UM2 temperature vary

[gluetolf 5]

Hi all

My UM2 shows strange behaviour. When I heat up to extrude, or when I print, the temperature varies +/-10°C around the temperature I set. Already tried different power sockets, checked wires, run UM2 through APC USV. What I didn't try yet is another power supply because I don't have one at hands. I also reset to factory defaults and tried several FW 13.12-14.07. Might I check something else?

Thank you

Gregor

 

[gr5 2,172]

It's normal for the printer to overshoot the target temp by 10C then undershoot by maybe 2C and then be stable. Is this what you mean?

Do you keep changing the target temp? What country do you live in (please update your profile location)?

 

[gluetolf 5]

No it varies over the whole print time by +/- 10C. I don't change the temp, it's the printer itself. I live in Switzerland. No I try to disassemble the nozzle, but I don't get it how to unscrew the M3x14mm slug without damaging it.

 

[gluetolf 5]

I got it solved by applying thermal grease Arctic Silver 5 to heater and temperature sensor. Some writing in German about it: http://www.3drucken.ch/2014/07/ultimaker-2-back-on-track.html

 

[anon4321 16]

Keep an eye on the AS5 and you might want to translate into German for the other thread. Most likely it will dry or lose its thermal conductivity - from http://www.arcticsilver.com/as5.htm

Extended Temperature Limits:

Peak: –50°C to >180°C

Long-Term: –50°C to 130°C

So the extended temp line is >180C which implies only for a short duration. The long-term temp is 130C so you will be well over that for even the coolest prints.

 

[titanoid 4]

Hi all!

Moved from http://umforum.ultimaker.com/index.php?/topic/8160-nozzle-temperature-range-while-printing-910-degree-celsius/, klausz and I had the same problem, he decided to carry on here to avoid double posting.

In my case, this temp issue caused periodic wavy walls, especially on small spiralized prints.

After applying the heat conductive paste, this happened:

 

Yes, my mouth is still open too... Temp varies now +/- 1°! (Before +/- 10°)

 

Since I received my UM2 three months ago, I printed quite some things, some came out better, some not. Learned a lot about all the beautiful parameters, prints getting better. (Learned the most around here, of course!)

At the moment I´m experimenting with the spiralize option and get this cute wash board surface.

Did a little research and found different things.

 

I must confess that I noticed the high amplitude of the nozzle temp before, thought that´s probably state of the art and I have to deal with it. Harrumph. Sorry for that.

 

Although I´m terribly happy with the improvement through the heat paste, some questions arose:

 

 

Since the thermal feedback between heater and sensor is obviously so much better with this paste, why do we find these little boors completely naked in the nozzle block?

 

That leads directly to the next question:

 

Is the software meant and able to deal with at least a little insonsistency?

I think yes, but some people discovered difficulties with the last firmware updates http://umforum.ultimaker.com/index.php?/topic/6613-firmware-surface-kwality-bug-1407-vs-1403/?hl=%20horizontal%20%20banding.

This is a little test print I made while following the software trail on my research:

 

 

 

Compared Cura 14.07 to 14.10 before applying heat paste.

(To add a little confusion, firmware on printer is 14.09, but Cura 14.09 had a little bug in spiralize, which Daid fixed. Thanks, Daid!)

So as you see, there is an improvement. The gcode provided from 14.10 seems a little smarter in handling the big temp amplitude. But if this amplitude is already minimized on the hardware side, the result would be even better, wouldn´t it?

Looking at bigger, more complicated models with infill and all that, a permanent temp fluctuation will be compensated.

If several peaks happen inside one single layer, the outside walls may look straight.

Or the walls show an irregular horizontal banding, and that, as we all know, leads us to a wild wide variety of possible reasons including belts and axes, vibration etc. Sure, all of this has to be checked.

But I wonder why this thermal feedback issue is not more common?

Or is the heat paste more common than I thought?

Would love to have some feedback on my thougts, if anything is nebulous, don´t hesitate to ask.

@klausz: I hope this works for you, too. Fingers still crossed, please report!

 

[mnis 11]

Hi

I use currently Cura v14.09 and have noticed the large temperature fluctuations, but it stabilizes within maybe the first 2 minutes, and keeps constant the predetermined temperature during the remaining time.

The thing with "Silver Grease" I think that's good, because even if the "Grease", the temperature can not stand, then remain 99.9 percent silver. :lol: ...this is better than nothing in any case, I think.

Maybe it also helps in addition something if you limit the maximum fan power in the Cura expert settings to 65 or 75 percent. I've been using for quite some time, at 4.5mm layer height, a maximum of 65 percent fan speed. This moderate setting is sufficient for good to very good results with my UM2. The left fan can cool down the nozzle greatly, especially when the nozzle tip protrudes very far from the sheet, this can be changed with the adjusting screw (insulator with holes).

What I can see from your last picture :eek: :roll: I actually know a lot more from very bad filament, a partially clogged nozzle and or irregular feeding by the material feeder. See through the Bowden, look directly onto the knurled wheel tracks in the filament. Can you identify any irregularities, such as small grinding marks?

Markus

 

[titanoid 4]

The thing with "Silver Grease" I think that's good, because even if the "Grease", the temperature can not stand, then remain 99.9 percent silver. :lol: ...this is better than nothing in any case, I think.

 

Better than nothing, yes, but such things tend to come back one day, boomerang-style. I don´t like that. So I took this one. Says to endure 300°C, that fits perfect! :-P

 

What I can see from your last picture :eek: :roll: I actually know a lot more from very bad filament, a partially clogged nozzle and or irregular feeding by the material feeder. See through the Bowden, look directly onto the knurled wheel tracks in the filament. Can you identify any irregularities, such as small grinding marks?

 

I can exclude this. Fortunately I discovered the benefits of atomic method and IRobertIs feeder pretty early, thanks to this forum!

As I realised this temp fluctuation (during the whole print) I reviewed all my prints so far. If you know where to look for, suddenly some patterns absolutely made sense. What you see in the last picture is the same regular pattern as in the little spiralized thing. Of course the "frequency" of the ripples depends on how long the printhead takes for a specific layer, so they are regular within a block of identical layers.

Nevertheless, thanks for the feedback!

Chris

 

[mnis 11]

A few more thoughts:

Perhaps the connection plug of the temperature sensor and or heating element is not seated correctly, or it is too loose. Also, the cable could be faulty somewhere.

There is also the possibility to temporarily install a much older version Cura including device firmware, for comparisons.

Also you could on the device menu: "Heatup nozzle" set the nozzle to a certain temperature. Then move the print head over wide ranges manually to check the temperature stability.

With an oven thermometer a rough comparison measurement could be performed.

And:

Are the objects shown by you somewhere downloadable, so there is the possibility to reproduce. The Cura settings used for the objects are also interesting.

Markus

 

[titanoid 4]

Whoops, terrible misunderstanding here:

My problem is already solved,

through applying heat conductive paste at heating element and temp sensor in the nozzle block.

 

The print on the right side represents the current state of my printer and, as you can imagine, I´m very very happy with this.

I just wanted to take a closer look at the symptoms of this issue visible in the prints.

As these symptoms (regular wavy outside walls) are very similar to e.g. z-wobble for the untrained eye.

And for noobs like me, before disassembling the z-screw, the issue in this thread should probably be checked first!

@mnis, sorry for inconvenience, next time I´ll try to make things clearer and shorter... :oops:

Chris

 

[mnis 11]

No, it's fine the way it is. Your contribution was very informative, it always takes a while until I fully understand something. :unsure: I should read more rested and attentive, hrmpf ... :oops: :ph34r:

:arrow: Your result is very convincing, and the next disassembly / assembly of my UM2-print-head I'll try it with some copper paste. :idea:

Thank you for sharing your experiences.

Markus

 

[klausz 1]

I have added Xilence X5 (according to manual) which reduced the temperature range to +/- 3 degree celsius but normally to +/-1 degree celsius (as Chris reported).

I also changed the nozzle (nozzle heater block) but still had problems.

Finally I opened again and changed the hot end isolator and the PTFE coupler. Nearly for sure it was the PTFE coupler. Now I have stable temperatures and it seems the difficulties are solved.

Using Taulman Bridge i was able to go up to the recommended 240 degree celsius (as there is nearly no risk that the temperature will go higher which causes problems) and the prints are at a quality level I never had before. :smile:

 

[mnis 11]

Oh, it's really amazing:

Cura v14.09 brought a negative change in the firmware, in relation to older firmware, the temperature stability of the Hotend (UM2) was no longer good. Perhaps it has always been similar unstable, at least to me, the thing is not so much previously noticed, and there are also some similar messages from other users.

A few days ago I made a maintenance on the print head of my UM2 and attached SilverGrease on heating sleeve and sensor-sleeve, something else was unfortunately not in reach.

But in fact the chosen Hotend temperature now remains almost constant even for small to medium speed changes and stabilizes generally much faster. During printing at a constant speed, the temperature fluctuations are according devices display now at +/- 1 degree, which was previously definitely much worse. Also on the printed objects may very well be seen with the naked eye a positive change. And as a little side effect for me it is now also possible to reliably work with PLA and temperatures well below 210 degrees.

Thanks @all

Markus

 

[titanoid 4]

Well, if someone stumbles upon this these days, a warning:

The thermal grease will dry and you will eventually hardly be able to release heater and sensor ever again!

I'm terribly sorry for that, as I recommended this so much.

Chris

 

[klausz 1]

Unfortunately I have to agree with Chris (titanoid).

When printing Taulman bridge it did not stick and got stuck everywhere because I was not at home to stop it. When I tried to remove the filament within the print head, I destroyed the temperature sensor (maybe I was not careful enough. Nevetheless I was not possible to take the temp sensor out and also I had to drill the remaining parts of the temp sensor out of the hole (iGo3D delivered the temp sensor very quick).

Finally I assembled everything. Now the machine is back working. (Maybe later I will add some comments on improvements, so that it is more easy to exchange the temp sensor.)

Coming back to the head paste. I liked i very much. The print quality had improved a lot. Not being able to remove the sensors anymore is unacceptable.

Looking for other solutions, there could be a calibration or a different copper paste (there is something with copper for brakes in the car - I have no idea if this should be considered).

Re temp sensor, in case you break it. Before you exchange put it back in the hole and try. Even when the temp sensor is still working, but the sensor is outside the hole, there is an error message. (Possibly for an activated heater without temp increase the program will conclude the sensor is broken).

The steel cylinder is not really necessary. Having time I will look for another heater or even better for a complete printhead (all the connectors should be near the print head for a quick exchange).

Also sorry from my side for recommending this 'glue-ing' heat paste. If there is no solution right now we should look for another solution.

 

[gr5 2,172]

You should be able to get a temp sensor from ultimaker for free - just tell them you had to break it to get it out when there was an issue with filament all over the head.

Alternatively you can get them here:

http://3dsolex.com/?page_id=579

but only if you also buy the Olsson block. The Olsson block has a hole so that you can PUSH the temp sensor and heater out! Nice design.

 

[Labern 774]

Just saw this today.

We used to use this paste at work but ended up stopping as it hardens at binds your sensors in place. Its really only good for in between seat sinks and electronic components not thermocouples and elements. Also if you do ever remove your thermocouple you have to remove all the excess paste from the hole before putting a new one. otherwise the old hardened paste will actually act as a slight insulator and give incorrect readings.