Print quality difference between bed ON and bed OFF

  • Hello, I would appreciate it if you can help me solve the following problem

    I'm having this error:

    Printing normally, with the bed set in default mode (bang bang), the print is not clean, every 1 or 2mm or so, can see a difference between layers that looks like a Z error, but there is no error in Z, and in general mechanically the printer works beautifully. The issue comes from another side.

    If in the middle of the print, I turn off the bed, the layers begin to be absolutely uniform and perfect, as you can see the photo.

    My configuration is as follows:

    Cartesian printer
    Duet Wifi
    Mk2 Bed (1.1 ohm connected on its 12v configuration) with glass
    PSU 12v 30a (clasic meanwell clone) regulated to the maximum voltage (delivery 15 volts)

    I tested by regulating the PSU at 12 volts (standard voltage).
    I tried setting the bed in PID mode at 100% 80% and 50%.
    I tried printing with bed at 115º 110º 100º 90º…. etc, and different materials.
    ... and the same thing happens, maybe a veeeery little better in some case, or puntually the only that varies is maybe that instead of seeing the difference between layers every 1mm, it occurs every 2mm or every 3mm or every 0.5mm.... that distance varies between different combinations of PSU voltage, and PID % power, and bed temp. but the difference between bed on and bed off is still there, and is big.

    This issue did not happen to me with Arduino / Ramps (same printer)

    Any ideas ?

    Thank you very much!

  • Only a simple guess:

    Try to monitor VIN in the WebInterface and how far it will drop.

    And do a M122 in the print to print out the Statistics.

    Take a closer look to;

    Supply voltage: min 10.8, current 12.0, max 12.1, under voltage events: 0, over voltage events: 0


    MaxReps: 4, StepErrors: 0, FreeDm: 218, MinFreeDm 146, MaxWait: 366839ms, Underruns: 9, 0


  • administrators

    We've seen this before when the bed is running in bang-bang mode. Switching to PID has always solved it. If the PID is unstable, for example because you haven't run bed heater tuning, then the PID control could oscillate and produce the same symptom.

    The cause is that the power voltage drops a little when the bed heater turns on, which reduces the hot end power. This is exacerbated if the wires between the PSU and the VIN terminal block are too thin.

    In the next firmware version, on the Duet WiFi and Duet Ethernet the heater power will be compensated for changes in the supply voltage.

  • Thank you both for your answers.

    I will check voltage drop.

    I have autotuned bed at 15v PSU and 100% PWM, and M307 values come from there, but I will run it again at 12v and test at 12v.

    Wires between PSU en Duet are 3.5mm and between Duet an bed are 4mm… but also will check wires and adjust blocks too.

    I will report soon....

    In the mean time.... a SRR may be help in this case? or will be the same ?


  • It may help if the PSU is not so culprint here. If this will drop down the same will be vaild with a SSR. The Wires should be thick enough.

  • administrators

    You should run bed heater tuning at whatever PSU voltage you normally use. You can see whether the bed is operating in stable PID mode by watching the bed heater LED on the Duet, or on the bed if it has one. If it is running in PID mode then it should flicker but not flash on and off.

  • Did the adjustements.

    -Retightened terminal blocks.

    -Two new PID configurations tested:

    1. PSU at 12v: Autotune runed for bed at 100% PWM, and default hotend PID settings
    2. PSU at 15v: Autotune runed for bed at 90% PWM, and for hotend at 80% PWM

    After print with test #1 settings:

    Supply voltage: min 11.7, current 12.2, max 12.4, under voltage events: 0, over voltage events: 0
    MaxReps: 4, StepErrors: 0, MaxWait: 484810ms, Underruns: 81, 0

    After print with test #2 settings:

    Supply voltage: min 14.0, current 15.0, max 15.4, under voltage events: 0, over voltage events: 0
    MaxReps: 5, StepErrors: 0, MaxWait: 155000ms, Underruns: 0, 0

    Printed part is OK on both tests.

    Two Questions:

    1. On test #2 voltage min was 14v over 15v… is it an acceptable drop? or could be better and more stable with a better quality PSU?

    2. Cant find VIN monitor on web interface. I measured voltage externally. Can I add it to Web interface ? or maybe VIN monitor was added to a latest release?

  • Hi,

    the Voltage-Drops seems to be fine and i high quality PSU will not change it.

    The "bigger" 1V drop of the Second-Test belongs to the Bed-Heater. Even with a 12V and 24V original Meanwell - PSU (both around 650Watt) i've noticed drops up to 1.2Volts. This has changed to max 0.3V drop after using SSR's for the HeaterBed

    Additional you should find VIN in WebIf if you look up right

    But if you meassure the Drop to on the PSU directly i assume this clone of Meanwell is more bad designed

  • Thank you!

    Dont have a VIN in web.
    Web Interface Version: 1.15a

    I will update it anyway.

    I measured votage with multimeter from Duet VIN terminal blocks.

  • Puh, it will be a great idea to update…

  • I know, I know…jajaja... its the default coming with duet, only a couple of month ago buyed, I must to upgrade all the combo.

    Thanks David, Thanks MoS.

    Problem solved. Great result.


Looks like your connection to Duet3D was lost, please wait while we try to reconnect.