get some debug info from the firmware? Marlin will spit out min/max feed rates and such. If I had to bet, I'd bet you're tripping over one of those settings.

I edited the max extrusion setting from e45 to e450 in the configuration.h of the latest marlin firmware, and that removed the "hand brake" setting from the netfabb code, allowing the really nice&fast prints I was used to.

I have to revisit your tool, since the profile I had created had unreasonable expectation settings in there.

On the other hand, where did you get your e865.888 as the steps/mm setting from? it is the exact number I found on my UM while measuring/calibrating my UM back in September/October 2011...

[ddurant 0]

Er.. Thought I just posted a reply but don't see it.. If it shows up, hopefully this one will won't contradict it..

 

I edited the max extrusion setting from e45 to e450 in the configuration.h of the latest marlin firmware, and that removed the "hand brake" setting from the netfabb code, allowing the really nice&fast prints I was used to.

Crap. Can't believe I missed that.. Makes perfect sense now that you figured it out.

Since the math is based on machines using E14, the G1 E values are all going to be considerably bigger than what you'd normally see when printing with sprinter/marlin and their E865 or E926 values.. Far more likely to trip over the default firmware limits and yet another reason why I think the current way of telling the FW how much you want to extrude is just wrong.. (that's a whole different story, though)

 

I have to revisit your tool, since the profile I had created had unreasonable expectation settings in there.

How so? Just too high feed or width or something? It should, in theory, work for just about any value you feed it, within reason.

If you can quantify how it's off and/or speculate on how to improve it, I'd be very interested in hearing about that..

 

On the other hand, where did you get your e865.888 as the steps/mm setting from?

I got it from you! It's what I use when I slice with SF now that I have the new bolt installed. Works great! (ty, btw)

If you're seeing that in my program, please grab a newer version. The forums here don't let me edit the OP ( :( ) but you can find it here: viewtopic.php?f=9&t=194&start=30#p882

[joergen 2]

One other thought... In a blog post about slic3r was a comment about using integer numbers for the layer height.

533.3333 steps is a mm in Z... Our favorite layer height is 0.1mm which is 53.33333 steps... Obviously something the stepperFW can't do... 54 steps is far more convenient... But it translates to 0.10125mm layer height... Anyway to integrate this into your software? Sort of a up/down arrow that translates the integer steps into mm?