Buildplate heating is not continuous

[jsieler 0]

Hi everyone, I am new here and my name is Joachim.
I have a recently occured problem with smaller prints that print well for about 20 to 30 minutes and then suddenly loose adhesion on the buildplate and start moving with the print head. I have monitored and logged the temperature of the build plate and found that it is not heating continuously but ranging between the desired temp, which is 61°C and about 45°C. This happens in cycles of about 20 minutes, meaning it heats up to 61 then over the course of approx. 10 minutes decreases to 45°C and then over the course of another approx. 10 minutes it heats up again to 61°C. This happens repeatedly for the entire time of the print process, which took about 20 hrs. Is this behaviour normal or could it be the cause of my adhesion problems? Ohh, and it's an Ultimaker 2+.

Thank you,

Joachim

Edited October 22, 2021 by jsieler

[GregValiant 1,351]

@Smithy and @gr5 are good with the Ultimakers and @Torgeir is a 2+ guy.  Load that model (if you can share it - or another similar model if you can't) and set Cura up for your slice.  Then use the "File | Save Project" command to save your work as a 3mf file.

 

Post both the 3mf file and the problem Gcode file here.  Somebody will review them and come up with some suggestions.

Edited October 22, 2021 by GregValiant

[Smithy 1,146]

Hi Joachim

 

no it is not normal. But question, do you have the stock firmware or tinkergnome installed on your UM2+

[jsieler 0]

Hi everyone, I have stock firmware and I upgraded two years ago from UM2 to UM2+. Everything worked fine for several months until suddenly prints that had been working don't anymore. The print works wonderful for about 20 minutes and then suddenly the adhesion seems to fail.

I'll do as you, GregValiant  suggested. I'l post the codes later on today.

 

thank you very much for your attention!

 

Joachim

[jsieler 0]

Here is a project file and the gcode file. The print works well for the first 3-4 cm but then the printed part starts moving with the print head.

UM2_cfg_tube_insert.3mf UM2_cfg_tube_insert.gcode

[jsieler 0]

Here is the printout of a part of the temperature log of the buildplate

[gr5 2,230]

It's not cura and it's not the gcode file. I searched the whole file and there are no temperature related commands of any sort in there.  Just fan speed.

 

You set these on the printer itself.  There is no option on the UM2 to cycle the temp.  Normally the bed temp doesn't move by more than 1C from goal.

 

So something is very strange here.  Maybe a bad power supply.  Does the nozzle temp also oscillate?  This pattern is strangely repetitve.  Is it possible it repeats once per layer?

 

Some versions of the UM2 firmware reduce the power to the printed bed if the nozzle needs much more power than normal.  I doubt this is the issue but it is a small possibility.  So the nozzle would demand more electricity in the same repetitive pattern.

 

Do you have a volt meter and know how to use it?  I'd monitor the voltage and make sure it stays at 24V during these dips in temperature.  It seems very unlikely that this would happen as well because the power supply is pretty smart and knows if it is failing and will turn off automatically.

 

Most likely there is a bad wiring connection.  Most likely it's where the connector is soldered to the heated bed.  That connector can come loose.  I'm not sure why it cycles like this but maybe when it heats up the connection expands enough that there is an open circuit and then when it cools down enough it reconnects.

 

Do you have soldering skills?  Do you know someone who does?  This is a 5 minute soldering job.  The bed comes apart quite easily.  Keep track of where the springs and washers went.  Be gentle with the cabling.  Reflow the solder for all 4 connections.

 

 

[jsieler 0]

Hi gr5,

thank you very much for your detailed reply and information! I also found no temperature values in the gcode because it is the UM2 gcode and AFAIK for the temperature settings the presets in the printer are used.

I have updated the firmware now and try a new print and monitor the build plate temperature every 30 seconds. If this didn't help, I will follow your advice and check the connectors of the build plate. I actually do not believe that the temperature in the nozzle also cycles, because I would expect that to result in patterns on the print part, which it does not. I have soldering skills and I can use a multimeter.

I'll report tomorrow, if the firmware update brought any changes. Thanks again!

Joachim

[gr5 2,230]

There are no known firmware bugs that would do what you see.  But it can't hurt to update the firmware.  There are tens of thousands of working UM2 printers and I've never seen this exacty cycling (although most people don't monitor it so carefully I suppose).

 

Hopefully it just needs a little solder.