I thought I'd add my experiences on this subject. I've built a very large printer. The heated bed and build surface have been two of the biggest concerns, but so far both seem under control. The build area is about 870 x 800, and I'm using 6mm aluminium as the structural element, with 20+ adjustable screws on the underside supported by a 40x40x4.0mm EA aluminium frame. Heating is done with a custom bed heater from China with holes for the screws - it's about 2300W on 240VAC. I looked at getting Borosilicate glass for the build surface, but I was quoted about $1200 + freight from the states (probably another $1200), and PEI sheet was pretty similar. I ended up using a large mirror from IKEA ($60 for 2!), which I cut to size. Before the heated bed went in, I had all kinds of surface adhesion issues, but since then it's been great.
Mesh bed levelling hasn't been perfect, I get about 1-2mm max error across the whole area, but given the size I'm not concerned with that.
So far I havne't printed anything particularly large, I'm still debugging the design, but I have a project coming up that will start to use more of the area, so I'm keen to see how it goes.

Hi All,

I've had a quick look around, and I can't find what I'm looking for.
I'm wondering if there's a way to connect a digitizing touch probe to the Duet 3 board for reverse engineering purposes?

Something like this: https://www.youtube.com/watch?v=FJ8aKo_qqos

Cheers

right... There's a gap in my understanding there. Thanks for clearing that up. I'll give that a shot this weekend.

@Phaedrux, I've run the pressure advance script (here) several times, and I've picked a value that gives the best results (0.025 in my case), but it hasn't helped. Based on the explanation here it seems that pressure advance is simply a multiplier applied to the acceleration - a higher number means more extrusion during acceleration/deceleration.
In my observations however, As I increased the pressure advance value, I noticed that the mid-line regions (constant speed regions) became under extruded - for higher values the lines wouldn't touch. For low values I noticed that the ends became shorter - and for really low values not quite touching the perimeter lines.

To be honest, I've only tested this "on the fly", and I haven't dedicated any real time to it, so it's possible that other effects are at play.

The main intent of exploring this avenue was to explain my observation that the motor went backwards slightly - job done.

Hi All,

I've skimmed over this topic a little, but having just figured out this problem I thought I would add my 2c.

I have a very large bed - roughly 800x800, with 16 manual adjustment screws (4x4 arrangement), and 4 independent Z motors.

In my case, I have (at some point) needed to manually level both the lead screws, and the manual adjustment screw - I have also used G32 and G29 bed levelling.

I partially agree that there is a hierarchy to bed levelling, which needs to be done after assembly, maintenance or a component failure (like a loose screw on a Z motor coupling for example?). With so many components there's plenty of opportunity for something to go wrong, so the correct procedure helps.

I manually adjust the Z screws with the hot end close to the screws first - this gets the nozzle within a reasonable range for G32 to work its magic. Then I run G29 to do a mesh bed levelling - now that all 4 motors are coplanar this gives me a good idea on the flatness of the bed. If the min/max deviation of the mesh bed levelling is less than 1mm (an arbitrary number I've decided to use), then I'll leave it be - otherwise, I'll adjust the manual levelling screws...

A more thorough method would be to look for a deviation greater than Xmm (some number) between any two neighboring points in the mesh bed levelling, but for now I'm satisfied with maximum deviation...

On smaller printers like your typical 300x300 size, then I think a similar approach is easy enough to do, so it doesn't really matter whether it's necessary - do what ever you feel works best for you.

Thanks, that worked.

So, a quick follow up on this.

Back story - the printer is a huge CoreXY built as a MVP (minimum viable product) - there have been a lot of short cuts taken, so I expect a bunch of problems.

The job I was printing when I posted this was for a casting pattern - it took almost 6 full days to print, and the quality was very average. The blobs and zits were the biggest problem, but there were several horizontal bands that were several layers thick - kind of like the magic numbers problem scaled up a few layers.

Anyway - I made the error of printing the whole job 2% too small :'(, so I'm now re-printing it. I took some feedback on board, tried a few test prints, and turned the settings up. I disabled the Accel and Jerk control in Cura, then tweaked the M5665 and M201 values on the duet while it was printing.
We're now printing at 150mm/s (max), Jerk is 750mm/s, Accel is 700mm/s^2, and layer height has been increased to 0.32mm. All with the intention of speeding the print up firstly - and hopefully to get some improvements in quality...

Well... Here's a before and after shot of the print at roughly the same point in the print.

Before

After

It's not perfect, but it's a huge improvement.

@CaLviNx "Best approach" is subject to perspective. I agree with @Piet here - you build the most light weight frame that you can (because it's cheaper) and you let the software/hardware smarts take care of the rest for you - because it can. It's very normal for machines to require commissioning before being ready for use.

Remember that 3 points is the most stable, only if you assume a rigid body. In real life "rigid" doesn't exist. All bodies are elastic to some extent.

So, a quick follow up on this.

Back story - the printer is a huge CoreXY built as a MVP (minimum viable product) - there have been a lot of short cuts taken, so I expect a bunch of problems.

The job I was printing when I posted this was for a casting pattern - it took almost 6 full days to print, and the quality was very average. The blobs and zits were the biggest problem, but there were several horizontal bands that were several layers thick - kind of like the magic numbers problem scaled up a few layers.

Anyway - I made the error of printing the whole job 2% too small :'(, so I'm now re-printing it. I took some feedback on board, tried a few test prints, and turned the settings up. I disabled the Accel and Jerk control in Cura, then tweaked the M5665 and M201 values on the duet while it was printing.
We're now printing at 150mm/s (max), Jerk is 750mm/s, Accel is 700mm/s^2, and layer height has been increased to 0.32mm. All with the intention of speeding the print up firstly - and hopefully to get some improvements in quality...

Well... Here's a before and after shot of the print at roughly the same point in the print.

Before

After

It's not perfect, but it's a huge improvement.

@engikeneer - Thanks for the tips. Long story short, I've come to the settings I use now as a result of resolving earlier problems - for example, extruder and Z skipping because jerk was too high. Some of these problems have been resolved at the source, some haven't. Sounds like I need to go back to first principles and re-tune everything.

@engikeneer - now that you mention it, I noticed that cura jerk is set to 12mm/s, but duet is 2000mm/min (33mm/s).

I had a problem a long time back where the extruder jerk or accel was too low, so the motion was fine, but occasionally it would have to wait for the extruder. From memory it was only on retracts though.

Sadly, this is a huge job - 100+ hrs, so I'll have to wait until it finishes before I can try anything else.

Here's the config.g file:

; Configuration file for Duet WiFi (firmware version 2.03)
; executed by the firmware on start-up
;
; generated by RepRapFirmware Configuration Tool v2.1.8 on Mon Mar 30 2020 11:14:18 GMT+1000 (Australian Eastern Standard Time)

; General preferences
G90                                       ; send absolute coordinates...
M83                                       ; ...but relative extruder moves
M550 P"Point Zero CoreXY"                 ; set printer name

M667 S1                                   ; select CoreXY mode

; Network
M552 S1                                   ; enable network
M586 P0 S1                                ; enable HTTP
M586 P1 S0                                ; disable FTP
M586 P2 S0                                ; disable Telnet

; Drives
M569 P0 S1                                ; physical drive 0 goes backwards
M569 P1 S1                                ; physical drive 1 goes forwards
M569 P5 S1                                ; physical drive 5 goes backwards
M569 P6 S1
M569 P7 S1
M569 P8 S1
M569 P4 S0                                ; physical drive 3 goes backwards
M569 P3 S1
M584 X1 Y0 Z5:6:7:8 E4:3                  ; set drive mapping
M671 X-60:-60:875:875 Y200:400:200:400 S5.0	    ; Set location of z leadscrews for independant bed levelling.
M350 X16 Y16 Z16 E16:16 I1                   	; configure microstepping with interpolation 
M92 X80.00 Y80.00 Z1600.00 E412.0:412.0         ; set steps per mm
M566 X2000.00 Y2000.00 Z20.00 E5000.00:5000.0   ; set maximum instantaneous speed changes (mm/min)
M203 X18000.00 Y18000.00 Z400.00 E5000:5000     ; set maximum speeds (mm/min)
M201 X800.00 Y800.00 Z100.00 E2000.00:2000.00   ; set accelerations (mm/s^2)
M204 P1200 T4000			  			  		; print and travel acceleration
M906 X1600 Y1600 Z1200 E1200:1200 I30           ; set motor currents (mA) and motor idle factor in per cent
M84 S30                                   ; Set idle timeout
M593 F80				  				  ; Set Dynamic Acceleration Adjustment
M572 D0 S0.06				  			  ; Set extruder pressure advance
;M915 P0:1 S5 F1 H200 R3			      ; Set up stall detection


; Axis Limits
M208 X0 Y0 Z0 S1                          ; set axis minima
M208 X740 Y800 Z750 S0                    ; set axis maxima Previous X860

; Endstops
M574 X1 Y1 S1
M574 Z0 S0                          	  ; set active high endstops
M591 D0 P5 C3 R5:120 E20.0 S1		  	  ; Configure Laser filament sensor for extruder drive 0.

; Z-Probe
M307 H3 A-1 C-1 D-1                       ; disable heater on PWM channel for BLTouch
M558 P9 H5 F120 T6000                     ; set Z probe type to bltouch and the dive height + speeds
G31 P500 X-21.6 Y9.4 Z0.85               ; set Z probe trigger value, offset and trigger height - refer to: https://duet3d.dozuki.com/Wiki/Gcode#Section_M564_Limit_axes
M557 X40:740 Y40:780 S78                  ; define mesh grid - AS DESIGNED
;M557 X270:530 Y270:530 S52               ; define mesh grid - FOR TESTING ONLY
M376 H10				  				  ; Set bed compensation taper (layers to fade out bed compensation)
;G29 S1

; Heaters
M307 H0 B0 S1.00                          ; disable bang-bang mode for the bed heater and set PWM limit
M305 P0 T100000 B4138 R4700 S"Bed"        ; set thermistor + ADC parameters for heater 0
;M140 H0 S75 R50
M143 H0 S120                              ; set temperature limit for heater 0 to 120C
M305 P1 T100000 B4138 R4700 S"Extruder 1" ; set thermistor + ADC parameters for heater 1
M143 H1 S280                              ; set temperature limit for heater 1 to 280C
M305 P2 T100000 B4267 R4700 S"Extruder 2" ; set thermistor + ADC parameters for heater 2
M143 H2 S280                              ; set temperature limit for heater 1 to 280C

M305 P101 S"Duet 2" T100000 B4725 R4700	  ; Duet 2 Drivers Over Temp Warning
M305 P102 S"Duex 5" T100000 B4725 R4700	  ; Duex 5 Drivers Over Temp Warning

; Fans
M106 P0 S0 I1 F25000 H-1 X255 B0.001             ; set fan 0 value, PWM signal inversion and frequency. Thermostatic control is turned on
;M106 P1 S0 I1 F25000 H-1 X255             ; set fan 1 value, PWM signal inversion and frequency. Thermostatic control is turned off
M106 P2 S1 I0 F500 H-1

; Tools
M563 P0 D0 H1 F0:1                          ; define tool 0
G10  P0 X0 Y0 Z0                           ; set tool 0 axis offsets
M563 P1 D1 H2 F0:1                          ; define tool 1
G10  P1 X-20 Y00 Z0                          ; set tool 1 axis offsets
G10  P1 S190 R150
G10  P0 S190 R150                              ; set initial tool 0 active and standby temperatures to 0C

; Custom settings are not defined

@Marshal - Its hard to say if your problem is the same. I blame my issue on the slicer. I only have very simple start code - no M32 commands.

M122 output below:

M122: === Diagnostics ===
RepRapFirmware for Duet 2 WiFi/Ethernet version 2.05 running on Duet WiFi 1.02 or later + DueX5
Board ID: 08DJM-9178L-L2MSD-6JKD6-3S86K-19D2N
Used output buffers: 1 of 24 (14 max)
=== RTOS ===
Static ram: 25712
Dynamic ram: 94484 of which 0 recycled
Exception stack ram used: 520
Never used ram: 10356
Tasks: NETWORK(ready,596) HEAT(blocked,1232) DUEX(suspended,160) MAIN(running,1664) IDLE(ready,160)
Owned mutexes:
=== Platform ===
Last reset 78:43:40 ago, cause: power up
Last software reset at 2021-05-20 18:12, reason: User, spinning module GCodes, available RAM 10404 bytes (slot 2)
Software reset code 0x0003 HFSR 0x00000000 CFSR 0x00000000 ICSR 0x0441f000 BFAR 0xe000ed38 SP 0xffffffff Task 0x4e49414d
Error status: 0
Free file entries: 9
SD card 0 detected, interface speed: 20.0MBytes/sec
SD card longest block write time: 211.0ms, max retries 0
MCU temperature: min 25.4, current 31.8, max 36.5
Supply voltage: min 11.4, current 11.7, max 12.4, under voltage events: 0, over voltage events: 0, power good: yes
Driver 0: ok, SG min/max 0/664
Driver 1: ok, SG min/max 0/880
Driver 2: standstill, SG min/max not available
Driver 3: standstill, SG min/max not available
Driver 4: standstill, SG min/max 0/1023
Driver 5: standstill, SG min/max 0/418
Driver 6: standstill, SG min/max 0/500
Driver 7: standstill, SG min/max 0/387
Driver 8: standstill, SG min/max 0/494
Driver 9: standstill, SG min/max not available
Date/time: 2021-05-24 19:55:28
Cache data hit count 4294967295
Slowest loop: 628.52ms; fastest: 0.07ms
I2C nak errors 0, send timeouts 0, receive timeouts 0, finishTimeouts 0, resets 0
=== Move ===
Hiccups: 0, FreeDm: 158, MinFreeDm: 97, MaxWait: 1487703ms
Bed compensation in use: mesh, comp offset 0.000
=== DDARing ===
Scheduled moves: 1909091, completed moves: 1909075, StepErrors: 0, LaErrors: 1454, Underruns: 0, 3
=== Heat ===
Bed heaters = 0 -1 -1 -1, chamberHeaters = -1 -1
Heater 0 is on, I-accum = 0.4
Heater 1 is on, I-accum = 0.3
=== GCodes ===
Segments left: 1
Stack records: 2 allocated, 0 in use
Movement lock held by null
http is idle in state(s) 0
telnet is idle in state(s) 0
file is doing "G1 F6000 X387.506 Y471.215 E1.89808" in state(s) 0
serial is idle in state(s) 0
aux is idle in state(s) 0
daemon is idle in state(s) 0
queue is idle in state(s) 0
autopause is idle in state(s) 0
Code queue is empty.
=== Network ===
Slowest loop: 204.40ms; fastest: 0.00ms
Responder states: HTTP(0) HTTP(0) HTTP(0) HTTP(0) FTP(0) Telnet(0) Telnet(0)
HTTP sessions: 1 of 8
- WiFi -
Network state is running
WiFi module is connected to access point 
Failed messages: pending 0, notready 0, noresp 0
WiFi firmware version 1.21
WiFi MAC address cc:50:e3:e3:b1:11
WiFi Vcc 3.28, reset reason Turned on by main processor
WiFi flash size 4194304, free heap 16232
WiFi IP address 192.168.0.112
WiFi signal strength -55dBm, reconnections 0, sleep mode modem
Socket states: 0 0 0 0 0 0 0 0
=== Filament sensors ===
Extruder 0: pos 0.84, errs: frame 187 parity 4 ovrun 71 pol 348 ovdue 0

Hi All,

I'm been debugging a new build, and I have a problem that I've allowed to hang around for a little while, but my current print is suffering more than others so I need it solved.

I suspect it's a slicer problem, so I've posted it one the Cura foums but I know there is plenty of knowledgeable users here who might be able to help.

So apologies for referring you to the Cura forums here:

https://community.ultimaker.com/topic/37053-blobs-and-zits-stuttering-movement/

Thanks and advance

The problem has a distinctly "hysteresis" feel about it. That is to say, that travelling in one direction does not result in the same path as movement in the other direction. Belt tension would be an easy culprit for this problem, but as such, it's the first one to be resolved.

Here's what I suggest. Isolate travel direction as the problem - Print a simple thin walled square (50x50 should do), go CW for 10-20 layers, then CCW for 10-20 layers. If hysteresis is the problem you should get the same bands appearing with direction change. I imagine you'll need to generate this code manually to control the travel direction.

If this occurs, you'll need to look for anything that gives you some uncontrolled movement - it could be as subtle as flexibility in the hot end, or a pulley (or many pulleys in the case of a CoreXY) that move slightly on their shaft. Remember you're looking for problems that add up to a total of a fraction of a line width - so perhaps 0.1mm or less.

Best of luck.

@sirkris, Firstly, this isn't a problem, it shouldn't effect your print. Someone else will be able to confirm, but I suspect the reference point is taken as the average (or some other normalization) of all z heights - which may mean no single point on your bed is Z 0.000mm.

Download your heightmap.csv file (from the /sys folder), and take a look at your Z values - you may be able to figure out how it's rationalised.

@Phaedrux, I've run the pressure advance script (here) several times, and I've picked a value that gives the best results (0.025 in my case), but it hasn't helped. Based on the explanation here it seems that pressure advance is simply a multiplier applied to the acceleration - a higher number means more extrusion during acceleration/deceleration.
In my observations however, As I increased the pressure advance value, I noticed that the mid-line regions (constant speed regions) became under extruded - for higher values the lines wouldn't touch. For low values I noticed that the ends became shorter - and for really low values not quite touching the perimeter lines.

To be honest, I've only tested this "on the fly", and I haven't dedicated any real time to it, so it's possible that other effects are at play.

The main intent of exploring this avenue was to explain my observation that the motor went backwards slightly - job done.

@CaLviNx "Best approach" is subject to perspective. I agree with @Piet here - you build the most light weight frame that you can (because it's cheaper) and you let the software/hardware smarts take care of the rest for you - because it can. It's very normal for machines to require commissioning before being ready for use.

Remember that 3 points is the most stable, only if you assume a rigid body. In real life "rigid" doesn't exist. All bodies are elastic to some extent.

Mystery solved.

I started working with @Phaedrux's suggeestion of using Slic3r, and I noticed that they didn't have a "Lines" pattern for the top/bottom layers. So in order to keep the test fair, I used "zig zag" in Cura to keep the playing field level.

During the test print, I immediately noticed that the stepper wasn't going backwards any more - So I came back to Cura and realised that "Travel" moves were turned off. You can see in the image below, small joining moves that extend horizontally. So, the motor is going backwards, but only because of the CoreXY geometry. A Cartesian printer wouldn't see the same effect.

Here's the same print, sliced with the "Zig Zag" pattern instead.

so after all that - did it help?
Yes - but only slightly.

The image below shows the Lines pattern on the left, and the Zig zag pattern on the right. I've used the lighting here to show that the lines pattern has slightly higher ridges, whereas the zig zag pattern is flatter. It's not a huge difference to the touch though, but its visually cleaner.
For what it's worth, this was done with a 0.8mm nozzle too, so the effect would be lesser on a 0.4mm nozzle.