Wat. Happened.

[kittka]

Hey guys, first time posting on the forum.

Just recently completed my Rostock MAX assembly, and had two good prints, calibration cube and a left jowl of a halo helmet. Had a few issues with bed sticking using the glass with hairspray. I'm used to Kapton tape on my Solidoodle, but was able to finally get good adhesion following recommendations on this forum.

I went to print another helmet jowl, and it was a four hour print, and I left for a bit. I came back after the four hours to find a half finished print, and a broken hot end with unextruded filament coiled on the floor. I was printing ABS at 195°C. I print on the solidoodle at slightly lower temp (190) and I've seen higher temps mentioned by others for this extruder type but the printing/laminating seems okay.

[img]http://i.imgur.com/IetFNkA.jpg?1[/img]

So what happened? As far as I can tell I the PEEK tube separated from the extruder, but I can’t tell why. It looks like there is burnt ABS wrapped around threading on the PEEK. I came to work after shutting it down and had no time yet to diagnose much.

Can this be cleaned and reassembled?

I have not experience with cleaning PEEK material, can it be soaked in Acetone without issue?

Thoughts on the cause of this break?

I'm using the stock Rostock MAX kit, with overhead bowden, heated onyx bed, RAMBo 1.5? newest version, not sure what else to mention on this.

[img]http://i.imgur.com/9ClbsZA.jpg?2[/img]
[img]http://i.imgur.com/DR8n0PA.jpg?1[/img]

[Eaglezsoar]

It looks as though the hotend overheated and melted the peek section. You can get the peek section at SeemeCNC.com. It is a thread in type connection.
So what caused the peek to melt? This happens when the hotend exceeds 240C for a while but the question remains what caused the temperature to climb
to those levels? I suspect that the problem is with the thermistor. While the printer was running the wiring to the thermistor either opened or shorted. That
is the usual cause of such failures. I would look carefully at the thermistor wiring. I don't think you can repair that damage without a new peek section, get
a new one.

[kittka]

Ha, I could see that being the case. I'd had some issues with the first thermistor, but it broke so it read as high resistance. I bought some thermistors from ebay, and those so far have been somewhat variable and not as steady on the temperature reported, compared to the original kit thermistor. I'll bet that the thermistor crapped out, or when off the rails, and led to the overheating.

Now, I guess I'll need to clean out the nozzle? I can't imagine that it didn't clog it with the overheating.

Thanks for the input!

[Eaglezsoar]

Soak the nozzle in acetone, the hardest part is going to be getting the peek out of the aluminum. It screws in on both ends, once you soak it awhile you might be able to unscrew it,
that part you'll have to figure out. I don't suspect the thermistor itself, but rather the wiring going to it. Also make sure that it cannot get out of the hotend. Best luck.

[Lochemage]

I'm pretty sure if the thermister disconnected during the print it registers as 0, the firmware will catch that because the thermister value goes below the low threshold (meaning it registers a temperature far lower than room temperature, which is roughly around 25 depending on your room temp) and safely turn the heat and extruder off.

It might be possible that you have the wrong thermister type assigned to it in your firmware; but from what I've heard, they aren't that dramatically different from each other. Reaching a temperature that high just because the type was incorrect is probably unlikely.

The most likely cause is your thermister stayed wired properly, but the actual glass bead sensor partly fell out and is no longer measuring the full temperature of the hotend. Measuring a lower temperature, the firmware would increase the heat to the hotend. Eventually it will get hot enough that the thermister reads the goal temp, but the hotend is actually much hotter.

I would double check and make sure your thermister is still very well secured in the little thermister hole underneath that high temp permatex. As an added safety measure, I would wrap kapton around the nozzle to hold it in place better (It has been my experience that the permatex is a poor 'glue', my thermister just easily wiggles and pops out along with all of the permatex).

[kittka]

Quick Update, especially for any future people encountering similar issues.

Does anyone know the thread size for the PEEK? I damaged the aluminum female threads getting the remaining PEEK out; and I need to chase the threads with a proper size tap. I am guessing M9x1? or M10x1?

The PEEK replacement part has been ordered. Should arrive soon, and I can repair the hot end. Looks like the RTV overheated as it became crunchy and fell off in some areas. I wonder how hot this thing actually got!?! Even saw oxidation/discolorization on the the thick part of the aluminum near the nozzle.

Looks like the thermistor is okay and not ruined, but it looks like the JST connectors I soldered weren't very robust, and this could have been part of the problem. Thanks to Eaglezsoar for this suggestion. At least I hope it was part of the unsteadiness I saw, if not part of the overheating problem. Once I fixed those, the temps looks quite steady and correct.

As Lochemage suspected, the thermistor was separated from the hotend head, but I couldn't determine if this happened before or during PEEK breaking. I must say I'm not thrilled with the attachment method in the assy instructions, even the assy document talks about the difficulties here. I agree some Kapton or something could help, I'll do that, thanks Lochemage. I replaced the thermistor with what I believe is a proper 100K NTC thermistor with +/-1% which I figured was appropriate.

[Eaglezsoar]

Send an email to [email protected] they are the only ones that are going to know what those thread sizes are.
Best of luck on all of this. Been there, done that.

[GarageBay9]

I had this exact same thing happen a few months ago.

Did you use high-temp silicone to install the thermistor, or did you get impatient like me and just tape it in the hole with kapton because you'd had to mess with it like four times and were sick of busting thermistors that were each $10 + shipping + days of waiting when you wanted to print right now, dangit?

If it doesn't have a solid bond with the lower body of the hotend for heat transfer, and there's an air gap or the tip of the thermistor can be pulled out of contact with the walls of its hole when the wires get moved during printing, it can give inaccurately low temperatures - but not low enough to trigger the automatic cutout. The controller gets a reading from the air in the cavity for the thermistor instead of the actual hotend temp from the metal, and starts trying to heat it up again.

Mine gave me a brief reading of around 290c when it failed with a pretty loud snap. Here's the thread I started, before I did a very careful teardown and some post-mortem testing on that hotend to track down the exact cause. http://forum.seemecnc.com/viewtopic.php ... 862#p18862

[kittka]

GarageBay9,

I did use the RTV, but each time I've had to fool with this I do get impatient. But i found some cheaper NTC thermistors on Ebay, and am using that with some success.

But with regards to the thermistor bond, the instructions are very clear to NOT coat the entire glass bead, and somehow get the head touching the bottom of the hole. I just insert the thermistor, bottom it out, then RTV it kinda pushing it into the gaps. Don't know how much air is left in there, but I'm sure there's some.

Like my Solidoodle, I've resigned to keep two nozzles to make one that works, and hopefully maintenance the 'down' hotend before the next issue pops up. If I ever get confidence in the nozzle, I might eventually attempt a dual extrusion system.