Inside the Ultimaker 3 - Day 1 - GCode

[korneel 176]

 

you are the best. next time i'll bring you beer.

also, is there a simple way I can get printer statistics ? like hours printed witha  core?

also : is there a way to easily change the current to the stepper motors?

 

Printer stats will be added in the API on a firmware update. We are already collecting things, but they are not yet exposed.

Currently the motor current is hard-coded in griffin/printer/controller.py

 

thank you!

when you say hard coded in controller.py.. does that mean i can change that file and it will respect the changes?

[Daid 306]

thank you!

when you say hard coded in controller.py.. does that mean i can change that file and it will respect the changes?

Yes, but you are doing something that you need to be very careful about.

[korneel 176]

thank you!

when you say hard coded in controller.py.. does that mean i can change that file and it will respect the changes?

Yes, but you are doing something that you need to be very careful about.

i understand.. but if i wanted to reduce the current that would be relatively safe no?

[erri 2]

(It contains "opinicus", "griffin-display", "jedi-marlin" as main components for example. I could go into the overall architecture in a different post)

I assume that "jedi-marlin" is the firmware that is running on the "atmega board" (jedi.hex).

The name suggests that it is Marlin based. If so, where ist the source code?

[gr5 2,170]

I have these and I love them. I feel your lack-of-sleep pain. I have a pink noise generator on my phone always running at night so I can slip these on and block out sound *and* light if I choose to.

https://www.amazon.com/CozyPhones-Sleep-Headphones-Travel-Bag/dp/B012YFVN90/

[neotko 1,417]

I have these and I love them. I feel your lack-of-sleep pain. I have a pink noise generator on my phone always running at night so I can slip these on and block out sound *and* light if I choose to.

https://www.amazon.com/CozyPhones-Sleep-Headphones-Travel-Bag/dp/B012YFVN90/

 

That might not help much for users where they get a fine for leaving the lights on at night https://ultimaker.com/en/community/23375-how-to-turn-off-or-dim-the-leds-of-the-ultimaker3#reply-162146

[Daid 306]

I assume that "jedi-marlin" is the firmware that is running on the "atmega board" (jedi.hex).

The name suggests that it is Marlin based. If so, where ist the source code?

 

It is marlin based. Based of the Ultimaker 2 firmware. I haven't put the source code online yet, didn't get around to do so yet.

Before the FOSS people start to scream things about pitchforks and torches, if you have an UM3 and want a copy of the Marlin source. Send me a message and I'll send you a copy. Same for the linux kernel, which isn't online yet due to the size of the repository not fitting on github.

(Yes, that's enough to comply with GPL, it does not state that the source needs to be online for everyone. Just that you need to provide it to people who have your product. Not doing this to be an ass. But because it is less effort right now)

[ahoeben 1,948]

Not doing this to be an ass.

 

I can confirm that only part of Daid can be considered "ass", and I never got to see it.

[erri 2]

Thanks @Daid for the info.

[neotko 1,417]

@Daid

I'll like to ask you...

I have change both extruders to a custom um3/bondtech and I'm using 1.75 with more mods.

The thing is that no matter how much I change the esteps jedi.json file I end up having one extruder doing less than the other. In the end the hotfix was to do this:

M92 Exxx after each toolchange. So I assume that UM3 is using Marlin X2 or a more advanced version than umo/um2 uses?

Does this marlin accept two esteps for each extruder? Why editing the json doesn't change both? I did ssh vi jedi.son and reboot after the change and I always get the same effect if I don't do M92 for each tool. Also tested M92 T0 Excx M92 T1 Exxx and seems to work.

Is there any plan at all to be able to access directly the printerboard to run test and experiments without needing to do the full gcode with header and all that? Is quite slow to do simple tests like just extruding 10cm of filament to test how much did actually was moved to calibrate at different temps and get an accurate estep #t.

I know the objective atm is to make all work, but I'll like to know if there are plans for the near future to allow direct access to modders. Also is there any chance I could get my hands on the new marlin used? It would help me fix or know what's causing this.

Anyhow my hotfix works, it doesn't work for cura since I can't append gcode to the print sequence unless I manually edit each print.

Edited December 2, 2016 by Guest

[Daid 306]

http://software.ultimaker.com/jedi/Marlin-2016.10.10.zip

That's the marlin as it is in the current firmware release. I see no reason why the e step change wouldn't work properly for both feeders...

The upcomming release if the firmware has a "talk to marlin" feature on the printer. We're using this during debugging as well. It's a python commandline tool that runs directly on the printer.

We pushed 3.4.90 to testing yesterday, this version contains the commandline tool. You can access it by logging in as "ultimaker" on ssh in developer mode.

Note that this testing version has a few known bugs, of which the most important one is that manual bed leveling does not properly set the Z offset between nozzles.

[neotko 1,417]

Thank you very much @daid !

[AnnieR 0]

Hi @Daid

 

I'm recently working on developer mode of Ultimaker 3. What I'm trying to do is to change the print file (usb already imported different files to the printer) by control of pc/remote selection by coding. I wonder how I would be able to achieve this function since I didn't find in headers defining/selecting the printing job.

 

Thanks,

Annie

[Daid 306]

For remote control, you don't need developer mode. You need to access the HTTP REST API.

[3dprntz 4]

I've tried looking around and can't find the gcode for the automatic bed leveling. I'm compiling a gcode that would require the printer to level the bed during the print more than once to keep the first layer height consistent.

[gr5 2,170]

There's no need.  Leveling will be consistent for months.  Even years.  I know because I've tested this. As long as you don't change print cores or add a layer of glue to the glass or similar.  Also, the leveling switch in a UM3 is consistent to much better than .1mm which is crazy surprising.

[3dprntz 4]

I've had issues trying it out. Usually I wouldn't need to level it, as I've had the same experience in the past. However, the first print has a perfect layer, but the following are too far from the plate. I tried turning off the auto leveling to see if there would be a difference and it gives me the same issue, which points me to the point that the leveling helps the first layer.

 

Thinking about it now, I can double check the full build plate leveling and get that as even as I can. But, I've done that already before. It would still be nice to know the auto-leveling gcode as it would be handy with projects in the future. From what I understand, its usually G29 on other printers, but can't be exact with the UM3.

[gr5 2,170]

This is a common problem on glass plate UM2 and UM3 printers.  The plate is tempered and produced in sheets about 2 meters wide if I remember right and UM gets the glass from the center but the glass is thicker in the middle.  Like a mountain range running down the middle of the glass in one axis.  It's subtle but you can see it with a metal ruler or straight edge.

 

Then we level on only 3 points where the 3 screws are.  So those 3 points are PERFECT.   So the glass is already thinner in the 4 corners and after leveling the front two corners are raised up.  This results in a plate that is extra low in the rear two corners.  So printing in the rear two corners is almost always too high and you get worse adhesion back there.

 

One solution is to just not print in those 2 corners.  Another is to raise the rear of the glass after leveling is done by just unscrewing the rear screw about 1/2 turn.

 

I think you can tell Cura to print the first layer lower or thicker or something maybe?  Then you could do a per-model-setting possibly?  Probably not.  You could edit the gcodes a bit to squish the prints in the rear corners a little more.

[3dprntz 4]

The print is being done on the front-center portion of the printer. I think I'll run a test without active bed leveling and have the subsequent prints with lower first layer and see if that helps. Is there no griffin gcode for active bedleveling? 

[3dprntz 4]

GitHub - Ultimaker/UltimakerMarlin: This repo contains the versions of Marlin powering Ultimaker printers. In case of questions please drop us a line: support@ultimaker.com

 

Looks like there are a couple here:

G30 - Probe Z at current position and report result.

G31 - Get/update capacitive sensor base level

M310 - Single cap sensor read

 

Could be one of these I suspect.