Makerbot replicator 2, stops extruding

[WWC Tech Lab]

Printer stops extruding at 13% completed every time. It'll keep moving but won't extrude filament. Heat remains constant. Thoughts? Is this a motherboard problem?

Thanks for your help!

[TobyCWood]

Definitely NOT a motherboard problem.

Check your temp when it fails.

Assuming it's OK... 200-215 for PLA...

Unload the filament after it fails and take a photo of the tip and post it here.

What are your retraction settings?

[Jake Clark]

Check the front fan to ensure that it's blowing the right way, ontop of what TobyCWood recommended; 

[WWC Tech Lab]

Thanks y'all!

Front fan is blowing well and in the proper direction. Photo of unloaded filament end attached. (It's white filament with sharpie marks on it that we were using to track progress through the plunger.)

[TobyCWood]

Too hard to see, but there is some indication of the tip enlarged i.e., the stuff melted too high up the extruder then bulged outward when pushed down and then shaped by the thermal tube. Hold the filament between two fingers and slide to the tip. If you feel a tiny edge towards the tip it could be heat creep.  Have you tried switching filament? What temp are you setting the extruder? What are your retraction settings?

BTW... we have had some very rare humid weather here the past few days. All 3 of my Rep2s were showing signs of heat creep-like clogs. Today the weather is back to normal and the clogs went away.

[Jetguy]

Please tell me this isn't such an old replicator 2 that it still has the plunger style feeder and you never upgraded to the free upgrade spring lever and roller style?

Because the plunger style wears out and that would be this fault, poor grip and failure to feed as soon as any retraction or friction builds up.

Not to mention you should before doing anything else perform the most basic extruder test, does it grip or does it slip?

Because if the feeder cannot 100% grip the filament, everything else is a waste of time.

The test is this easy:

Use the filament load menu and load a fresh clipped end of good filament into the extruder normally, again using the load filament menu.

One it is loaded and extruding, do NOT yet press the button to stop extrusion. Grab the filament with your fingers going into the top of the extruder (not the hot noodle out the nozzle) and stop it from going into the extruder while the motor is still pulling. Good grip will mean the motor will thump and skip steps. Bad or failed grip, the motor will not skip steps and the gear will grind a divot in the side of the filament. One you are satisfied you understand the test, stop extruding.

Bottom line: if you cannot pass this step, you have to clean your drive gear, replace or repair or upgrade your extruder feeder until it can properly grip the filament and pass this test.

[Jetguy]

https://store.makerbot.com/parts-accessories/x5-rep2-extruder-upgrade-complete-kit/

They did give these away but it's been so long, you may have missed that free upgrade period.

https://support.makerbot.com/learn/makerbot-replicator-2/printer-maintenance/adjusting-the-delrin-plunger_11372

https://s3.amazonaws.com/download.makerbot.com/replicator2/MB1476_Extruder_upgrade_Support_NEW.pdf

On Thursday, April 20, 2017 at 9:32:33 PM UTC-4, Jetguy wrote:

Please tell me this isn't such an old replicator 2 that it still has the plunger style feeder and you never upgraded to the free upgrade spring lever and roller style?

Because the plunger style wears out and that would be this fault. 

[Jetguy]

The reason why I'm harping on this is the failure. 

If you said it gets to 13 % complete, the extruder is not putting out plastic and is thumping or ticking- well then we would assume nozzle clog or hotted jam of some type and know that the feeder was working.

But that's not what was said. So the obvious assumption is, the extruder gear grinds a divot in the side of the filament. Like a car getting stuck in the mud and digging deeper as you keep spinning the tire, the same thing happens to the filament and drive gear. So once it starts slipping it cannot ever self recover and begin feeding again. This is why it's so critical to maintain your feeder and ensure from time to time and especially when there is a problem you have proper grip. The most basic rule is it must skip steps rather than guiding the divot from slipping.

Again, the way this works, there might be yet another fault, heat creep, damaged nozzle, clogged nozzle, or other dozens of failure that cause it to be harder to push plastic through the nozzle. But before that, the feeder has to never slip, Because as long as it grips and worst case, skips steps, the instant the force blocking the nozzle is removed- the feeder self feeds again completely on it's own mid print. But if it grinds a divot, that's game over. It will not self recover.

[Jetguy]

Also, please understand why the upgraded extruder exists and even in the adjustment of the plunger directions I linked, they advise you to upgrade the extruder.

Because it works like this. The original and absolutely marginal stock drive feeder system used a Delrin plastic plunger to press the filament against the drive gear to make it grip. So that means the motor constantly fights the friction of having to slide the filament past this plunger and push the filament into the nozzle, and pull the filament from the spool. Then because there is no or limited spring action here, as filament diameter varies from one type or spool to the next, as the plunger wears over time, now the feeder plunger puts less and less pressure pushing the filament against the drive gear. Finally, the straw that breaks the camel's back is when it slips and grinds the divot, again because there is no to limited spring pressure here, the pressure of the filament against the gear drops to zero. In other words, it slips a tiny bit, that grinds away the side of the filament, that reduces the pressure against the gear so it grinds easier now, that wipe out any remaining pressure and grinds the divot until the motor can basically spin free all day long and the filament just sits there not moving.

With the spring lever and roller bearing upgrade, now we have a nice ball bearing pressing and rolling the filament against the drive gear. This is way less friction than the plunger and now allows more of the motor force to push the filament into the nozzle system actually preventing minor jams that may have skipped steps before. Even as filament diameter varies, the lever and spring maintain a constant pressure on the filament against the drive gear. Under worst case conditions, even if there was minor slippage and the drive gear begins to grind a divot, the spring and lever and bearing all keep putting pressure on the filament into the drive gear, still trying to get that last bit of grip until there is just nothing left.

But all that said, even the upgraded system can fail. The levers can crack or warp from heat and apply less pressure than what was intended as it deforms. The drive gear can get get the grooved filled with dust, and then slip, and the creates more dust, and eventually it cannot grip. And, given enough hours and rolls of filament, the teeth on a drive gear can lose grip over time from wear and tear. And that's why the basic extruder test of does it grip or does it slip tells you if the "system" of the feeder is working. Any one or multiple components could fail, but when they all work correctly in unison, it should be so strong of grip the filament cannot slip and the motor is forced to skip steps.

[WWC Tech Lab]

My bad- I used wrong language in my last post.

Yes, we upgraded to the spring loaded lever extruder a long time ago- spring is still strong and the roller bearing moves freely. We also have good grip on the filament at load and a visual assessment of the toothed gear suggests that it's still in fine shape. 

This is happening with all our filaments (from a variety of makers) and across a temperature range of 205-230C. Retraction is currently set to 1.3mm.

Could we talk more about "heat creep?"

Thanks everyone!

[Jetguy]

OK, so there are parts of the extruder system that are hot (namely the heater cartridge, the heater block of aluminum, the nozzle, the thermocouple temp sensor) and then there is the cold end (the mounting bar, the motor, the heatsink, the feeder parts, the fan) and there is ONE part that connects the HOT and COLD systems together called the thermal barrier tube. It's called the thermal barrier because it's a threaded tube of stainless steel and machined to a specific shape and cross section (necked down diameter) just above where it screws into the heater block. Stainless is chosen material because it has a low rate of conduction of heat. This is the same reason why you can have stainless cookware, with a stainless handle on a pot, and the pot can be extremely hot and yet, you can hold the handle without a glove. Again, 2 factors at work here, the shape of that machined tube is made extra thin walls to limit cross section of the tube and further increase the thermal resistance thus making a barrier to prevent larger amounts of heat from the smoking hot heater block and nozzle where the filament melts, from conducting up the mechanical connection of the heater block to the rest of the extruder system.

Why is this so important and why are you having jams and why is it called heat creep?
Again, basic thermal rules apply here. The tube slows down conduction but unlike the name implies, it cannot prevent ALL heat transfer. Given a long enough time and no cooling, the 2 sides of the thermal barrier would eventually reach the same temp as the heater block. That's basic physics. What keeps the cold side cold is that a heatsink, cooling bar, and the large contact of the threaded cold end section of the tube mate together thermally making good contact so any heat rising up the tube is conducted to the cooling bar, slightly raising it's temp, that in turn tries to heat the heatsink, but because air is blowing past the fins and is cooler, it transfers this heat to the air. So what happens is, from a start of a print, the entire system is cold at room temp. When the heater kicks in, the block and nozzle get hot, as does the hot side of the thermal barrier into the heater block. Over time during the start of the print, heat slowly conducts up the tube raising the temp of the tube, and that conducts to the cooling bar, and that tries to raise slightly the temp of the heatsink. In an ideal situation the cooling of the heatsink and thermal connection to the cooling bar the entire system stabilizes at a temp that is higher than room temp, but well below the softening temp of PLA. Remember- other factor here, PLA has 2 different critical temps (well actually most thermoplastics have them too) but the point is, PLA softens but does not "melt" at relatively low temps. It becomes like a cooked wet noodle. You are trying to feed that into the system and create pressure inside the nozzle. So again, what happens, is over time the heat creeps up the thermal barrier tube, raises the temp of the incoming filament path, and the filament softens and becomes sticky mess that cannot be pushed easily into the rest of the hotend where it is hot enough to melt and be liquid. That's the trick and specialty of PLA. It must be cold right up until it hits the actual melt zone directly inside the heater block and nozzle. If it is heated BEFORE this, it causes jams. Between motor heat, botched assembly of the heatsink to cooling bar thermal conduction contact, incorrect fan or airflow, seal chamber ambient temps, or poor conduction of the thermal barrier tube threads to the cooling bar threads, you will get PLA jams a few minutes after a print starts because again, cold parts that need to stay relatively cold are heating up over time after a print starts.

[WWC Tech Lab]

Jetguy- this is excellent, thank you. 

We've had the hot end apart so many times addressing thermocouple wire breaks and such, that we've probably disturbed the natural order of things. Next step, I think, is to dis- and re-assemble to make sure that everything is tight. We do have some aftermarket ceramic insulating tape installed that I'm pretty sure is thinner than stock, so I'll double that up as well. 

Any visual clues of heat creep on the thermal barrier tube itself, once I get that out?

Thanks again!

[TobyCWood]

So your retraction setting is kinda high and long, fast retractions exacerbates the problem. It's not always heat related, but that's the most typical. You get the same kind of effect when the filament goes a tiny bit over tolerance, melts too quickly even at a lower heat... and/or there's too much unneeded up and down movement.

Try setting retraction down to 1mm... make it slower too... I have my default on my Rep2s at 200 mm/s retraction speed.