UM2: Nozzle not heating up anymore

[sebastian 0]

Hi everyone,

my printer seems to be broken. I already filed a support ticket but as it is very urgent, maybe someone here has a hotfix idea.

A few days ago, after a failed print, the nozzle didn't heat up anymore. I checked whether some cables went loose as we transported the printer before that for a workshop. Everything seemed to be fine so I just pushed gently at the temperature sensor and the heating cartridge cables at the hotend. That helped and the nozzle did heat up again. So it seems like a defective contact there.

After that, a few prints went perfect but then the problems started again. I monitored sudden ~50 degree temperature jumps during prints so I thought something was wrong with the temperature sensor.

I disassembled the hotend and carefully pushed at the cables again. The problem seems to be a loose contact at the heater cartridge (see picture), not at the temperature sensor. Again, a few prints went well but now I can't get the printer to heat up anymore.

 

Seems like I need a new heater cartridge? Anything I can do to fix this? Is it possible to disassemble the heater cartridge from the heater block?

 

Sebastian

 

[illuminarti 18]

In addition to the collar that joins the white teflon part to the heater block, there is a thin threaded rod that locates the block in place. I think you can access that by looking down through the top metal plate of the hotend assembly. If not, unscrew the collar, so that the block comes free from rest of the head. Then unscrew that long rod (it has a hex driver slot in the end of it). Once you loosen that, the heater and temp sensor should slide out of the block.

 

[gr5 2,170]

If you remove the nozzle you need to heat it up to 180C first (240C for ABS) so that the threads will turn. The collar that has holes on it - you can put a thin screwdriver into those holes and spin it - the nozzle will go up or down. You want the nozzle to go down until it falls off and then you can remove the threaded rod illuminarti speaks of. Or maybe you can get to the threaded rod from above. Either way you only need to loosen about 5 full turns before the heater and temp probe slip out.

For the temp to shift 50C "suddenly" it has to be the probe, right? It should be around 100 ohms when at room temp (more like 108 ohms). I would put a multimeter on the probe wires at the other end (disconnect probe from PCB and connect to multimeter instead) and then jiggle all those wires at the head to see if resistance changes suddenly).

Here is a table that relates temperature to resistance for the probe in the UM2:

http://www.intech.co.nz/products/temperature/typert.html

 

 

[sebastian 0]

Ok, so now I definitely need a new heater cartridge....

I tried to unscrew the long rod illuminarti described which was pretty difficult since the hex driver slot was already completely rounded. I never touched that screw before so that must have happened during assembly.

The temperature sensor slided out easily but the heater cartridge did not. So I tried to get it out with tongs which immediately resulted in a destroyed heater cartridge. It seems like one of the cores in the wire was already in a very bad condition (what probably caused the problems...) and the other one just snapped off. I still need to get the heater cartridge out of the heater block...

I did a quick test with the temperature sensor which seemed to respond okay.

Sebastian

 

[jonnybischof 60]

...

For the temp to shift 50C "suddenly" it has to be the probe, right? It should be around 100 ohms when at room temp (more like 108 ohms). I would put a multimeter on the probe wires at the other end (disconnect probe from PCB and connect to multimeter instead) and then jiggle all those wires at the head to see if resistance changes suddenly).

...

 

I've had a similar problem just recently. What you're describing sounds much like it..

In my case, the cable that connects the temperature probe to the electronics board broke (slack joint). The temperature reading just stopped because there was no more connection. The printer assumed that the temperature still was the same as the last read value - which resulted in uncontrolled heating up or cooling down of the nozzle, clogging and burning it up in the process.

You may want to test that cable, too. Just to be sure. This is an awful problem because it's hard to identify. The easiest way to test the cable is to join the two wires together on one point, and measuring the resistance between the two wires on the other end. Should always be close to zero, but if it changes significantly if you move the wire, then you've got a slack joint and need to replace the wire.

My advice: Use high-flex cabling, and use 0.25mm2 wires instead of something thinner. These are pretty hard to break...

But I'm afraid this won't be heeded because hflex is like 5 times more expensive than standard cabling...

 

[sebastian 0]

Received the parcel today and installed the new heater cartridge. Everything is working fine again and I'm a happy little printer now...

In case somebody needs to know, it is a 24V/25W heater cartridge in the Ultimaker².

 

[illuminarti 18]

Yay! I'm glad you're back in business! Sorry it took so long to get you the parts.

 

[sebastian 0]

Hi guys,

after installing the new heater cartridge, my printer seems to have problems adjusting temperatures.

Here is a temperature curve from OctoPrint:

Notice the high amplitudes. This leads to more uneven print surfaces.

For comparison, here is the chart for a new Ultimaker² which is located right beside the first one:

 

What could be causing this?

 

Sebastian

 

[Dim3nsioneer 557]

Suboptimal PID parameters... you may try to use M303, but please only if you feel safe with it...

 

[jonnybischof 60]

I'm investigating similar problems on my machine. It could be electrical interference on the temperature sensor wiring. I have that strange problem that my heated bed temperature keeps dropping after a few minutes, but when I touch the sensor wiring it normalizes, then after a minute or so starts dropping again.

Going to take some time until I can really say what's happening - I'll have to hook up my oscilloscopes to take some measurements.

Having electrical interferences can lead to bad readings which will toss the PID out of balance. Or, it could of course simply be a bad configuration as Dim3nsioneer says.

 

[sebastian 0]

I reconfigured PID parameters using Repetier Host. That helped! Thank you.

 

[gr5 2,170]

Those curves didn't look too bad but it is definitely PID issue. My UM2 overshoots once - then slowly recovers.

Can you publish the "old" and "new" PID settings? Someone else might want to try your values.

 

[sebastian 0]

Sure, here are the settings:

Factory:

P=12.15

I=1.0

D=37.02

New settings:

P= 12.40

I=0.89

D=43.28

Sebastian

 

[gr5 2,170]

Those are so similar that I would say there is no difference. In general, oscillations are made worse if you increase P or I or decrease D. But your changes are so small - I would have expected P or I to be cut in half or D to be doubled in order to dampen those oscillations.

Anyway, I'm glad you got it working.

 

[pm_dude 27]

Got a similar issue last night. Started having some bad prints with Colorfabb PLA/PHA. I wanted to test with a new nozzle and discovered the nozzle was pretty badly assembled.

- Loose nozzle

- Nozzle not fully screw in the hot-end insulator

- The M3 slug was so stuck I had to use vice-grips to remove it because the threads where completely damaged.

- The Heater was only partially inserted in the heater block

- Teflon insulator is still good

 

Doing this caused damage with the heater wire. Orange light started to appear when heating the nozzle.

 

Should I try to cut the wire where the burn occured and re-solder them?

 

[cor3ys 28]

Hi PM_dude

Yeh my 3mm slug came out like that and had to use some pliers to undo it, but i have pleanty of 3mm rod and lucky the nozzle thread is still ok, just waiting on a new heater cartridge arrive. (found a UK suplier of UM2 spares :smile: )

 

[pm_dude 27]

I have a new nozzle kit ready to be used so that's not the issue at the moment.

I really need to get that heater working again and since the issue seems to be in the wire I which I could just cut/solder it there to see if its just that or the entire heater that need to be replaced. :-|

 

[gr5 2,170]

How hot does that part of the wire get? I would tape the heater and temp sensor together with kapton tape (can handle something like 600C) and then turn it up to 250C and see if that brown spot is a problem.

If so then yes, you might as well cut it all open and solder a new wire in. But use lead solder (hard to find now) which can handle up to 300C. Typical electric circuit solder melts around 200C I believe. Or maybe 1 inch up the cable is below 200C?

 

[pm_dude 27]

I think I don't have access to this type of soldering material anywhere near. I'll try it and if it fry again I'll know :S.

Support was kind enough to send a new heater cartridge and nozzle just now. Thanks Simon

 

[pm_dude 27]

The solder material wont stick to the connector or to the wire.

An interesting fact is that the place the wire burned is not so random after all. This is where the connections between the wire and the heater are done.

If I can't solder that wire at this location then I wont be able to solder it when I get the replacement heater cartridge.

Any clue what soldering material I should look for more precisely? Mine is high-tech rosin core solder and contain 62% tin, 36% lead and 2% silver.

[jonnybischof 60]

As George already mentioned - soldering anything that close to the heater cartridge is not an option - Unless you find high-temperature soldering tin, which is not only difficult to find but also difficult to work with.

You could use crimping ferrules instead:

Twist the wires together, put a ferrule (thin metal "tube") over the spot and crimp it with pliers. This way you don't need to solder anything and the solution is high-temperature resistant.

The ferrules I linked above are probably too small (they're meant for the thermistor wires, not the heater), but you'll find them in any electronics components shop. I usually recommend Mouser electronics.

While you're shopping, you should also get heat-resistant insulators, for example these from Alpha-wire (check which diameter you need).

That's more or less what the heater wires are usually insulated with. I just did a quick search here, you might find something more suitable (cheaper?).

 

[pm_dude 27]

Thanks Jonny.

 

Here's what the connection look like.

 

Its really weird that the connection with the wire is made at that place. Its really a weak spot to do this in the wire itself without protecting the connection from bending. Even worst, we bend the wire at this precise place to fit it in the print head.

 

Crimping ferrules looks interesting. I'll try to find some locally to test but since the replacement heater is in the mail already I wont order anything online. I cant twist the wires together since one end is not the wire but the connector itself but it should still work.

 

 

[gr5 2,170]

It's hard to do this "in mid air" so I recommend you pull the wiring completely out. Sorry - just do it. It's worth it. You dont' have to feed the new heater wire back through as you can just wire tie it to the outside of the black net bundle.

If solder won't stick to a metal surface it is usually because of a thin layer of oxidation. But often if you scrape off the oxidation it re-oxidizes the layer below in milliseconds. So you have to scrape off the oxidatoin "under water" or under "molten solder". The trick is to put the wire/metal on a strong surface like stone or wood, cover the tip of the iron with a drop of solder and then scrape/scratch the surface back and forth with the iron. After a little while you should see the solder sticking to the wire/surface in small areas. Keep going you don't want it to touch in only .001 square millimeters. Keep going until it seems like the solder is really sticking in plenty of spots.

Of course the crimping ferrules is probably a better option.

 

[pm_dude 27]

Clearly with that solution working midair wont work

I'll have to take it out anyway.

 

[jonnybischof 60]

Ah, no I see it. Maybe the producer used standard solder which came loose and caused the failure? Not all heaters are made for temperatures above 200°C (even though the fiber glass insulation suggests it..)

You can still use crimping ferrules, no need to twist the wires. It's better, but not necessary to twist them. Just place both the wire and the cartridge pin inside the ferrule and crimp it hard