extruder not enough power to push the filament through?

[Greg MacDonald]

Makerbot 2x. Printing w ABS and normal medium settings.

Summarizing a long and frustrating problem. 

Problem: prints fail after about 30 mins to 90.    (btw, I'm trying to run fairly large shapes (+/-6.5"x3"x3"high)


Symptoms During Print:  could see inconsistent flow after a while and then hear a periodic clicking sound. Then stops extruding all together with a click - click - click (+/-100 Bpm for the musicians out there).

Seemed like: .... it was a clog, but i don't think so.  i could easily push filament through the heated nozzle. it came out clean and straight.

What I've done: (beyond unload, clear, reload, retry.....)  I took the head apart and put filament into the heated nozzle by hand and it passed right through.  I lifted the motor off to the side and ran the load sequence, it takes the filament, but if I apply even the slightest back pressure, the motor clicks.   I adjusted the tension within reason and it changed a little, but not dramatically.  But tight or loose, any back pressure and the motor stops turning and produces a click........and repeats as long as there is any kind of back pressure.  This has happened on both sides I believe, though I've not done this same thing on the left side. ( I have had similar failures on that side).

BUT!:  When I shut the thing off for a while, power back up, then feed the filament it sucks it through like a champ.

Is it possilbe it's a bad motor?  (or pair or motors)
Is it possible that when the motors heat up they lose their strength?
Is it a controller problem?  (sure seems temperature sensitive though)

If anyone can solve this for me, I'll have my daughter make you cookies. 

signed....."Printless in Seattle"

[Scottbee]

There are some design and manufacturing issues inside of the extruder on the 2X.  For many if not most, they won't manifest themselves in such a way as to cause a failure (when the bot is new), but over time they can cause some problems.

While there are many things that may be causing this, my immediate suspicion is that you're getting some "heat creep" inside of the extruder that is causing the pinch roller to bind and lock onto the filament.  The first thing to try is to take the extruder apart and check the shoulder screw that mounts the pinch roller.  The nut is the "wrong size" for the molded-in capturing feature, and on my machines they didn't use any Loctitie.  Carefully remove the screw, apply some blue Loctite to the threads and reassemble the 5 pieces (load arm, idler arm, bearing, shoulder screw, nut). Make sure the nut doesn't spin, and don't tighten it too much or the idler arm will crack.

That may resolve your problem.

If you can't follow my instructions, let me know.  I think I have some pictures of the internals.

I like oatmeal raisin and chocolate chip....  ;)

[Eighty]

I noted this same problem on my Rep2 a long while back (maybe a year ago). Never did put my finger on it. The issue still appears on long prints (>8 hours). I can usually squeak through it by placing a fan on top, blowing air across the gantry. I think it has to do with excess heat, but not sure exactly what.
The symptoms usually start with a little clicking from the extruder stepper. I seems like low power to the stepper motor, but again, I never have found the root cause.
One lingering theory is that perhaps the contacts at the motor connector loosen up (due to the heat), and result in loss of amperage. Dunno.

[Ryan Carlyle]

This is an interesting issue because the stepper is clicking later in the prints, which means it's NOT an issue with filament softening. If it clicks on the first layer then you've trammed the build plate too close to the nozzle. If it clicks during loading then you have a nozzle clog. If it clicks while printing into a section that is warping upwards, that's obviously the curled plastic blocking the nozzle. But if you can push filament through easily by hand, and the stepper clicks with minimal back-force while loading, that's indicative of a problem with the extruder or the stepper.

Steppers DO produce less force when they get hot. This could explain why the print starts ok but fails after a while. But it shouldn't be getting that hot. I've even run my R2x with the chamber above the rated ambient temp of the steppers and it works fine.

So either the stepper is producing less torque than it should, or the extruder is needing more force to push filament than it should.

Does the other extruder work? That's the first thing to check. And try different filament to see if that makes any difference.

If everything on the machine is 100% intact, then there are a couple simple things you should check before taking anything apart:

• Check for an accumulation of filament dust or junk in the gear wheel.There is a little window on the side of the extruder that you can look in. Blow it out with some canned air if necessary. 
• Check for dulled drive gear teeth. The teeth should look sharp.
• Check the set screw on the extruder. If it is tightened down too far, then the stepper is working hard to flatten the filament and that robs your pushing power. Do a load/unload cycle, and look at the filament that comes back. It should have distinct bite-marks from the gear teeth but NOT be significantly flattened. I run mine with a very small amount of flattening on the side opposite the teeth marks. Loosen or tighten the extruder set screw on the top of the extruder as needed.
  

Then there are possible hardware issues:

• Stepper failure -- this would be very unusual. Try swapping steppers. 
• Extruder mechanical problems like Scott B mentioned. Does the idler bearing spin properly? Note that the spring in the extruder can get compression-set if you use the lever often. There's really no reason to ever use the lever.
• Bad/loose wiring connection to the stepper motor. Wiggle the connectors to the steppers and the motherboard. Look for any signs of burning/melting.
• Failed stepper driver (botstep). This would probably only affect one side. Try swapping all the wiring connections at the motherboard.
• Power supply issues sending insufficient power to the steppers. This would probably also cause issues when all three heaters run at the same time.
• Exotic stuff beyond my knowledge, like digipot settings. 

[Joseph Chiu]

It can also be the case where, for solid parts, you have been slightly over extruding, just barely, but the excess eventually reached a point where the gap beneath the nozzle has gone to zero.

--
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.
For more options, visit https://groups.google.com/d/optout.

[Jetguy]

Couple of comments.

If an extruder clicks, that means it's skipping steps which contrary to belief is a good thing when compared to the alternative, the filament drive slipping  and grinding the filament. This also debunks the heat theory and the fan cooling other were talking about in the filament drive area.

I have actually seen a real world test where stepper motors actually did get stronger with higher temps. The permanent magnets inside DO have an optimum temperature. I'm trying to find that article but the testing was rather thorough.

Now, this goes back to OTHER things I know about the system. I don't trust the digipot system and there is enough proof that many times here in the this forum, there are reports of incorrect current values being sent. We also know the highly debated act of printing over USB can and does have the chance of a missed command. The complaint here is that mid print, suddenly the extruder motor lost torque after a pause or some other event. Well, if you know HOW the system works, there are many ties for power management that there are writes to the digipots to lower current. If that write to increase the current back from a pause mode is missed or garbled, the digipot remains in a lower current mode and the stepper drive skips steps.

many of these reports say that power cycling the bot, which BTW writes the values as part of the start sequence to the digipots back to the default current levels.

So ding, ding, I think the digipots are the hidden answer to these weird but sometimes repeatable sequences.

[Dan Newman]

Well, then an interesting detail to know is whether or not the bot is
running Sailfish since Sailfish reads back the setting from the d'pot
to ensure that it was set correctly. Mind you, when the command is
lost altogether (as MW will do over USB), then there's nothing the firmware
can do: it isn't a mind reader and cannot execute commands it does not
receive.

Dan

[Scottbee]

You guys have all made very good points... and I hate to add variables.. but I can't help myself:

As I mentioned, the extruder on the R2X has some dubious design qualities.. it can work well, but it can also bind up and screw the pooch.  I'm working on a complete tutorial on how to "file away" the errors that were made in the design and manufacturing of those plastic pieces.

Here's another one that folks don't often think about, but I had it happen recently and it left me scratching my head until I figured it out:  If there is a bind or lack of lubrication on your Z-axis rods you can also end up with an extruder problem.  The system assumes that the Z-axis moves down one layer thickness every index.. and that it is "weight biased"... no backlash and always following the command.  If you have a bind, due to slop in the lead screw, the stepper motor will turn properly but the platform will not actually move down.  So it will try to lay the new layer down directly on top of the previous layer with no Z-gap.  If your extruder is set up properly this can and will cause hydraulic lock and stepper stall (clicks).  

I'm not saying this is your specific problem, but if you've recently made changes to your gantry (aluminum arms or wraps) or have dry Z-axis rods, it's something to think about and look into.  It's something that may occur later in the build, and not while Z is close to "0" (due to bowing of the Z-axis rods).

[Christopher Doll]

Thank you for the thoughts - I've been assisting Greg throughout this process and will draw up a list of items to check as we proceed.

- so far Makerware is the only interface we've used to generate projects, and we've mostly used the canned settings (modifying things like temperature and speed here and there based on previous assistance and suggestions from Makerbot support). I've briefly looked at the deeper settings available via the ReplicatorG utility, but haven't used it yet.

We will take a closer look at the extruder assemblies, and I may find a way to mount an additional fan above the motor. The buildup of heat is quite suspicious.

We found a problem with the left filament guide assembly - the screw holding the capstan in place was driven in way too hard, slightly cracking the arm. Makerbot is replacing that piece. The right side didn't appear to be affected in any way, although I will be revisiting these entirely. I've tensioned the holder several times, suspecting this to be a possible area of concern, but never beyond what would appear to be "too tight".

Our builds have curled, and somewhat interfered with the extruder as it passes over lifted edges - however, this hasn't prevented the 2x from proceeding (the failure seems to be a separate event, but I'm not certain). Curiously, the two prints that mostly finished (I'd hesitate to call them successful), proceeded after a curling event.

Thank you everyone!

Christopher Doll

[Eighty]

Dan,
In my case, I've been running Sailfish since the day you gave me a Beta version to test (1/26/13 to be exact). So I'm thinking it might be either (a) a firmware problem in both MBI and Sailfish firmwares, or (b) a physical issue that is independent of firmwares used. Option (b) being more likely.

For me, no pause is necessary for the issue to arise. Simply an extended period of printing.

[andre]

" I lifted the motor off to the side and ran the load sequence, it takes the filament, but if I apply even the slightest back pressure, the motor clicks"

That right there isn't right.  I can literally punch a hole in my finger with the filament doing what you did.  My extruder motors have enough torque where I can't keep my finger pressed on the bottom of the spring feeding assembly without feeling some pain  :)

I suggest swapping around some components.  Swap the botstep with any other...  Swap the motor connectors with the other extruder.  Print from the SD card.

[Greg MacDonald]

Thanks too all for the help.  Amazing help.

couple of comments in reply. 

BTW, running on the card, not USB.

We have been having trouble with both sides similarly.  I like the idea of swapping sides, but I think it will yield the same results.  the back pressure sufficient to stop the gear (and make it click) is really low.  (w/out having a hand way to measure, its about the force needed to pick up a tennis ball if you had a shoe lace tied around it......if that even. )

I can see the "d" section/notch (flat side) of the axle on the motor jumping so can rule out that the gear might not be securely fastened. Checked it anyway....tight.

Tightening the tension screw has little or no effect on the on the clicking.

We've had this unit for only 3 weeks. The left side extruder was doing this first and when we opened it the first day, I saw that the surrounding plastic housing on the nut was already cracked. We abandoned that side in favor of the right because we were intent on getting a print out.  But both sides have been behaving the same way and I suspect the cracked block is just a coincidence. At one point we thought the unit may have been dropped in shipping because a number of things were kind of out of alignment, like the Z-axis and the door of all things. We'd realigned the z-axis gear in the back to the best of our ability and it doesn't seem to be affect this other problem.

have now completed taking the left head apart and checking the alignment, and strength of the pull.  I couldn't reasonably hold the filanment back it pulled so hard. (nothing like what I was reporting). Tearing back into right side to check. But it appears that a reboot has re-set whatever was the problem.  (maybe the heat, and/or the control may be the narrowed area of concern). 

Was thinking of loading matching ABS into both sides, slicing my model into layers and having it alternate. ......may be a better idea than reality.  We'll see.

thx again.  Either way, I will keep reporting on this.  I had a a previous unit (Rep2) and we had identical issues, but hadn't discovered this until replacing it with this 2x.

GREG

On Friday, March 21, 2014 7:05:16 PM UTC-7, Greg MacDonald wrote:

[Jetguy]

Man, that is telltale if I ever heard it.

We have a discussion going on over in the Jetty group (Sailfish) on a slightly different problem but the symptoms are spot on.

We know the digipots that set the current for the stepper drivers default to mid value. We also know the intended set value is supposed to be 118 (on a 0-127 scale)

We know from the datasheet of the digipots that they default to mid value in the case of brownout or other events as well.

We know the firmware writes the values to the digipots on startup.

We know the default start.gcode lowers the values to the digipots during heating so as to not overload the PSU.

We know that when printing over USB and maybe some other glitches, commands can be skipped and never executed during a print

 

So if you have a proven situation where you started printing and then found the motors weak clearly indicating reduced current, then a power cycle of the bot happens and the problem goes away, I don't see how that cannot be the digipots. It reads liek a textbook answer.

Again, either the vlaues to raise the digipots back to the normal current is being missed or interfered with, or the digipots are experiencing a glitch reseting them to the default mid value.

It's just even more proof a power cycle sets everything "right".

[Jetguy]

Unfortunately, I don't know an easy "fix" for the problem that doesn't complicate things.

 

I've never thought the idea and implementation as they did it was right. There are very few good reasons to lower current setting and if it causes problems (and this sure looks like the root cause) then any advantage of having software controlled current settings is negated.

 

A simple change to the gcode, something I sent Makerbot well over a year ago, could maybe prevent this.

Simply heat, then home. It's not rocket science. The problem is the users expect something to happen when they hit print. If the machine just sat there warming up, there'd be complaints as to why it was waiting. Or, you could even compromise. Fake home it, disable the motors, preheat, home it for real, then print.

But here we are, well over a year later and the same old system bites yet another person.

 

 

 

On Monday, March 24, 2014 12:25:29 PM UTC-4, Greg MacDonald wrote:

[Scottbee]

The front door being out of alignment doesn't necessarily mean that the unit was dropped.  It just means that it was assembled by the same crack-head who built the two that I have......

[Dan Newman]

On 24/03/2014, 9:55 AM, Jetguy wrote:
> Unfortunately, I don't know an easy "fix" for the problem that doesn't
> complicate things.
>
> I've never thought the idea and implementation as they did it was right.
> There are very few good reasons to lower current setting and if it causes
> problems (and this sure looks like the root cause) then any advantage of
> having software controlled current settings is negated.
>
> A simple change to the gcode, something I sent Makerbot well over a year
> ago, could maybe prevent this.
> Simply heat, then home.

And that's what Simplify3D does, in part I suspect, because they listened
to you and Wingcommander.

Dan

[Greg MacDonald]

wait.....as a new guy to all of this.....Simplify3D is an aftermarket replacement for makerware in the case of the Makerbot printers?  I see it supports makerbot (including the 2x).

couple of questions. 

1.  With a big meeting looming tomorrow that I'd like to print for, can I get it, learn it, and print with it in a day? (cost is ok)
2.  Do you feel that the default better deals with the issue I've been wrestling with?  (assuming I don't have a lot of time for experimentation and discovery.


thx. GREG

[Dan Newman]

On 24/03/2014, 10:36 AM, Greg MacDonald wrote:
> wait.....as a new guy to all of this.....Simplify3D is an aftermarket
> replacement for makerware in the case of the Makerbot printers? I see it
> supports makerbot (including the 2x).
>
> couple of questions.
>
> 1. With a big meeting looming tomorrow that I'd like to print for, can I
> get it, learn it, and print with it in a day? (cost is ok)
> 2. Do you feel that the default better deals with the issue I've been
> wrestling with? (assuming I don't have a lot of time for experimentation
> and discovery.

Will S3D work better for your models? No clue. Different slicers have
their strengths and weaknesses and what matters to you may not matter
to me. Some of the folks here who use their printers commercially
tend to use several slicers and know what works best for different
types of models they work on.

Short answer is thus I have no idea if S3D will be an improvement or
not for you. I also do not own a 2X and as such have never used it
with a 2X. (Indeed, I've not tried using it for dualstrusion either,
not even on my R1 Dual.) I've only used it on my R2 w/HBP. I've
had good success and do like the slicer, but I've only used it for
about 3 - 4 weeks and that's not enough time with it for me to be
comfortable making recommendations.

As to learning curve, I would not recommend jumping on a new slicer
unless you have at least half a solid day to spend experimenting with
it first and printing calibration cubes and otherwise making basic
tweaks. Then a half day to try models akin to what you're working
with to see if you can get the results you want. And if the models
take hours to print, then that half day may stretch into a week if
you find yourself iterating on profile changes.

If all you need is to heat then home, then you're better off in the
short term figuring out how to change the start gcode for the slicer
you are currently using.

Dan

[Greg MacDonald]

yes......logical I guess.  Got a little excited there for a second.  shouldn't be experimenting too much 24 hours before something important. Though, not doing so may mean no model for the meeting, which I'd like to avoid. 

Since, rebooting the printer seems to reset things and since it doesn't usually fail till sometime during the print, I'm hoping that there is some time to print some of the model. if I can get it out in pieces we can glue it together, hit with a can of primer and get thru the day.  (the shape is basically an extrusion). 

ok, thanks, I may roll the dice anyway on S3D later on today. worth it if it solves. it.

Thanks, GREG

[Christopher Doll]

On first glance at both files (start.gcode and start-abs.gcode), they are now performing the HOME-LowerStepperCurrent-HEAT-RaiseStepperCurrent pattern:

In both files I'm seeing this:
(*** Set the homing position ***)
G92 X152 Y72 Z0 A0 B0 (Replicator Home Position)

G1 X-112 Y-73 Z150 F3300.0 (Waiting Position)
G130 X20 Y20 A20 B20 (Lower stepper Vrefs while heating)

.... then there's all the settings for heating the extruders/build plate followed by the very last line:

G130 X127 Y127 A127 B127 (Set Stepper motor Vref to defaults)

Since this is my first glance through the configuration files, could you take a peek and check me on this? It appears to be setting the current back on each of these profiles I've saved. Now if this is being ignored by the Makerbot 2x is another story altogether. I'm less clear about the path this code takes once it goes from the SD to the Makerbot.

Thank you!

[Scottbee]

My "canned" S3D R2X profiles do the same thing....

[Dan Newman]

On 24/03/2014, 12:14 PM, Christopher Doll wrote:
> On first glance at both files (start.gcode and start-abs.gcode), they are
> now performing the HOME-LowerStepperCurrent-HEAT-RaiseStepperCurrent
> pattern:

This thread is starting to get a little jumbled. "They" meaning who exactly?
MBI and MakerWare? The S3D slicer?

> In both files I'm seeing this:
> (*** Set the homing position ***)
> G92 X152 Y72 Z0 A0 B0 (Replicator Home Position)
>
> G1 X-112 Y-73 Z150 F3300.0 (Waiting Position)
> G130 X20 Y20 A20 B20 (Lower stepper Vrefs while heating)
>
> .... then there's all the settings for heating the extruders/build plate
> followed by the very last line:
>
> G130 X127 Y127 A127 B127 (Set Stepper motor Vref to defaults)
>
> Since this is my first glance through the configuration files, could you
> take a peek and check me on this? It appears to be setting the current back
> on each of these profiles I've saved. Now if this is being ignored by the
> Makerbot 2x is another story altogether. I'm less clear about the path this
> code takes once it goes from the SD to the Makerbot.

For S3D and a Rep 2 w/HBP,

01. G90 ; use absolute coordinates
02. M83 ; relative extrusion distances
03. G21 ; set units to millimeters
04. M127
05. M140 S100 T0
06. M6 T0
07. M104 S240 T0
08. ; **** Replicator 2 start.gcode ****
09. M103 ; extruder off
10. M73 P0 ; enable build progress
11. M70 P2 (Simplify3D Creator)
12. G162 X Y F3000 ; home XY maximum
13. G161 Z F1200 ; home Z minimum
14. G92 Z-5 ; set Z to -5
15. G1 Z0.0 ; move Z to 0
16. G161 Z F100 ; home Z slowly
17. M132 X Y Z A B ; load current position from EPROM
18. G1 X-145 Y-75 Z30 F9000 ; move to wait position off table
19. G130 X20 Y20 Z20 A20 B20 ; lower stepper Vrefs while heating
20. M70 P1 (priming nozzle...)
21. M6 T0 ; wait for extruder to warm up
22. G130 X127 Y127 Z40 A127 B127 ; default stepper Vrefs
23. G1 Z0.4 ; position nozzle
24. M108 R5.0 ; set extruder speed
25. M101 ; start extruder
26. G4 P3500 ; purge nozzle
27. G1 X-140 Y-70 Z0.15 F1200 ; slow wipe
28. M103 ; stop extruder
29. G1 X-135 Y-65 Z0.5 F1200 ; lift
30. M73 P1 ;@body (notify GPX body has started)
31. ; **** end of start.gcode ****

Line 5 starts the HBP heating while line 6 waits for it to come to temp.
Line 7 then starts the extruder heating. Only after line 7 does the
start gcode consider homing. (It should raise the VREF currents first
since it is assuming they are at operational level which they may not
be; for instance, bad VREF default values in EEPROM.) On line 19, it
does then drop the VREF currents while waiting for heatup to finish
for the extruder at line 21. But the majority of the waiting for
heatup was for the HBP and that was done before any homing at lines
12 - 16).

Dan

[Greg MacDonald]

Still looking at the last post and the code info and not the purpose of this post.

I discovered that on the original problem of losing power in the extruder motors, if turn the thing off and back on, (re-runnning the start up sequence in doing so) and run the prints straightaway, I was having more luck.  At least it was printing with a higher though not 100% success rate. 

Since then, I've re-uploaded the firmware in the 2x.  (first re-downloaded the Makerware software, then ran the upload from makerware).  the one test after printed to completion.  (though I should note that it was a much simpler and shallower build.)  I need to test it on a larger shape to see if it solved things.

I had also been running into some odd other errors including:
A read error on the SD card.
A 'detected errors and stopped the print' error (paraphrasing) that occured at about the 80% mark.
Trouble using a new SD card.

At the suggestion of Makerbot support, I re-upped the software and firmware.  So far......that seems to have made a big difference.

I will continue to test and see if this in fact solved it and will report. 
thanks, GREG

[Bonekollector]

S3D is worth looking at - but not for a job you're doing right away. It does a lot of things well out of the box & the supports it generates are superior to Makerware IMHO. It also gives you access to a lot of settings that you can tweak. I have a Rep2, so I dont know about the 2x, but I did have to fine tune the settings a bit to get the results I was looking for (primarily increasing the extrusion multiplier as well as lowering the print speed). But I really like the latest version a lot. Its ability to apply multiple processes to the same model is crazy cool.

[esg6589]

Just a quick word in regard to sd card errors, several of my prints died mid way through citing this as the issue until I stopped using the included sd card on my 2x. Something tells me there is more to the original problem, but that is a simple thing to eliminate right off the bat.