Firmware 1.18 provisional change list

Avatar

By dc42 13 December 2016 11:42

Administrator · 2,256 comments

With firmware 1.17 in the Release Candidate phase, I am already planning firmware 1.18. With the firmware getting so complicated, I am looking for features I can remove as well as features to add or improve. The provisional change list is:

  • Simpler heater tuning process which avoids the need to second-guess the PWM and calculates the model parameters with the print cooling fan both on and off.
  • Add feed forward term to temperature control to better maintain hot end temperature when the print cooling fan is turned on or its speed is changed. Possibly add a feed forward term to compensate for extrusion rate too - are there any Volcano users who could test this?
  • Support the Ethernet version of the Duet WiFi that is under development.
  • Add babystepping support.
  • Remove option to set bed probe points using M557 with P parameter. You can use a bed.g file instead if you are currently using them.
  • Remove temperature coefficient option from Z probe trigger height. It was added to better support ultrasonic height sensors, but I don't think anyone is still using them.

I will consider making other small changes at the same time. Projected release date is mid to late January.


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By Alex9779 13 December 2016 13:30

Member · 64 comments

My two cents to "Remove temperature coefficient option from Z probe trigger height":
This can also be interesting for inductive sensors IMHO. And I think a lot of people use them. Though a temp relation formula or sheet is not available to the cheap Chinese probes...

Avatar

By Zesty_Lykle 13 December 2016 15:28

Member · 85 comments

Can I ask for a backup option?
Full zip of the system that then can be downloaded?


Lykle
Design, make and enjoy life

Co Creator of the Zesty Nimble

Avatar

By Nyl0cke 13 December 2016 16:15

Member · 77 comments

I've got a volcano, whats up?

Avatar

By T3P3Tony 13 December 2016 16:17

Administrator · 688 comments

Lykle - That is probably something we will ask ChrisHamm to put in the WebInterface

Nyl0cke ??


Duet Wifi Hardware Designer
www.duet3d.com
www.think3dprint3d.com

Avatar

By Nyl0cke 13 December 2016 16:20

Member · 77 comments
dc42 wrote

  • Add feed forward term to temperature control to better maintain hot end temperature when the print cooling fan is turned on or its speed is changed. Possibly add a feed forward term to compensate for extrusion rate too - are there any Volcano users who could test this?

Last edited by Nyl0cke (13 December 2016 16:21)

Avatar

By T3P3Tony 13 December 2016 16:24

Administrator · 688 comments

ahh missed that!


Duet Wifi Hardware Designer
www.duet3d.com
www.think3dprint3d.com

Avatar

By croadfeldt 13 December 2016 16:53

Member · 49 comments

I have a volcano, 40 watt heater and high volume / pressure fan. So I can test whatever you'd like. I know I see a +/- 2-3 degree spread on my volcano at times, so would certainly appreciate that.

Avatar

By dc42 13 December 2016 19:49

Administrator · 2,256 comments
croadfeldt wrote

I have a volcano, 40 watt heater and high volume / pressure fan. So I can test whatever you'd like. I know I see a +/- 2-3 degree spread on my volcano at times, so would certainly appreciate that.

Thanks! To start with, please heat up your hot end until the temperature is stable. Then extrude filament through the volcano nozzle at the maximum rate you are ever likely to while printing, for about 30 seconds, while monitoring the hot end temperature. Continue monitoring it after you stop extrusion. I'm interested in how much the temperature drops when you start extrusion, and overshoots when you stop, to see if it is worth adding a feedforward term.


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By croadfeldt 13 December 2016 20:26

Member · 49 comments

Results will be here in about 10 minutes when it's done uploading.

https://youtu.be/8uE4DUzVU9E

Avatar

By dc42 13 December 2016 20:53

Administrator · 2,256 comments

Thanks, looks like the under/overshoot is about 2C.

Are you using 1.75mm or 3mm filament? If it's 1.75mm, are you able to extrude at 10mm/sec or greater when using a large nozzle on the Volcano?


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By Adamfilip 13 December 2016 21:02

Member · 100 comments

Ability to update PanelDue firmware from Webcontrol 😁

Avatar

By croadfeldt 13 December 2016 21:03

Member · 49 comments

1.75mm filament.

How large of a nozzle do you want to try? I can go from .4 up to 1.2mm.

Avatar

By croadfeldt 13 December 2016 21:42

Member · 49 comments

Will try out the 1.2 and let you know.

Avatar

By dc42 13 December 2016 22:08

Administrator · 2,256 comments

What I am thinking is that if you print with a 1.2mm nozzle at e.g. 0.6 mm layer height and 1.5mm extrusion width, then at 100mm/sec printing speed the extrusion rate for 1.75 mm filament would be about 1.5 X 0.6 X 100 / (1.75 X 1.75 X pi / 4) which is about 37 mm/sec. Maybe you can't reach this extrusion rate, but even 50mm/sec printing speed would need nearly 20mm/sec. Melting filament at this rate may cause a significant drop in nozzle temperature, which is why adding a feed forward term may be worthwhile.


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By Adamfilip 13 December 2016 23:46

Member · 100 comments

Sounds like you should have a 3mm volcano.

Avatar

By croadfeldt 14 December 2016 01:21

Member · 49 comments

We are at the limits of what the extruder can do here. 25mm/s is definitely too fast, 20mm/s still has few stepper skips. But for the purpose of this testing, we should be good. Don't mind the snapping sound, that's another issue I've brought up with E3D, I believe it's the bearings or binding in the driven gear assembly.

1.2mm nozzle - 1.75mm filament - 300mm extrudes at 1500mm and 1200mm feedrates with the fan at %50 and %0.
https://www.youtube.com/watch?v=1P9KXk67aPk

1.2mm nozzle - 1.75mm filament - 900mm extrudes at 1200mm feedrates with the fan at %50 and %0.
https://www.youtube.com/watch?v=yADr5m_hg9M

Last edited by croadfeldt (14 December 2016 01:22)

Avatar

By T3P3Tony 14 December 2016 01:44

Administrator · 688 comments

Chris - Off topic: in watching these videos I saw a lot more of your YouTube channel. I find it hard to not sit and watch my printers printing, now I get to sit and watch your printer as well - live!

Are you using Octoprint just for the video streaming or for control as well? if the printer on the stream running a Duet? Sorry forall the questions, just about to share this and want to be accurate.


Duet Wifi Hardware Designer
www.duet3d.com
www.think3dprint3d.com

Avatar

By croadfeldt 14 December 2016 02:21

Member · 49 comments

Hey, glad I could help feed the addiction or bring the zen as it were. 😁

I need to update my description as I've shifted away from Octoprint and now use just the Duet Web interface for all the on screen stats and job control. I do like Octoprint's G-Code preview and being able to display that, but with the recent OBS changes and inclusion of the browser plugin, I have a single browser session for DWC in OBS that I carve up into the on-screen display you see in the stream. Prior to switching though, I used Octoprint for the onscreen stats, so if you go back a few months you will see Octoprint used to control my RAMPs setup and for a while the DuetWifi.

Avatar

By dc42 14 December 2016 08:31

Administrator · 2,256 comments

Thanks, Chris! Looks like you are getting about 8C temperature drop when you start extruding, and 8C or 6C overshoot when you stop depending on whether the fan is off or on. Definitely worth adding a feedforward term if any one wants to print at those sort f speeds with a Volcano.

Your temperature indication looks rather jittery. Are you using a thermistor, thermocouple, or PT100?


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By croadfeldt 14 December 2016 15:15

Member · 49 comments

You're welcome, I'm using a PT100. I've noticed the same thing that it jumps around a lot. The temp reading does stabilize eventually.

Full disclosure - I suspect I have a 5 voltage issue on the board or my power supply though, as my screen backlight has been flashing constantly now for days. Will address that separately if needed after I find my multimeter... Anyhow, if the PT100 is driven by the same 5v rail, it might account for the jitter.

Avatar

By dc42 14 December 2016 19:36

Administrator · 2,256 comments

The PT100 daughter board runs from the 3.3V rail.

I have heard of another user seeing the PT100 readings jump around when the PT100 wires followed the extruder wires over some distance, so perhaps you have a similar interference issue in the cabling. The other possibility if you are using a 2-wire PT100 connection is that you have a connection in your PT100 circuit that isn't perfect.

I use a 4-wire PT100 connection and the wires follow the extruder wiring for about 300mm and the hot end wiring for 1m. Despite this, my temperature readings are rock steady.

I suggest you try monitoring the temperature readings under the following conditions:

1. With the extruder motor off (send M84 E0) and the hot end off.

2. With the extruder motor off and the hot end heater active.

3. With the extruder motor on (e.g. after extruding, with the M906 idle current factor set to 100%) and the hot end heater off.

That will tell you whether the extruder motor current and/or hot end heater current is interfering with the PT100 reading.


Duet WiFi hardware designer and firmware engineer
http://www.escher3d.com
https://miscsolutions.wordpress.com

Avatar

By (In)Sanity 17 February 2017 14:13

Member · 279 comments

Trying out this new firmware now,  I still sometimes swear bed comp doesn't do what I expect it to.   I'm thinking it's just caused by my Z axis motor and 2mm pitch lead screw just not having enough torque to hit any decent microsteps,  more like 1/4 stepping.

Last edited by (In)Sanity (17 February 2017 14:14)


Duet WiFi Powered FFCP with E3D legends hotend system.  BLTouch grid leveling.

Avatar

By (In)Sanity 17 February 2017 14:40

Member · 279 comments

Right wrong or otherwise on my part I just had a heater fault with 1.18 trying to print some PETG.   I had the first layer set to 240C without cooling and then 245 for the second layer with cooling.   I got a heater fault pretty much the instant it tried to switch to layer two saying the temp wasn't rising the 0.3 deg C;/s that was expected.   My logic was I wanted the print to stick on the first layer and then bond to itself on subsequent layers.   Trying the same print with flat 240C which I'm sure will work.


Duet WiFi Powered FFCP with E3D legends hotend system.  BLTouch grid leveling.

Avatar

By (In)Sanity 17 February 2017 15:12

Member · 279 comments

Sorry for the post flood,   is bed comp set by default to tapper off ?   Doing a test print now with a piece of tape on the lead screw so I can see any movement.  The first layer it was waving around,  now it's doing nothing at all running the same span.   

Thanks,

Jeff

Last edited by (In)Sanity (17 February 2017 15:13)


Duet WiFi Powered FFCP with E3D legends hotend system.  BLTouch grid leveling.