Clearing Filament Jam Without Opening MOD-t (UPDATED)

[ajf]

So your filament is jammed. You can't print or even unload the filament.  There are a couple of threads in this forum describing how to disassemble the head and clear the jam.  Theoretically that method would void your warranty, though frankly, NewMatter would be nuts to try and enforce it at this point.

There is another way to clear the jam which doesn't require disassembly. But first, a little bit about what is happening:

The nut where the hotend attaches to the head (you know, the one mentioned in the cutesy video about changing the hotend) also serves as a transition and guide to get the filament from the extruder to the hotend. Well, due to what I'm going to stride out on a solid steel platform and call a fundamental, flabbergasting design flaw, that nut is getting warm enough for PLA to soften. And when more filament is pushed in behind it by the extruder, the softened PLA will expand and get stuck.  The servo driven extruder is not nearly powerful enough to push or pull the filament once it gets jammed in the nut.  Trying to run the load/unload sequences or even my clearnozzle.gcode will do nothing but cause the extruder's hobbed bolt/drive gear to grind filament, which will only make matters worse. So, DON'T!

To clear this type of jam, your'e going to need a soldering iron or some other way to apply heat directly to the nut. If you don't have a soldering iron, a large flat head screw driver heated by your gas or electric stove might work too... But, really, you bought a 3D Printer (crappy as it may be), you're a maker now..., get yourself a decent soldering iron - it will come in handy! (I have several, but this Aoyue 937 is quite nice for the price. A cheap 40W unit will work too.) We only need to heat the nut to about 160-170ºC to soften up most PLA. 

The idea is to apply enough heat to the nut to soften the PLA that's jammed so you can pull it out.  Here are the steps:

• First, detach the filament feeder tube from the MOD-t print head.
  
• Then, pull out  five or six inches of filament from your spool, so it protrudes from the top of the print head such that you have enough to grab and pull by hand.
  
• Now, cut off the filament and detach the spool and holder from the MOD-t.
  
• Disconnect power/USB and lay MOD-t on it's back.
  
• Apply heat to the retaining nut with your soldering iron or alternate heat source. I set my digital soldering iron to 350ºC to make short work of the job of transferring enough heat.  
  
• Be careful here as the main body of the print head is plastic and you don't want to touch it with your heat source.
• Nor do you want to overheat the nut, as that could damage the internal interface to the extruder housing which is also plastic.
• Eventually, you should be able to pull the filament out by the slack you left above the MOD-t head. It may help to have a second pair of hands, one set to pull while supporting the head and another to apply the heat source.
  

Note: You can try to avoid this type of jam from happening by always unloading filament immediately after a print.

More about the cause:

Looked in to trying to mitigate this with slicer settings. This isn't solvable with gcode because the jam is happening after gcode has finished running.  The auto cooling done by the slicer is kind of irrelevant here, but it doesn't hurt to leave the fan on all the time with PLA and that's why I've changed the Slic3r config to do so. (You can just go to Filament Settings>Cooling and enable 'Keep fan always on' in Slic3r without having to reload the config.)

The problem here is that the nut is aluminum and when the print stops the still warm hotend sinks all it's heat in to the nut. This could be mitigated by having the firmware keep the fan on after a print until the hotend has cooled. That assumes that the user doesn't turn off the MOD-t after use though. 

Basically, this is a horrendous design flaw. Fatal really. It will eventually effect every user, even with a firmware fix.

It's possible that adding a heatsink to the stainless steel tube of the hotend, on the threads directly above the heater block, could be a fix. (I'll give that a try and report back.)

Update (12.16.15 9:39 pst): The threads on the hotend's stainless steel tube are M5x0.8.  I couldn't find any aluminum in my (ridiculously disorganized) shop that was tapped for M5, nor could I locate my M5 tap... (Shut up you! That's not funny...) I did however have some M5 stainless steel t-slot nuts which I use for building my larger linear delta printers. Stainless doesn't make for a great heatsink, while I could improve the performance of the t-slot nut for this use by adding some thermal paste to the threads, think I'll wait for the new tap and die set to get here tomorrow and build an aluminum heatsink. However, the t-slot nut is helping a bit, I haven't had a complete jam since adding it (and, I've been trying!) The worst that's happened is that I've had to tug on the filament during the unload routine.

Update (12.17.15 6:00 PM PST): I attached a thermistor to the flange of the aluminum retaining nut that accepts the hotend and monitored the temperature during and after a print with the stainless steel t-slot nut acting as a heatsink.  During the print, the temperature held steady at about 31-32ºC.  Once the print ended, the head moved up and the fan shut off, the temperature began to quickly moved up to 82ºC as the hotend dumped it's heat into the aluminum nut.

I have now attached an aluminum heatsink (as seen below) and run another print.  This time the temperature of the aluminum retaining nut stayed at about 24ºC, just above the ambient temperature, during the print.  The post print temperature only rose to about 52ºC, which, even accounting for the fact that it would be somewhat higher internally, would not be enough to cause the filament to jam in the nut. Update: On further testing, with different parts, the temperature post print does go to 77ºC.  Will need to try a larger aluminum heatsink.

Update: (12.18.15 7:35 PM PST): Tried a larger aluminum piece as a heatsink, just a flat piece bar stock (pictured below.)  It managed to keep the post print temperature at the retaining nut to a maximum of 70ºC. Yet, even that still requires manually pulling the filament out to unload.  The problem is that the bar stock doesn't have enough surface area to compete with the retaining nut and the aluminum mounting plate to which it is attached in terms of drawing heat.  Unfortunately, the only proper heatsinks I have on hand are too tall to fit between the hotend and print head.  Will have to make a trip to WeirdStuff this weekend and find something suitable...(Update: Couldn't find suitable heatsink at WS, will have to hack something together.)

Update (12.20.15 3:59 AM PST): I managed to fabricate a larger heatsink (48g of aluminum) which was sufficient to resolve the issue. However, it's an impractical solution for the average MOD-t owner.  Reluctantly, I have decided to update the Slic3r config to force the fan to full power at the end of a print and wait for the hotend to cool down to 28ºC before actually reaching the end of the gcode file.  This is a clunky solution and obviously doesn't help a user who is printing from the cloud service or using a different slicer (though the same fix could be added.) As mentioned above, a firmware update that keeps the fan on until the hotend has cooled would also mitigate this issue. 

The real solution would be a redesigned print head which incorporates dedicated active cooling for the hotend feed tube (heatsink w/ always on fan) and new mounting nut that provides a thermal break.

[Atrix]

Is anyone able to check this as my printer is currently down, when using new matters printer software does the fan stay on all the time? 

(Maybe) If it does then that might be stopping the blocking behind the nut as using the Slic3r or Cura they both have an auto off after a certain temperature has been reached and that might be letting to much heat back up the metal feed tube cause the block.

Just to help this method you can unscrew the plate the hot end is attached too. So before you start this process unscrew all 8 screws and use the load filament feature this will push the plate down and showing the block. Once enough filament has come through, pull the power. Then cut the remaining filament (leaving a small amount to grab) left in the print head and using the load filament again allow to come out. Then start this method as it could cool on the way back up the filament feeder and cause an internal jam

[ajf]

I was just looking in to if this issue could be mitigated with gcode.  It really can't and I'l get to why below, but I have just updated the Slic3r config to keep the fan always on.  

I would not recommend unscrewing the mounting plate (at least for the casual user.) For one thing you don't need to, plus it's time consuming and a pain, there is also a chance of breaking the wire that's attached.  Also, pulling the filament out from the the top of the extruder may help to clean the drive gear a bit.

This isn't solvable with gcode because the jam is happening after gcode has finished running.  The auto cooling done by the slicer is kind of irrelevant here, but it doesn't hurt to leave the fan on all the time with PLA and that's why I made the change to the Slic3r config.

The problem here is that the nut is aluminum and, when the print stops, the still warm hotend sinks all it's heat in to the nut. This could be mitigated by having the firmware keep the fan on after a print until the hotend has cooled. That assumes that the user doesn't turn off the MOD-t after use though. 

Basically, this is a horrendous design flaw. Fatal really. It will eventually effect every user, even with a firmware fix.

[Jackson He]

ajf:

What if you add gcode to the end-code to essentially do an unload, such as retract the filament by a certain amount that does not get the filament completely out, but out of the nut under discussion? I don't know if this will screw up the calibration routine that runs at the beginning of the next job, though, because it may expect the filament to be in the hotend.

Best regards,

JH

[ajf]

Think running the unload routine is the better solution at this point.

With gcode not only would there be the issue of having to re-prime the hotend, but also the filament would deformed and there's no guarantee it wouldn't jam again. 

[Aaron Mahler]

If you want a quicker way - use a small blowtorch (the small butane kind used for soldering, hot air shrink tube heating, etc) right on the shaft where it enters the nut. With the printer off, apply the blue flame to it for just a few moments while pulling upward on the filament. Blam - released. 

YMMV. I'm sure this idea might make New Matter apoplectic... but I'm not seeing any problems after doing so and it keeps me from having to take the whole thing apart. 

 - Aaron

On Tuesday, December 15, 2015 at 9:16:01 PM UTC-5, ajf wrote:

So your filament is jammed. You can't print or even unload the filament.  There are a couple of threads in this forum describing how to disassemble the head and clear the jam.  Theoretically that method would void your warranty, though frankly, NewMatter would be nuts to try and enforce it at this point.

There is another way to clear the jam which doesn't require disassembly. But first, a little bit about what is happening:

The nut where the hotend attaches to the head (you know, the one mentioned in the cutesy video about changing the hotend) also serves as a transition and guide to get the filament from the extruder to the hotend. Well, due to what I'm going to stride out on a solid steel platform and call a fundamental, flabbergasting design flaw, that nut is getting warm enough for PLA to soften. And when more filament is pushed in behind it by the extruder, the softened PLA will expand and get stuck.  The servo driven extruder is not nearly powerful enough to push or pull the filament once it gets jammed in the nut.  Trying to run the load/unload sequences or even my clearnozzle.gcode will do nothing but cause the extruder's hobbed bolt/drive gear to grind filament, which will only make matters worse. So, DON'T!

To clear this type of jam, your'e going to need a soldering iron or some other way to apply heat directly to the nut. If you don't have a soldering iron, a large flat head screw driver heated by your gas or electric stove might work too... But, really, you bought a 3D Printer (crappy as it may be), you're a maker now..., get yourself a decent soldering iron - it will come in handy! (I have several, but this Aoyue 937 is quite nice for the price. A cheap 40W unit will work too.) We only need to heat the nut to about 160-170ºC to soften up most PLA. 

The idea is to apply enough heat to the nut to soften the PLA that's jammed so you can pull it out.  Here are the steps:

• First, detach the filament feeder tube from the MOD-t print head.
  
• Then, pull out  five or six inches of filament from your spool, so it protrudes from the top of the print head such that you have enough to grab and pull by hand.
  
• Now, cut off the filament and detach the spool and holder from the MOD-t.
  
• Disconnect power/USB and lay MOD-t on it's back.
  
• Apply heat to the retaining nut with your soldering iron or alternate heat source. I set my digital soldering iron to 350ºC to make short work of the job of transferring enough heat.  
  
• Be careful here as the main body of the print head is plastic and you don't want to touch it with your heat source.
• Nor do you want to overheat the nut, as that could damage the internal interface to the extruder housing which is also plastic.
• Eventually, you should be able to pull the filament out by the slack you left above the MOD-t head. It may help to have a second pair of hands, one set to pull while supporting the head and another to apply the heat source.
  

Note: You can try to avoid this type of jam from happening by always unloading filament immediately after a print.

More about the cause:

Looked in to trying to mitigate this with slicer settings. This isn't solvable with gcode because the jam is happening after gcode has finished running.  The auto cooling done by the slicer is kind of irrelevant here, but it doesn't hurt to leave the fan on all the time with PLA and that's why I've changed the Slic3r config to do so. (You can just go to Filament Settings>Cooling and enable 'Keep fan always on' in Slic3r without having to reload the config.)

The problem here is that the nut is aluminum and, when the print stops, the still warm hotend sinks all it's heat in to the nut. This could be mitigated by having the firmware keep the fan on after a print until the hotend has cooled. That assumes that the user doesn't turn off the MOD-t after use though. 

Basically, this is a horrendous design flaw. Fatal really. It will eventually effect every user, even with a firmware fix.

[Peter Traugot]

AJ, You mention "Weird stuff". Am I to assume then, that you live in the Bay area? I used to frequent that place almost once a month. Used to get some cool, old SUN personal workstations, and gut them for mod cases back in the day. Haven't been there in years, but what a great toy store :)

We should get together and geek ; )

[ajf]

Palo Alto.  I used to go there a lot in high school and when home from college, mostly to pick up cool old school (50-60s) test equipment and Apple IIes. Would also salivate over the NeXT, SGI, and SUN boxes which were too expensive for me at the time - would have cut into beer money:-) Probably for the best, as dad trashed most of what I had acquired 'cause, apparently, they were taking up precious attic space needed to store jars of random nuts and bolts ]:-|

Started going back a few years ago, and have found some great stuff for building CNC machines.  I also hit Halted and Alan Steel & Supply once in a while, not as good as WS but still pretty cool.

I'm always up for a visit...

[Atrix]

This is really good work so far, would it be possible to increase the surface area of that bar by cutting slices into it or in such in a way that it uses the fan as well?

would copper be more efficient  

[ajf]

I kind of gave up on the whole added heatsink idea because I didn't think it would be broadly applicable. Not everybody is going to have the materials and equipment needed to make their own... The final part I made was a longer bar with a pre-made heatsink attached, worked fine, but cumbersome and not easy to replicate.

[Peter Traugot]

Nice. I'm in Foster City. Yeah, I saw those NeXT machines piled up. I used to work for 3DO (game company) back in the day. NeXT was right around the corner from me.

Same with the SGI machines. what a shame. Such cool hardware. I was with EA when they got their first "Reality Machine". I think it went to 22 cores at the time.

Is there a way to PM on this board?

[ajf]

Ha! I was a big fan of Trip back in those days, though not much of a gamer. What ever happened to him...? It's all quite sad what's happened to the Valley. I worked at a startup in the basement of a downtown PA Digital building for a while..., it's now a law office *shrug*. I get a bit choked up every time I walk by... Those were the days!

No PMs, but you can send private email using the message actions drop down button.

[Peter Traugot]

Lemme know if you got the mail I sent.

[Peter Traugot]

Trip went on to lose money for lots of investors through 3DO (I was there for roughly 8 years), Digital Chocolate, and an obscure little game company called IF. It had good intentions, but it was like "Edu-tianment" attempts back in the day.

On Monday, December 21, 2015 at 1:22:41 AM UTC-8, ajf wrote:

[Christopher Leger]

It was a brilliant idea to use a soldering iron!  I've been experiencing progressively more difficult filament unloads over the past few weeks and I've been hesitant to change color for fear of getting into the unload jam that folks have been reporting.  (for the record, I have NEVER had a hands-off unload like they show on the videos - I have always had to pull)

Well, this weekend, it finally happened - I ran the unload procedure 2 or 3 times and pulling with a lot of force, but no release.  I was not looking forward to a complete teardown, so I decided to give your idea a try.

I pulled out my Hakko iron with the fattest tip running at 750 degrees touching to the metal shaft and I was able to release within about 1-2 minutes.  A reload following that was problem free, so I give this solution a thumbs up!  Thanks.

On Tuesday, December 15, 2015 at 9:16:01 PM UTC-5, ajf wrote:

[ajf]

Glad it worked. It would be better to warm the aluminum nut directly, the stainless shaft will reduce heat transfer and you could dislodge the teflon liner while heating it at those temps/duration. 

Once this starts happening, it's much more likely to happen again as residue from the filament will remain on the mounting nut.  I'd highly recommend using the updated Slic3r config which leaves the fan on at the end of a print until the hotend cools or adding similar end gcode to whatever slicer you are using. Avoid the cloud interface like the plague...

[Jacob Ramsey]

So for better unloads, the filament should be unloaded as soon as possible after finishing a print? For a less buggy release using New Matters native unload functionality.

[Jacob Ramsey]

Is there such Gcode for Cura? This is the program I have been using most of the time.

[ajf]

In Cura you can edit the end.gcode in the Start/End-Gcode tab if you've selected Expert>Switch to full settings... 

end.gcode should look something like this:

G92 E0
M106 S255 ; Turn fan on to 100%
G0 Z130 F300; Go high
G0 X77.7 Y52.3 F3000
G92 X0 Y0 ; Change Offset back to center
G0 X0 Y50 F3000; Go front and center
M109 S28 ; Wait for hotend to cool
M104 S0 ; Turn off temperature
M107      ; Turn off fan
M84        ; Disable motors

The two lines that do the cooling are:

M106 S255 ; Turn fan on to 100%

and

M109 S28 ; Wait for hotend to cool

I'm not very familiar with Cura and haven't actually tested with this yet. Will do so and report back.  However, I'm concerned that Cura doesn't allow you to turn off auto cooling and just leave the fan on all the time, at least not as far as I can tell.  That's an issue because in my previous testing the MOD-t firmware would not turn the fan back on again after it had been turned off.

[ajf]

The above end gcode does seem to work fine with Cura.  And, I think my issue with having the fan not turn on again was related to the print head going to max Z height, which for some reason causes the firmware to shut everything down. 

NewMatter uses this line in their end gcode to lift Z after a print finishes:

G0 Z130 F300; Go high

The normal way to finish a print would be to home Z in firmware.  In my case, because I use 1/4" thick glass as the bed this absolute Z height (Z130) was too high.

[JP]

So I've tried all the suggested fixes here, soldering iron held to the nut on the head, no dice..I've still got a bit of filament stuck up there.  I was going to try a small screwdriver/awl heated with a torch for a few minutes to see if maybe i can work it out that way..i've emailed support but their response time is WAY slow...and I'd rather not be without a printer or have to void the warranty...

Anyone else got any sure-fire ways to fix this??

[ajf]

It's going to be difficult to remove any stuck filament that broken off inside without opening the print head to detach the ground wire connected to the aluminum plate where the nut is... 

You could try removing the hotend, then heating the nut and trying to push a length of filament through the entry hole on top of the print head to dislodge the bit that's stuck.

Also, not that the warranty is only 90-days, so if you are close to or past that, opening it up is the best option.  If you do choose to open it, I recommend:

1. Remove hotend.
   
2. Lowering the print head a few inches.
   
3. Laying MOD-t on it's back (with the build tray removed of course).
   
4. Unscrewing the five screws the hold the bottom on.
   
5. Pulling the bottom off; The vertical smooth rods are attached and will pull out of the print head.
   
6. Unscrewing the top of the print head and swinging it aside.
   
7. Lift out the fan and set to one side (you can detach the wiring or not).
   
8. Use 1/4" socket on a driver to remove the nut on the grounding screw where the white wire connects to the aluminum plate (be gentle with the wire.)
   
9. Remove the 8 screws on the aluminum plate (The extruder is held in place four around the hotend mount, so either disconnect the wiring harness on the extruder so you can remove it entirely. Or be prepared to support the extruder such that you don't put any stress on the connector/wires or damage the encoder disc)
   

Reverse the steps to reassemble.

[Peter Traugot]

What is encoder disk meant to do? I'm completely unfamiliar with this. I mean, my garage door opener has a similar dusk, but I digress.

[ajf]

The encoder provides speed and position feedback to the driver to allow for controlling the motor accurately. 

[nappen]

Just happened to me after my 5th or 6th print. And drat, my soldering iron is in storage. Guess it is an excuse to go get a new one :)

Very frustrating experience. 

[JP]

I spoke to NM support, they're shipping me a new printer, 3 weeks estimated delivery...seems like if this is a design flaw this could possibly happen again?  The tech suggested that maybe the filament I used wasn't 100% PLA and contained some other materials that weren't fully heating up and causing the jam.  I'm not sure but the fact that the heat dissipates up the feed tube into the carriage seems a little odd...

He did say that this was something they are aware of and are working on it, so I guess that's good...

In the mean time, I am without a printer until I get the new one and I'll have to be more careful when printing..

[ajf]

Ugh. How utterly frustrating. Sorry you've had to go through this... Please don't let it dampen your enthusiasm for 3D printing!

I have emailed and chatted with support a couple of times and they are very nice and seem genuinely interested in helping resolve issues. Feel sorry for them as it looks like they are being overwhelmed at this point.  That being said, I don't get the sense that the support folks are particularly well versed in 3D printing. During one chat I asked if they were planning to offer connectivity via standard 3D printing host applications, I had to then explain what a host application was..., only to get the response that no they would not because it would mean opening up their firmware. For the record it would not, but that would be a good thing as there are some of us out here who could fix many of the firmware issues in short order.

The jamming described in this thread, and in your particular case, has nothing to do with the quality of the filament. It is completely due to the poor design of the hotend, the mount and the firmware. All of this was quite avoidable and implementing a firmware fix (though not the ideal solution) would take all of about 10 minutes.

[JP]

AJ, thanks for the reply, this doesn't desuade me from 3D printing, nor the MOD-t for that matter, being my first printer (and only thus far) I was seriously considering purchasing a 2nd one until the clog happened.

As with anything new and getting in on the bleeding edge (while I realize 3d printing isn't new per se, but New Matter is to the playing field and it's fairly certain they have some hurdles to overcome) there will be hiccups that get sorted out along the way.

One thing I noticed ironically today I recieved a Maker email about a new hot-end they've developed with less clogging:

https://store.makerbot.com/smartextruder

They speak about how they've extended the PTFE tube up the feed nozzle to prevent clogging, I'm hoping this is something that maybe New Matter can learn from and only enhance their product.

The biggest issue I see with a MOD-t versus any of these other DIY models is the lack of user-exchangeable parts (aside from the extruder and print bed)

For a hobbyist and DIYer I wanted to get into 3D printing with a good rock solid machine, and aside from this hiccup, the MOD-t has been just that and I've been impressed.  Are there better machines out there for same/less cost?  Possibly..but at the time I backed the project (almost 2 years ago now) for the cost they were one of the only fully assembled and backed with warranty and support that I found and I honestly think I made the correct choice for an entry-level unit.

[ajf]

Glad your still enthusiastic! 

Needless to say that I'm somewhat less impressed with the MOD-t, but I'll save further criticism for another day and thread...

I've never used a MakerBot myself, but they were kind enough to send out a rep and a Replicator2 to be in a booth I had at a hackathon a couple years ago. Seemed like a well built machine, didn't have the SmartExtruder yet though...

Agree that the lack of user serviceable parts is a big problem with the MOD-t. Technically, it's only the hotend that can be user replaced as the "extruder" would also include the coldend (drive motor part.) And it's actually that part that needs to be serviceable most of all.  Over time the drive gear will start to get a build up of filament scrapings in it's teeth and lose the ability to push filament.

The usage of PTFE in the SmartExtruder is quite different than how it's used in the MOD-t hotend. In the SmartExtruder it's used to reduce friction and guide the filament from the drive gear into the hotend. It doesn't actually go very far into the hotend and nowhere near the actual heated area because that's a really bad idea. There is a transition from the PTFE to a very precise and highly polished stainless steel feed tube, probably right at where the cooling fins end, that then guides the filament into the melt zone at the nozzle.

In the MOD-t hotend the PTFE, unfortunately, is used to avoid the precision machining of the stainless feed tube.  It's constantly exposed to heat and that's not good.

[Jacob Zdvorak]

So I've had my third filament jam.  First time I sent the printer back and got a "new" printer.  Worked (I use that loosely) for a dozen prints then jammed again.   I used the method outlined here with success.  Another half dozen prints and it jammed again.  Once again I used this method and was able to remove the filament.

This time however I was not able to load new filament. I pulled off the hotend and shinned a light up through the hole where the hotend attaches(where the nut is) in the carriage and looked down through the top hole where the filament goes in. (Please excuse my lack of correct terminology, new to all of this)  I could clearly see the light coming through.  No blockage that I could see.  I could see what looked like little gears or teeth which I am guessing push the filament through.  With the hotend still off, but plugged into the wires, I started the load sequence to see if I could see those teeth moving.  They were not.  I tried pushing some filament in to see what would happen and it would not take the filament.  Also out of curiosity I shoved some filament into the hotend which I had plugged back in and just let dangle from its wires and the filament heated up and came through like normal (hope this wasn't bad for it but this is how I learn).

Is it possible there is a piece I cant see keeping the drive gear from working?  I saw that @ajf said "Over time the drive gear will start to get a build up of filament scrapings in it's teeth and lose the ability to push filament.".  Is it possible that is what is happening?

So after all of that my question is, are there any suggestions?  I am split 50/50 on sending it back.  I got into this hobby after a lot of reading and fully expected to have problems and needing to service the printer myself.  The problem is the design of this printer makes it a pain to access the cold end/drive servo while also voiding the warranty.  I'm leaning toward saying forget the warranty lets break this thing open and learn how to service it myself.  Anyways, thanks for any suggestions!

On a quick side note which I may start a new thread about; I wonder if there is a way to take the carriage apart and modify it or leave it apart to make servicing easier.

[ajf]

First thing to check if the filament won't load is that the print head isn't maxed out at Z height. If the grommet on the top of the print head is touching the retaining nut on the screw the motors won't work. You can manually move the head down using the white gear on the back of the print head.

I'll be very frank here, I would suggest you return your MOD-t now. It's never going to be a reliable printer without an extensive, pretty much complete, redesign. And, more importantly, I'm pretty sure now that the people running New Matter don't care one bit that they've made a horrible product. They've lined their pockets with VC money and that seems to be all they care about. 

You can get much more reliable and configurable printers for what NM wants for a MOD-. Even at the crowdfunding prices, you could easily build a better machine without too much trouble.

regards,

aj

[Jacob Zdvorak]

Thank you for the reply ajf!

If they offer me a full refund at this point I will gladly take it and put it towards a new printer.  I've always been interested in the delta style printers.  Now that I've seen the potential of 3D printing I need one that works! =)

[ajf]

Jacob, my suggestion would be not to wait for an offer, but make a demand.  I've been hearing privately from people in a similar situation getting pushback from NM on refunds. That kind of makes me angry as both a 3D printing evangelist (self-proclaimed!) and a product guy.  If they give you any trouble, let me know...

regards,

aj

[nappen]

I had the jam, ordered the soldering iron...mine is in storage. But called NM whilst waiting. After going through the standard protocol stuff, I pulled the hot tip and there was a small tiny bit of filament on the gear side that I could get my needle nose pliers on and they had me pull it through from the extruder side. There was a divot in the filament-looked like a bite out of the filament. They don't know why it happened, but nothing looking melted. They thought maybe bad g-code, but I used astroprint with the mod-t profile, so who knows. Has been working since (2 days) except on the first print after doing it, the hot tip dove into the tray leaving a hole.

I considered asking for a refund, as well, due to a lot of issues with the shipping etc, but am trying it for now.

[Shahar Zrihen]

Sorry if I'm hijacking the post. 

I've been having similar issues with my unit.

It ran well for 3 weeks or so, even changing spools mid print without a glitch until it got stuck once. I contacted support and they did a skype session with me and taught me how to fix this. they replied quickly and scheduled a session within 2 days. 

since my first filament jam I had to open the printing head every time I change filaments and even when it breaks mid print. 

Now I have a new issue - after the last change I started a print and it got stuck in the middle and "air printed" above the material. I had a jam in the hot end. I opened the printer head and cleared it, cleared the hotend and did a load. it loaded fine and melted away properly. 

I started a test print and it got stuck once again and started "air printing" again. 

has anyone encountered this type of jam?

I've emailed the support twice and got no response for almost 10 days now. I'll try to call them tonight (it's a bit comlicated due to hour difference)

[ajf]

There are a few things that could cause what you are describing, perhaps in combination.

• Hotend is bad and cooling down mid-print.
  
• Drive gear is slipping due to build up of filament shavings from grinding.
  
• PTFE liner in hotend has been damaged by prior jams and/or the heat sink issue described here.
  
• Filament has absorbed moisture and swelled at points causing it to jam. (least likely, I think.)
  

If it's the hotend, you'll need to request a replacement.

The drive gear can be cleaned with the supplied brass bristle brush or a hard bristle toothbrush, etc... It's easier to clean if you remove the extruder from the print head, this should help with that. It can be done without removing the extruder, just a bit of tight space to work in though.

In the unlikely event that it's swollen filament, you can dry it out by putting the spool in an air tight container with a couple of cups of uncooked rice (best to have the rice in a cloth bag or dish towel or otherwise separated from the filimanet.) If you are in a particularly humid environment, this is also a good way to store filament when not in use.

I'm hearing from a lot of people that response times from NM support are long.  Suspect they are being inundated by requests.  They don't seem to be posting responses at the support site of late.

[Jacob Ramsey]

I am having a hot end shipped, as my current one stopped working. They claim it to be an improved design. Any one have insight on the improvement?

[ajf]

I got my replacement last week.  The only difference is that the nozzle is stainless steel rather than brass.  Well, the PTFE liner in the new one falls right out, wouldn't call that an improvement though...

[Jacob Ramsey]

If the PTFE liner falls out, just put it back in?

[ajf]

Yes.

Knowing better, I went ahead and tried removing the steel tube from my old hotend cold. Which snapped it.  Probably of the best since now I have excuse to replace the tube with one that has a polished interior and won't need PTFE at all.

[Tyler Braddick]

What exactly did you do to get a replacements? 0.7.0 isn't helping my constant jamming so I feel like maybe it's my printer

[JP]

AJ, 

let me know if you get the tube replaced and what you get it replaced with..if it works well I'd like to try it myself :)

[ajf]

Will do.  Still looking for a stock part that might be suitable.  It'a also probably going to require a modification to the mounting nut or a custom replacement.

[ajf]

I only got a replacement hotend. Just emailed support and sent them a picture of the damaged part.

[JP]

well 2nd printer filament jammed up in the carriage.   i can see the spinning feed mechanism when i look in from the top, and from underneath there's a blob of filament that's stuck in there.  hot-end is clear it's up in the carriage again.

I believe this is a printer issue and not a filament problem, this is different filament than I used before and I noticed after the 0.9.0 firmware update, after i print from gcode file (advanced > print file from app) the fan doesn't stay on after the print finishes.   Unless I'm there to manually unload the filament, it seems like this will keep jamming (this print ended while I was down eating lunch, it sat just long enough to congeal and now I've got a 2nd RMA to do)

[Peter Traugot]

Ack! What a terrible design, sealing the feeder up like that. I mean, it's one thing to have industrial design and a zero maintenance approach, but NOT when the machine is so prone to jams!

--
You received this message because you are subscribed to a topic in the Google Groups "MOD-t" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mod-t/Pgdlp3kf2pQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mod-t+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mod-t/211cd9a6-dd0a-4107-a8e7-916e73c419ab%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Greg L]

On Saturday, January 30, 2016 at 12:28:58 PM UTC-7, JP wrote:

it sat just long enough to congeal and now I've got a 2nd RMA to do)

Is the only option in your case to RMA? I think I'm currently in the same boat with a jam in the feeder ... sucks.

[JP]

At first it was, then I thought to myself heck, I'll attempt the tear down and clear it myself..it was rather stressful but i figured it out so no need to RMA.

[3DPrinterJedi]

I too have just called in my second RMA for the same damn issue. I'm a newbie to 3D printing and right after I formed a 'plug' on the first filament gopping up\breaking off and getting stuck in the pathway (in a place unable to get to) I called it in and New Matter sent a replacement Mod-T. I did not even get through the first print and clogged it again over night! At least with this new RMA they are sending out a replacement this week and will have me send in Mod-T#2 with those packing materials (though I kept everything from before).

New Matter wanted to look at the gcode and profile\settings I had for Cura and they told me they were wrong and provided what I will need in the future. My friend suspects that I had too much retraction and it is pulling the hot filament back in too far as it spatially relocates the extruder across the build plate and is hardening before it can extrude more PLA. This recoil is likely forming a plug as the poster described. I'm curious if I take a soldering iron and touch that carriage nut long enough will it heat the plug of PLA and allow me to push a pin through it to loosen it and then maybe I can come in from the top of push down with new filament to get things moving again?  There is nothing I can grip from the top or bottom so I'd like to dislodge this plug somehow (if it happens in the future again with Mod-T#3).

[Christopher Leger]

can you please provide the Cura settings that they stated were "wrong"?

[ajf]

The cause of the jamming in the retaining nut is most definitely not related to any settings in the slicer.  It can be mitigated by the end gcode generated but the slicer as I've described and implemented in my Slic3r config (and the Cura one that I have yet to put on github, but is in theses forums.) The cause it the faulty hardware design. 

[3DPrinterJedi]

Here are the settings I had that they said were incorrect:

[3DPrinterJedi]

Specifically she said my start gcode was wrong.  She gave me the attached profile to use. I also attached the New Matter screenshots on the settings to use.

[3DPrinterJedi]

I intend to make a heat sink too to minimize this from happening on the next Mod-T that arrives. I plan to cut (I guess with a hacksaw) a large heat sink from an old junk computer's CPU, drill a small hole in the middle and then find a tap and die kit that has an M5 bit to carve out the threads to screw it onto the extruder assembly shaft. As you say if we can prevent the heat from the nozzle from going up and inducing that retaining nut it should reduce the likelihood that PLA will melt and get pulled back up to where it can form an immovable plug.

I'm still learning, but what is a recommended retraction amount in Cura? I want to avoid the mechanism pulling too much filament back up in the pathway also. It's shitty I have broken two MOD-Ts over this same issue! I want to print things!!!

[Derek Schulte]

A 10-32 tap will usually suffice for an M5.  Also note that M5 is the standard thread for bicycle accessory mounts like water bottles and panniers, so it's actually fairly common.  If the aluminum "nut" that is retained in the perforated heatsink is getting too hot (or you believe that it is too hot), then chances are that the spring washer (inside the carriage) is no longer keeping the nut adequately mated to the heatsink.  This often occurs after rotating to nut a lot.

[Ben Cichanowicz]

I got the second replacement Mod-T yesterday and made this modified CPU heatsink thin and threaded. I invoked a print job and immediately did a filament unload when it finished. The heatsink was just warm. I removed the extruder/heatsink and checked that the pathway was clear. I then put the extruder and heatsink back. 

That was my first time using a tap and die kit. I messed up one side so that's why the fins are pointing down. It appears to have done the job. I hope this minimizes any chance of forming a PLA plug inside the carriage again. I'm so over that!

--

You received this message because you are subscribed to a topic in the Google Groups "MOD-t" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mod-t/Pgdlp3kf2pQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mod-t+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mod-t/32eb6ba3-9e7b-44a1-9256-1cd8b1c0e8c2%40googlegroups.com.

[ajf]

The problem with this is that it blocks the fan from cooling the print, which will result in poor print quality with PLA.

I'm going to replace the existing fan with a couple new ones. One for cooling the hotend and the existing mounting plate/heatsink, and the other vented to blow directly on the filament as it's being extruded.

[3DPrinterJedi]

Yeah I was wondering about that. Maybe I can go ghetto and put a small fan off to the side to blow a gentle breeze over whatever I am printing?

[ajf]

Lot of people do that, totally works.

Going to pick up some 24v fans and see if I can come up with a cheap and easy retrofit to kill both birds. With the current, huge, blower fan gone there should also be enough room to add a controller board to spoof temp (and maybe other stuff,) so I can print other materials like TPU (NinjaFlex, etc....) Also, going to mod or replace print head cover so I don't have to disassemble the machine just to get access to the extruder.

[JP]

I've already taken to modifying my carriage to prevent easier access...basically cutting apart flat areas where theres no need without compromising support.  I might have to reinforce some areas but that shouldn't be a problem..

[ajf]

Pictures or it didn't happen...

My current plan is to laser cut a new top plate out of acrylic and print mounts for magnets to hold it in place. None of my local sources had small 24v blower fans so I had to order them from Amazon and they won't arrive until 02.22.

[Ben Cichanowicz]

Yeah I'd like to see pics of the access holes you cut! Awesome you guys have done the work to uncover the root cause and then mitigate it. I recommended to New Matter that they should machine these heat sinks for everyone. If I made mine with lots of holes would it still absorb heat but allow the blower to get through?

On Thursday, February 18, 2016, ajf <ajf...@gmail.com> wrote:

Pictures or it didn't happen...

My current plan is to laser cut a new top plate out of acrylic and print mounts for magnets to hold it in place. None of my local sources had small, 24v blower fans so I had to order them from Amazon and they won't arrive until 02.22.

On Thursday, February 18, 2016 at 7:07:00 PM UTC-8, JP wrote:

I've already taken to modifying my carriage to prevent easier access...basically cutting apart flat areas where theres no need without compromising support.  I might have to reinforce some areas but that shouldn't be a problem..

--
You received this message because you are subscribed to a topic in the Google Groups "MOD-t" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mod-t/Pgdlp3kf2pQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mod-t+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mod-t/07e220da-f541-4498-a6f0-e50a16212cca%40googlegroups.com.

[ajf]

For my final heat sink iteration I used a piece of Al bar stock to offset the heatsink. Don't have a picture of it installed, it was clunky but worked (48g and lots of surface area):

New Matter is working on a permanent fix, not sure when that's going to roll out or how retrofits will be handled. 

[Ben Cichanowicz]

Oooo I could probably do that too to get it out of the way of the blower and probably use the tap better this time. I barely carved enough treading on the heat sink by itself

--
You received this message because you are subscribed to a topic in the Google Groups "MOD-t" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mod-t/Pgdlp3kf2pQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mod-t+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mod-t/464dd125-abaa-4993-b172-709f0f8af070%40googlegroups.com.

[Norman Wolfe]

hell yeah. i totally had this problem and my soldering iron fixed it in no time. the only difference was i put the printer into loading mode while heating it in order to make it just flow throw whenever the jam had cleared. worked like a charm. cheers!

[Christopher Leger]

With the demise of NewMatter, I've started thinking about the future. I'm on hot end #2 and this will not last forever.
I'll likely get another printer, but I'd like to keep my ModT running as it has been really pretty flawless for a year or two since the firmware bugs were ironed out.
I recall that someone posted a teardown of the ModT hot end. Can someone point me to it? I can't locate it.
Can't hurt to start looking for a source or a compatible end.