Replicator 2 Driving Me Crazy with Constant Jams - MakerBot No Help at All!

[Rob Jones]

So my Replicator 2 has been working great for several weeks - lots and lots of builds using a combination of OEM and aftermarket filament with no problems at all. Then it jammed. Now for the last TWO DAYS I've been trying to get it working again and after 50 or attempts, I have gotten exactly ONE good print.

I'm out of ideas, theories and patience. And MakerBot themselves are no help at all. I submitted a support ticket - no reply. I tried calling and after nearly an hour on hold, I gave in to the constant nagging to leave a message for a call back. When I entered the prompt to leave a message, the mailbox was full!!

Here's what's going on, and what I've tried to do to fix it:

I have the newer style spring loaded extruder and both fans are working correctly. With the extruder temperature set to 230ºF, the filament will jam within 15 minutes of beginning a print. This is evidenced by a fairly loud clicking noise from the extruder, the filament no longer being fed into it, and "air printing" with no filament coming out of the extruder nozzle. When I unload the filament, it is chewed up from the drive wheel on the stepper motor digging into it. There is often a semicircular divot dug out of the filament from the extruder drive wheel.

If I unload the filament and reload it, it will load fine. If I leave the Load script running, filament will feed with no problem whatsoever. The jams only occur during an actual build.

I experimented with different temperatures using a build file that has worked flawlessly before now. At 215º it jammed within 2 two minutes (before the raft was even finished). At 245º the result was about the same as 230º. At 250º it went a little longer before jamming. I know it's way too hot for PLA, but I also tried 260º to see if it would make a difference. It didn't jam until about 22 minutes into the build, but still jammed.

I completely disassembled the extruder, cleared all PLA out of the nozzle and barrier tube with a propane torch, removed and cleaned the heater element, replaced the ceramic insulation, thoroughly cleaned the heater block, and carefully reassembled everything. It made no difference at all and there is no telling how much filament has been wasted so far.

Is there anything else I check or try before I take this thing out behind the woodshed and put a bullet in it?

Is there another 3D printer on the market with similar capabilities, uses the same filament, and is roughly the same price as the Replicator 2? For the most part, it's an awesome piece of technology - until it isn't, then I feel like a beta tester and I'm really tired of the lousy support from MakerBot. If there is an alternative that won't negate my considerable investment in filament, I'd really like know about it. I've been seriously considering a 5G Replicator, but after this experience, I have my doubts that would be a wise purchase.

[Dan Newman]

On 22/05/2014, 3:36 PM, Rob Jones wrote:
> So my Replicator 2 has been working great for several weeks - lots and lots
> of builds using a combination of OEM and aftermarket filament with no
> problems at all. Then it jammed. Now for the last TWO DAYS I've been trying
> to get it working again and after 50 or attempts, I have gotten exactly ONE
> good print.
>
> I'm out of ideas, theories and patience. And MakerBot themselves are no
> help at all. I submitted a support ticket - no reply. I tried calling and
> after nearly an hour on hold, I gave in to the constant nagging to leave a
> message for a call back. When I entered the prompt to leave a message, the
> mailbox was full!!
>
> Here's what's going on, and what I've tried to do to fix it:
>
> I have the newer style spring loaded extruder and both fans are working
> correctly. With the extruder temperature set to 230ºF, the filament will
> jam within 15 minutes of beginning a print. This is evidenced by a fairly
> loud clicking noise from the extruder, the filament no longer being fed
> into it, and "air printing" with no filament coming out of the extruder
> nozzle. When I unload the filament, it is chewed up from the drive wheel on
> the stepper motor digging into it. There is often a semicircular divot dug
> out of the filament from the extruder drive wheel.

Sounds like you are overextruding -- putting out too much filament and as
a result, the nozzle cannot eject more plastic because there's too much on
the print already. Have you ever calibrated your extruder by adjusting
the feedstockmultiplier (MakerWare) or filament packing density (RepG)?
It's a different value for ABS (~0.85) vs. PLA (~1.00). [Makerware uses
a single value for both -- they split the difference between the two and
use an average value which is pretty silly actually.]

Some folks even recalibrate for each spool of filament. It primarily
changes with the softness/hardness of each plastic. (The value is primarily
compensating for how much the pinch gear digs into the filament thus
changing the effective turning radius of the gear and thus how much
plastic is fed in per motor revolution.)

> I experimented with different temperatures using a build file that has
> worked flawlessly before now. At 215º it jammed within 2 two minutes
> (before the raft was even finished). At 245º the result was about the same
> as 230º. At 250º it went a little longer before jamming. I know it's way
> too hot for PLA, but I also tried 260º to see if it would make a
> difference. It didn't jam until about 22 minutes into the build, but still
> jammed.

Increasing the temp likely did two things

1. Reduced the viscosity a bit making it easier for the nozzle to eject
more plastic,

2. Kept the plastic of the prior layer a tad more soft/pliable a bit
longer possibly again making it a little easier to eject plastic.

> Is there another 3D printer on the market with similar capabilities,

If the issue is indeed extruding too much plastic, then the issue is the
same with all printers. It's just a matter of the manufacturer educating
you about this issue and using a slicer which understands the difference
between ABS and PLA and doesn't use a lame value split between the two.
In the Cupcake and Thing-o-Matic days, MBI used to be on top of this and
educated people about it. However, with the Rep 2 they want everyone to
just think it's all plug-n-play and they stopped talking much about the
need to do calibrations.

Dan

[Clinton Hoines]

Are you using clear PLA? If you are switch to a different PLA I never had any luck at all with clear.

If not once you clear your nozzle again put a few drops of olive oil on the filament when you feed it in. Sounds crazy but it works for PLA.

[Robo]

Even 230 is too hot for most filaments, there's a lot of people here wondering why Makerbot uses that as their default. Turning the temp up even higher to attempt to remove clogs can actually make it worse - I had the best results just letting it sit there heated at normal temp, and then using guitar strings above and below the nozzle to work the clog out. (smaller than .4mm, I also play guitar so I have extras)

Overextrusion and slow speeds can contribute to the clogging in the first place, others can describe that in better detail than I can, I would just say you should work on tweaking your settings with some test cubes.

[Ryan Carlyle]

Unfortunately, many of your troubleshooting steps probably made things worse. Going to 260C with PLA in particular was a mistake. If your nozzle wasn't full of carbonized crap before that, it definitely is now. Torching alone is not guaranteed to get filament /dust deposits out and won't help at all with contamination or tip deformation blocking the nozzle.

There could be a lot of different things wrong now. My serious recommendation is to "reset" your printer. Replace your nozzle. (They're cheap -- $5 from QU-BD.) Go back to a Makerware default profile, except print PLA at 210C instead of 230C. Use a new spool of a different color and/or vendor of filament. Put some canola oil on the filament before you load it. Make sure it extrudes straight down when you load it. Put fresh blue tape or whatever on your build plate and tram it with a leveling print or proper feeler gauge instead of a business card. Then immediately print a 100% infill 20x20x10 box and check for over/under-extrusion. Adjust feedkstockMultiplier as needed to calibrate your extrusion volume. THEN try a real print and let us know what happens.

[Rob Jones]

Thank you so much for the replies and suggestions. I had no idea the calibration was needed - or even possible. Thanks again, Makerbot!

I looked up how to print the calibration cube and tried it at 210º with the feedstockmultiplier set to 1.00 with a 100% infill. Guess what? Still jamming!

I have a new nozzle on hand and swapped it out before trying the calibration print.

The only clear PLA I have is the roll that came with the printer.

What should I tweak in the config to get it working properly? feedstockmultiplier? feedDiameter? Anything else?

[Rob Jones]

I upped the feedstockmultiplier to 1.05 and changed the diameter to 1.73 and got a really nice looking test cube. I had no idea the Replicator could produce such a nice finish (very smooth with sharp corners). It sure would be nice if this information were more readily available! That would have saved me a TON of aggravation and the Swear Jar wouldn't be nearly full.

It's been a long day and I don't feel like waiting an hour to see how a "real" part prints. So I'll try one in the morning and post the results.

On Thursday, May 22, 2014 6:36:38 PM UTC-4, Rob Jones wrote:

[Dan Newman]

On 22/05/2014, 7:08 PM, Rob Jones wrote:
> Thank you so much for the replies and suggestions. I had no idea the
> calibration was needed - or even possible. Thanks again, Makerbot!
>
> I looked up how to print the calibration cube and tried it at 210º with the
> feedstockmultiplier set to 1.00 with a 100% infill. Guess what? Still
> jamming!

You measured your filament diameter before slicing and input the value?
That's important as well. I have plenty of 1.75 mm filament which is 1.82 mm
and some that is 1.68 mm. The slicer needs to know the actual diameter to
do its job. And the diameter can slowly change even on a given spool.

Dan

[Ryan Carlyle]

Basically, volume calibration is something you do for each TYPE of filament and any time you change extruder hardware. It's a parameter that represents how far the drive wheel teeth bite into the filament. ABS and PLA have different values but most spools of PLA will have about the same number for your particular printer.

If you measure your filament, then calibrate packing density / feedstock multiplier, in the future with similar filament all you have to do is measure and input the filament diameter. The main volume calibration stays pretty consistent as long as you measure filament diameter regularly.

Yes, Makerbot should tell people this stuff.

[Rob Jones]

Yes, I measured the filament with a digital caliper. There is some variation on different spools from about 1.67 to 1.73. I set it for a good middle ground so I hopefully won't have make a new build file for every part if I change filament.

It's great that the printer is so tunable, but incredibly frustrating when you don't know it can be tuned, or how to tune it. This information really ought to be more readily accessible.

I printed 3 test blocks and all worked flawlessly - but took less than 15 minutes. Around 16 minutes seems to be the threshold where prints begin to fail. So I printed a part that takes about 45 minutes and 16 minutes into the build, it failed (as usual). This is driving me crazy because the printer had been working fine until a couple of days ago.

I have attached a copy of the build file with the tweaked settings. Would somebody be willing to look at it and let me know if they can see the problem with it?

On Thursday, May 22, 2014 6:36:38 PM UTC-4, Rob Jones wrote:

[Franc Falco]

I've only quickly scanned the replies on this so my suggestion may have already been mention?
But, did you take extruder heat-sink off when you first cleared the block - If so, could you maybe have put it back on the wrong way around?

I had a very similar problem and it turned out to be exactly that.

[PrettySmallThings]

Installing the cooling fan backwards is also a common mistake. Sticker should be pointing in not out.

[Dan Newman]

On 23/05/2014, 7:51 AM, Rob Jones wrote:
> Yes, I measured the filament with a digital caliper. There is some
> variation on different spools from about 1.67 to 1.73. I set it for a good
> middle ground so I hopefully won't have make a new build file for every
> part if I change filament.

That works sort of as long as you are printing small, not very tall pieces.
Even then, you won't have as good of surface finish as you would get if
you actually measured the filament and then sliced using the correct diameter.

The overextrusion is cumulative so the more plastic output for a piece,
the worse it gets. You want to set the diameter correctly each time you
slice. And if you switch filamanet you may then need to reslice.

Dan

[Dan Newman]

On 23/05/2014, 8:26 AM, PrettySmallThings wrote:
> Installing the cooling fan backwards is also a common mistake. Sticker should be pointing in not out.

A more definitive rule is to look for the molded direction arrows on the sides of the fan: there
should be two, the first showing the direction the fan rotates and the second the side from which
air exits the fan, the direction of airflow. MBI is at times haphazard as to which way they
install the fans: blowing onto the heatsink or away from it. And you may get different opinions,
however it's my understanding that it should be blowing onto (towards) the heatsink.

Dan

[Bryon Miller]

This is exactly what I would try doing.  Make sure your fan is blowing air towards the heatsink and not away.  I put my extruder back together incorrectly like that once and it caused jams to happen often.  Once that was corrected, the problem was instantly gone.

[Rob Jones]

I checked the fans and they are installed correctly.

I made the test cube a little bigger so that it would take longer to print in order to see if it would fail. At about a 30 minute it worked great. So I copied the config file and pasted it into the config file for a part I'm having trouble with. The only parameter I changed was the infill (changed from 1.0 to 0.2).

Less than 30 minutes in, the build failed. How come the test cube prints fine, but my part fails every time? I'm using the exact same config file - except for the infill density. What gives?

Could this be related to the filament I'm using? I checked it's diameter and it's very consistently 1.74 - 1.76mm, so it should be fine. But I did two 2-hour prints in a row with another filament and had no problem. I measured that one and it averages around 1.68mm. So it would seem that the slightly thicker filament is causing a problem. I incrementally upped the diameter in the config file to as high as 1.82, but it still fails within 30 minutes. But then a test cube prints perfectly.

Oh, and I did finally get MakerBot on the phone. They had no useful advice to offer.

Anybody want to buy a Replicator 2 - CHEAP!

On Thursday, May 22, 2014 6:36:38 PM UTC-4, Rob Jones wrote:

[PrettySmallThings]

Yes, there are bad rolls of PLA. Even sometimes from reputable brands. Sometimes it's inconsistent diameter, but it can be the formulation itself.

If you haven't tried a different roll of PLA, you should. Personally, it's the first thing I try when I have a failed print.

[PrettySmallThings]

[noelani72]

I too use MBI and after market 1.75 PLA in my rep2 (with the extruder mod) and have had this problem a couple times.

My Rep 2 is nearing 1,000 hours and this is my fix - clean the thermal barrier tube..

During the preheat at 230 degrees, with the extruder drive assembly out of the carriage, I work a piece of PLA down in that tube.

Insert until you feel pressure, pull out, snip end off. 

Repeat.

I do this a good dozen times until I hear a suction sound.

Wow, an unbelievable amount of gunk is left behind in that tube and coats it inside.

In my near 1,000 hours, according to my service sheets I keep on a clipboard, I have done this once using nothing but MBI PLA at the 600 hour mark (discovered when changing PLA colors).

I have done this three times in the past 400 hours - and I am using nothing but ToyBuilder labs PLA.

I must have the mother of all Rep2 printers because mine runs darn near perfect. I love it.

My projects as of late run 15hrs each and I have no qualms leaving the machine unattended and never get an air print.

Occasionally, due to the size of my prints, they'll start to pull my blue tape off the glass build plate, but that's to be expected sometimes.

[Bryon Miller]

How much is "cheap"?

[Rob Jones]

That's going to depend on whether or not I can get the thing to work :-)

On Thursday, May 22, 2014 6:36:38 PM UTC-4, Rob Jones wrote:

[Ryan Carlyle]

Test cubes print fine because they don't require any filament retraction. Retraction is when clogs form in a thermal barrier that has become too hot. Basically molten filament is being pulled up into a cool section of the thermal barrier tube, where it solidifies into a plug that the extruder doesn't have enough power to push down.

If your prints always fail at 16 minutes, that's because it takes the thermal barrier 16 minutes to heat up to the point where PLA starts clogging. Honestly, PLA is a really terrible material for hot ends because the glass point is so low. This makes it clog extremely easily. (But the same property makes it low-warp so it's very popular.) Your problem -- like most air printing with PLA -- is probably excess heat / insufficient cooling in the extruder mechanism. There are a lot of possible causes for this, like the fan pointing the wrong way as people have mentioned. The most popular (and surprisingly-successful) solution is adding canola oil to the filament. This helps cool and lubricate the filament in the thermal barrier so it doesn't clog. Exact same reason why you put oil in a frying pan -- to act as a heat transfer fluid and prevent sticking. People often report a night and day print reliability difference when they add a filament lubricator and a little oil.

Other things also help with heat-related clogs, like printing at a lower temp and pointing a fan at the printer. But often there's some mechanical issue that is preventing proper heat-shedding from the thermal barrier or cooling bar... and finding that issue will fix your problems.

And yes, if your filament diameters are more than a few hundredths different, you should use different diameter settings. The difference between 1.65 and 1.85 is huge to a 3d printer. Basically every time you switch filament, the first thing you should do is measure the diameter and plug that into your slicer for the next print. Then write the diameter on the spool and remeasure occasionally in case the diameter changes along the length of the spool. Or buy premium filament like Toybuilder Labs where diameter is consistent enough that you can just input 1.75mm and get good results.

[John Carbone]

first off i would like to address everyone on here who talks about the heat of the extruder and telling people to not go over 230. Where are you getting your info from ?? hopefully not the morons at customer service at makerbot !!?? I generally print 240-250 on my rep 2 and have little to no failed prints. second off to fix your issue take some baby oil and put it on a paper towel and let the filament pass through it before going into the clear tube. your getting jams because there is no lubrication left in your nozzle.

[Bryon Miller]

Wow,  that seems like a high temp.   I used to print at 230, but 215 and 210 are doing fine,  now there's no stringing on my prints.   This was caused by using too much heat.

The reason I used 230 was because that was the default and I thought it was perfect until sometime pointed out that high temps caused the subtle stringing I saw in my models.  Until then,  I thought this was just how things were printed.

On May 27, 2014 6:41 AM, "John Carbone" <ecdpai...@gmail.com> wrote:

first off i would like to address everyone on here who talks about the heat of the extruder and telling people to not go over 230. Where are you getting your info from ?? hopefully not the morons at customer service at makerbot !!?? I generally print 240-250 on my rep 2 and have little to no failed prints. second off to fix your issue take some baby oil and put it on a paper towel and let the filament pass through it before going into the clear tube. your getting jams because there is no lubrication left in your nozzle.

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/we952p_fciA/unsubscribe.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ryan Carlyle]

Makerbot's PLA can survive at 240C because of a slightly different formulation. But most PLA brands will go to hell at 240C. It's really a terrible idea. Stringing and nozzle clogs from carbon deposits if the filament stops moving. Plus a harder time keeping the thermal barrier cool.

There is one exception -- if your printer has a junk value stored in the EEPROM for "temp offset" which can mean your temp readings are flat-out wrong by an unknown number of degrees. It's a hidden calculation offset that is not user-editable and serves no real purpose in the heater code. This has happened before so people print at wacky nominal temps while their true print temp is pretty normal.

[James McCracken]

Because I don't see it mentioned anywhere here... and because you can print calibration cubes but nothing else... does your non-cube test print have a lot of tool moves in it?  Places where the extruder is moving but not extruding?  This typically causes retractions, and can retract hot filament into the cool part of your extruder, which can cause jams...

Depending on your slicer and software there are different ways to adjust it but basically there's a retract extra distance setting which says when you're travelling to jerk the filament out a little to prevent it from dripping at the nozzle... that retraction can be overagressive and cause the exact issue you mention.  Or your part can be hell on that code; I once printed something with lots of little towers, I would jam and air print within 3-4 layers of hitting the towers...

On Tue, May 27, 2014 at 10:15 AM, Ryan Carlyle <temp...@gmail.com> wrote:

Makerbot's PLA can survive at 240C because of a slightly different formulation. But most PLA brands will go to hell at 240C. It's really a terrible idea. Stringing and nozzle clogs from carbon deposits if the filament stops moving. Plus a harder time keeping the thermal barrier cool.

There is one exception -- if your printer has a junk value stored in the EEPROM for "temp offset" which can mean your temp readings are flat-out wrong by an unknown number of degrees. It's a hidden calculation offset that is not user-editable and serves no real purpose in the heater code. This has happened before so people print at wacky nominal temps while their true print temp is pretty normal.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to makerbot+u...@googlegroups.com.

[MovieMaker]

you may have a bad roll of filiment. Out of 12 that I purchased, three were bad.  They drove me crazy until I figured out it was them. The jammed the hot end, the extruder and made horrible messes. Always make sure that you have good filiment.   my 2 cents.