Posts made by Pumlux

@chownki
I had some similar error messages on my Duet 2 Wifi months ago.
All was working fine and suddenly some phases may be disconnected.

To make it short, in my case the connection of the stepper was weak.
I checked it a couple of times and I thought the connections is fine, but when looking in detail, the crimping on the cable was not perfect and I could pull the cable out of the contact shoe.

To be 100% save I soldered the cable to the connector shoe and this error message did not appear again.
So shame on my crimping style.

You may check with a complete new cable on your driver giving the error message.

I got the parameter B1 example from duet3d.dozuki. It might be a worth to change the B1 parameter to B0 in the toubleshooting doku for BL-Touch.

Modify your M558 and G31. The BLTouch behaviour is controlled by the M558 and G31 commands in config.g. Here is what an ideal set of commands should look like.  


M558 P9 F100 H5 R0.2 T6000 A5 B1 


This sets probe type 9 for BLTouch, 100mm/min probe speed, 5mm dive height, 0.2 second pause time before probe move, 100mm/s travel moves between probe points, probe up to 5 times , or until 2 consecutive probe results match, and disable heaters during a probe move to limit interference. G31 P25 X10 Y10 Z2 Sets the trigger sensitivity to 25, the X Y and Z values should be measured to match your setup.

@DC42
David you are fast and you are right.
Thanks for your helpfull advice.

I had the M558 with B1 parameter (got is from setup example for BL-Touch) and wasn't aware of the impact of the B1 parameter.
Keeping the heater on as default is the better choice. Not sure if the homing will be that much more accurate with heaters switched off. Would need to check this.

With heaters off during homing I run into undertemperture issue with no extruding because of low temperture when the printhead was located very high over the bed. This is solved now.

Bug or feature ?

Temperture is dopping during a long time consuming move / home all .

I'm running a Duet Wifi 1.01 board and ugraded some weeks ago to RRF3.
I noticed that the heater seems to be not heating if a time consuming move is made.

In detail :
I heat up the bed to 50 degrees and the extruder to 200 degrees and than I do homing all.
During the move to home all axes the heater seems to be not heating, the time for homing was aprox. 2 min 30 sec and the extruder temperature drops from 200 to ~140 degrees. (see screenshot below ) when all the moves for homing are completed the temperature is rasing again to targeted 200 degrees.
(I see this temperture drop on the long time z-axis movement, not sure about the x and y axis)

In addtition, If the duet is heating up to the target temperature e.g. from 20 degrees to 200 degrees, I can not do printhead moves until temperture is reached (not always, I don't know what makes the difference here) .

It looks like moves and heating can not be done in parallel, not sure if this behavior comes from the duet firmware or from the web firmware.

I just noticed it on RRF 3.01 RC4 and RC5.

Solved on the extrusion multiplier issue.
The problem was in front of the PC

I had mixed T0 with drive 2 some time in the past and I need to change the extrusion multiplier for Extrusion 2 .
When doing this, also the displayed extrusion length changes in the expected way.

Completly my fault.

Closed

Regards
Pumlux

Hi all,

I'm facing since some days a strange problem and can not find how to fix it.
I noticed that suddenly from one print to the next print my setup seems to be not correct any more. The walls are not touching together and the hole print
object it not strong enough. It is more visible when the print contains retraction. The gap is minimal and I tried to fix it by increasing extrusion multiplier but this seems to have minimal to no effect.

To test this I print 2 lines
0.8 x 100 x 0.4 mm
1.6 x 100 x 0.4 mm

I tried with extrusion multiplier of 20% , 100% and with 400 % and the drive display in the webgui says always 38.4 mm.
I messured the thickness of each line printed with a caliper and they had been all 0.3 mm.

For me it looks like I can not influence the extrusion multiplier at all.

For my setup, just some small details.

Selfbuild printer MendelMax like with Duet Wifi and Titan Extruder using about 60 cm of bowden between coldend and hotend (e3D Chimera) with filament cooling.
I printed with 83 °C bed and 185 °C Hotend.
Firmware 1.18.1

I did not retry to reload the firmware yet.

Did someone had a similar issue or a tip for me ?

Regards
Pumlux

Hi Strips,

did you check your wiring for the y-stepper ?

I had a similar issue , one with the also the y-stepper and one somewhere with the power supply setup.
After I crimpt the cable with new connectors, my issues disappeared completly.
I used already a good crimping tool, but it looked like I worked not good enough on the first time.

Just from my expierence.

Regards
Pumlux

Looks like the link in the first post ist broken.

I did not see a default gateway in the list. Wouldn't this be requirred for accessing from different network , aka port forwarding from public Internet / VPN ?

Regards
Andreas

Did anyone had success to access the Duet Wifi from the internet ?

Let me describe again what the goal to achieve is.

Access the Duet wifi (having a private IP address) via the public IP internet over a router.
Normaly this requires a port forwarding rule on the router, something like
forward (external IP port) (Duet wifi external) to (ip of Duet Wifi) port 80
e.g. take 2000 as external IP port and the duet wifi ip address as 192.160.0.100

This would allow you to access your Duet Wifi over the internet .
With the duet 0.8.5 this works perfectly, with the duet wifi the connection can not be established.

The explanation from W3DRK sounds resaonable, but makes me unhappy

@ tomasf
I think this depends on where your VPN tunnel terminates and what kind of IP address your VPN network card gets assigned. As long as this ip address is in the same subnett as the Duet Wifi is, the explanation from W3DRK is still valid for your and his statement.

Damn stuff with periodic errors, currently it is working as expected and I can not troubleshoot.

Keep you posted.

@Qdeathstar : I use no haispray or other adhesives for the first layer.

Hi all,

I got some new strange behavior from my IR-probe. Does anyone had something similar ?

My configuration :

David adapted my IR-probe for high reflecting materials (to be less sensitive).
I currently have a natural color FR4 plate on my print bed where I had roughened the surface with sandpaper.
I had also to adjust the angle of the IR LEDs mechanically, because roughened with sandpaper influenced the read out values.
This was working fine for many weeks.
Suddenly when the printer calibrates axis for a new print, the nozzle was going into the bed.

It shows the read out values from the probe starts vor greater high with 0 and jumps than when moving the sensor to the bed to a value ~ 396 - 450 (but defenitely below 500).
Moving the sensor nearer to the bed, at some distance the value increases a bit, but still below 500 . Moving further downwards, the value decreases and at some point it is 0 again.

I put a white piece of paper on the surface, but still the value is below 500.

I had the printer powered off over night and next day, all was working again fine. Until some days later I had the same issue, with even lower values.
In detail, I calibrated all axis manually and start than the print process, I heated at the beginning the hotend and bed up
and than the print start with calibration of all axis again. This time z-axis run into the bed again.

I did the z-probe readout with a piece of white paper again and it showed the z-probe value is even below 400 (as the highest read out value).
One run of the read out values as example :

probe Value
height readout
7.28 = 0
7.27 ~ 400
6.87 ~ 420 (LED turns on)
4.71 ~ 420
4.70 ~ 220 - 250
4.68 ~ 30 - 60 (LED turns off)
4.50 = 0

The key point I noticed, the value is always below 500 and this makes the calibration with a trigger value of 500 not working.

What can be done to avoid this situation ?

*update -1- *
I moved the x coordinate about 5 mm and currently I get again read out values > 500 and more stable. This may be a workaround, or just the probe works as should for some time again.

Regards
Pumlux

Hi,

on older firmwares on the duet 0.8.5 it was possilbe to edit some files on the SD card to have e.g. the language setting and some more GUI settings set
permanently. With the new web GUI everything is placed in a .bin file. ( Not sure if this was the modell.js or the language.xml file)

I miss the possiblility to change the default language and some GUI behavior. I start always with Englisch and ATX power controll missing.

Is there still a way to say the GUI the default settings ?
Anyone who can provide a hint to me ?

Are you sure your mount for the z-probe is thight enough to reproduce the height exactly ?

When I used a printed mount for my z-probe with a clamp in fixing, my height also was changing.
After using a metal mount system with having the z-probe screwed to it, the height is realy reproducable .
You already mention you have variances in about only some 1/100 mm.
The remaining factors are heat elongoation and the bad results when I hit the bed with the nozzle for some reason.

One short update here, may be a info that could go into the wiki also.

I use the switched +5 V power ot the ATX PSU to power the Duet and for switching on the duet I need to
keep the PS_ON lead from the ATX PSU grounded manually for ~3 seconds, after the blue LED near the wifi module
flashes a bit brighter and goes off.
Than the duet PS_ON pin goes to GND and I can release my mechanical push button.

Note : I decided to go this way, because I do not want all the electronics to consume power without any purpose.
I have (currently) no plan/need to switch the printer on remotely without being there to watch the print start.

Hi eddygara,

did you solve this issue already ?

I had been fighting for z-axis calibration for a long time and can confirm this is very important for the printing.
The first layer must be best on the hole print.

Some of the outcome that I find during my calibration cycles so far.

1. You need to be sure that the probe mount is good enough to have a reoccuring precission.
   I had printed my first z-Probe mount myself and that was always changing the hight of the IR_sensor for some 1/100 to 1/10 mm resulting in the printnozzle being much to high. I change to a metal mount and this has solved this chaning z-Probe height issues

2. You need adjust the z height parameter in the G31 command in the config and after saving the config.g you need to power cycle the duet (if not already suggested by the firmware anyway) Please be sure, when using a sheet of paper the z value of the nozzle is not 0 but t he thickness of the paper ! You may see that the printed surface is still not close, than reduce the nozzle height in som 1/100 mm steps.
   Note I : The height of the nozzle is only set for this one point you adjusted, this is sufficient if your print bed is 100% flat and horizontal. If this is not the case, you need to calibrate the bed or use autocalibration ( that is another story)

Note II : If the nozzle is too high, you need to increase the value for the Z parameter of G31, telling the firmware at the triggered height position the nozzle is a bit higher over the bed. Than the firmware knows the bed is a bit deeper under the nozzle.

Normally if you have set the G31 Z height correctly ,the nozzle height should be always set correctly and reproducable.
You may want to heat up the nozzle and the bed to the temperature you use normaly, as there might be some heat elongation, depending on the materials you have used.

I just tried it.
I have put the command M80 into the first line of the config.g file.

This works as expected, when pressing the push button for > 3 seconds. Looks like the Duet Wifi needs some seconds for booting
up and reading the config.g file.
May be some change of the hardware design is responsible for the different default behavior of the board during powering on.

Anyhow it works now as expected, many thanks for your fast support David.
update
One thing I need to mention. If the M80 command is inserted into the config.g file and +5V VSB ( standby) power is conencted, you can not switch off the board by
doing a reset, as after reset the config.g is read and the board powered complete on again. So this is only a solution in a case like mine, powering on via push button.

Regards
Andreas

Hi eddygara,

did you solve your issue here ?
You need to look in the config.g file for the M558 command David is referring to.

I just checked on my printer and the probe is LED is blinking two time when set to P0
You should use
M558 P1
to have it correctly working with all features.

One think I noticed on my printer, if you have +5 V VSB (standby power) connected to your Duet Wifi, you need to power cylce the PSU to see the probe blinking.

Regards
Pumlux

Hi David,

thanks for the advice.

PS_ON signal seems to be working, but a bit different than from the 0.8.5.

I connected the PS_ON , +5 V BSB and GND to the external 5 V connector and the Duet Wifi is powered.
I can use the M80 to power the PSU and M81 to power off, I can do the same with the ATX buttons from the Duet web frontend.
Power ATX on and off.

But I can not controll the PS_ON signal with just a short triggering of the PS_ON to GND like on the 0.8.5.
Looks like the default behavior on the Duet Wifi is PS_ON off and on the 0.8.5 was PS_ON on.

Is there a different way (pin) that I can use to have a normal power on hardware button ?

Is anyone using the PS_ON functionality with the Duet Wifi and can provide some details how he has wired it ?

I have currently no clue what might be wrong here.