Prusa Mendel extruding dots

I've had my reprap for nearly a month now, it worked great at first but gradually began making dots rather than lines, so there is something that has gradually gone worse, I just can't tell what. I've made a tiny video for this on my very own reprap (prusa mendel), in the description there are images on other similar prints. [www.youtube.com] - You don't have to watch it but it will help you understand my problem significantly.


Background: At first, when I just bought it, the heater was blocked inside so the guy I bought off fixed it for me by drilling in to it from the bottom with the same size the hole had (0.35mm). After that it ran smooth, later it stopped responding to my computer and today I got it fixed again, this "dotting" thingy began for nearly a week ago or so and has gotten alot worse..

And the guy who I bought this off from is on vacation for a month so I can't get any help from him.

So perhaps there is another user here who has had some similar problem who can yell out a primal scream while aiming his point finger towards the screen and scream "THE SOLUTION IS....." (while actually saying the solution rather than dots.... I hate dots)

Edited 1 time(s). Last edit at 07/19/2012 05:45PM by injisera.

That was the sigh of dispair in that video. I have uttered it many times.

What are you printing? PLA or ABS?

What is that sludge you're printing on?

What temperature is your hotend set to?

Are you sure the hotend thermistor is OK? it looks strangely discolored in the video. Check the resistance - it should be around 90K at room temp (assuming EPCOS 100K)

Is the filament stripping? I'm assuming it does.

Can you feed filament through by hand with the hotend at temp?

What electronics are you using?

What software?

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

What are you printing? PLA or ABS?
>ABS


What is that sludge you're printing on?
>"ABS Juice", acetone mixed with some ABS


What temperature is your hotend set to?
>235 degrees celsius

Are you sure the hotend thermistor is OK? it looks strangely discolored in the video. Check the resistance - it should be around 90K at room temp (assuming EPCOS 100K)
>The plastic is melting when it's at 235 degrees (according to the printrboard), and it has worked many times before.. so I strongly believe that the problem is not the thermistor. But I'll have a look at it once I get my hands on a multimeter.

Is the filament stripping? I'm assuming it does.
>I'm sorry, stripping, english is not my native language and translating this gets my rumbo dumbo.

Can you feed filament through by hand with the hotend at temp?
>Yes I can, I'll make a tiny video of that later.

What electronics are you using?
>"Printrboard
Rev. C (by JDS)
printrbot.com
Laine Walker-Avina 2012

What software?
>slic3r 0.8.2 and Pronter(Printer Interface, their logo looks like this smiley ;-P)

Edited 1 time(s). Last edit at 07/19/2012 06:12PM by injisera.

injisera Wrote:
-------------------------------------------------------
> What are you printing? PLA or ABS?
> >ABS
>
>

That's quite a bit of juice. Maybe a little less.

> What is that sludge you're printing on?
> >"ABS Juice", acetone mixed with some ABS
>
>

That sounds about right. Maybe try 240C.

> What temperature is your hotend set to?
> >235 degrees celsius
>
> Are you sure the hotend thermistor is OK? it looks
> strangely discolored in the video. Check the
> resistance - it should be around 90K at room temp
> (assuming EPCOS 100K)
> >The plastic is melting when it's at 235 degrees
> (according to the printrboard), and it has worked
> many times before.. so I strongly believe that the
> problem is not the thermistor. But I'll have a
> look at it once I get my hands on a multimeter.
>

Is the hobbed bolt chewing a divot into the filament so it no longer extrudes?


> Is the filament stripping? I'm assuming it does.
> >I'm sorry, stripping, english is not my native
> language and translating this gets my rumbo dumbo.
>
>

That is good. When the filament comes out what does it look like? straight down or on an angle? If the filament comes straight down you probably dont have a jam.
> Can you feed filament through by hand with the
> hotend at temp?
> >Yes I can, I'll make a tiny video of that later.
>
>

OK
> What electronics are you using?
> >"Printrboard
> Rev. C (by JDS)
> printrbot.com
> Laine Walker-Avina 2012
>

Good
> What software?
> >slic3r 0.8.2 and Pronter(Printer Interface, their
> logo looks like this smiley ;-P)

When you send the next video, make sure you show the extruder and the hotend. It looks like you have a J-head. Is this right?

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

Allright, less juice!

240C, will use this moar!

No it is not chewing a hole in the plastic (the feeder), no "divot"

Yes, I do indeed have a J-head.


[www.3dprintershop.se]
This is my 3dprinter.. he took a picture of mine and put it up as the official "good-looking" printer.. hehe.

Printrboard RevC
Extruder Jhead 0.35 mm nozzle
Heatbed
LM8UU

[www.youtube.com]


aaaand


[www.youtube.com]

Edited 1 time(s). Last edit at 07/19/2012 08:45PM by injisera.

OK. So the nozzle isn't blocked and the temp is good. If it had a blockage the filament would shoot out a very weird angle.

How tightly are you clamping the extruder shut? I'm assuming that when you extrude (using pronterface) into free air you have no problem.

How does the hobbed bolt look? No excess plastic clogging the teeth?

Also, how tightly is that nozzle into the PEEK barrel? It needs to be fairly tight. Just looks like a lot of exposed thread in the video.

vs.


Upping the extrusion multiplier and getting better results may be pointing to a calibration problem. When was the last time you calibrated the extruder?

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

If you look at the object, it is full of holes, so there simply isn't enough plastic extruded for some reason. When the flow rate is much too low the filament gets stretched so much it snaps and forms blobs instead of lines.

---

[www.hydraraptor.blogspot.com]

BTW it is much easier to debug such problems when using colored filament: Black, White, Yellow, Orange - anything that shows up well against the color of your bed (red in your case). Clear (Natural) ABS/PLA is the worst for debugging such problems because it is very difficult to see (especially the first layer which is the MOST important layer).

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

@rhmorrison
Allright, I will use my white ABS

@nophead
"much too low" - ahaa, good to know that I'm way off then

@akhlut
I've never calibrated it, the guy who I bought off did all that for me, I'm not a total "retard" or anything, but I payed full price and everything so I wouldn't get problems like these.. So if there's some soldering / programming that needs to be done then I can handle that, I did get the "firmware" that was being used on the board, and I did get the arduino program so I'm capable of changing some stuff, it's merlin btw.

I'm fairly sure why it's not working, I was looking over some blog posts of some people who had built repraps, to see their mistakes, and I noticed a guy having a problem that looked alot like my problem. So I read about [blog.lincomatic.com] and I realized that there was supposed to be a screw somewhere, and this all explains why it has gradually gotten worse of the course of a month or week or every time I used it. Because at first it was tight, it sat great, but then it got warmed up every time I used it, which in turn made it soft, it's shape got more loosened because it was in constant motion, drifting off, and finally, the circumference of the hole got bigger and bigger, the grip got worse and worse until finally, this bad. And it also explains why it's bad at quick movements on the extruder, because while doing the outer circle, it first stops to move there, then it begins, and it slowly gains it's full speed, and until it does that it drops dots.. and then a fine line!... it's all so.. clear now.. because it also explains why the plastic is dripping between movements when it's rotating backwards. I atleast hope that it's that screw that is the problem...

So what I'm in need of now is some information regarding a screw (and probably a bolt) of where I can buy that thing that should go into the hole for the.... ah whatever.. I'm redicilously bad at explaining, even worse in english. So here's a tiny video showing EXACTLY what I need. [www.youtube.com]

You need an M3 by 6mm grub screw. http://www.orbitalfasteners.co.uk/en/products/m3x6-socket-set-screws-cup-point-stainless-steel-a2-304-

---

[www.hydraraptor.blogspot.com]

After putting some lipstick to the tip of the rotating part of the stepper and on the gear it's holding. And later putting it to work, I did hope to see the gear / stepper to drift off from each other. But sadly the lipstick was intact, this means that the gear got great grip and it also means that I was wrong, this also means that a screw won't fix the problem, it also means that I don't know what the problem is, except for the problem being close to the extruder.

[www.youtube.com] Video showing how it rotates correctly while having the lipstick intact.

To all you weird guys: The lipstick is not mine, it's my girlfriends.

So thanks but no thanks nophead :/

When you tell pronterface to extrude can you easily stop the large gear with your hand and watch the stepper shaft spin inside the small gear?

So that isn't a problem, but toss a grub screw in there anyway - it may save you a headache down the line.

Edited 1 time(s). Last edit at 07/20/2012 03:11PM by akhlut.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

OK, take a marker and measure 100mm of filament from the top of your extruder and mark that length on the filament. Then tell pronterface to extrude 100mm. Does the mark make it to the top of the extruder? If not then you need to calibrate it.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

when I press extrude while holding the big gear, then the whole machine rather jumps a step backwards(one tooth-gear back) instead of the shaft from the stepper rotating in there. So it sure does have -great- grip on that little gear.

I'll do the marking now!

When I asked for 100mm on the extruder, I had 84.8 left roughly


So to see if I made all the calculations correct, I marked 10cm from the top of the feeder, like the last filament I could see, 10cm from there, I marked, I pushed the 100mm extrude button, left I had 8.48 cm. mm.. cm... confuse the shit out of me even if it's just one 0 here and one 0 there.


I pushed the button again knowing it would only take 1.52cm (probably). And I held my finger on the filament, close to the gear, and I felt this "bump".. it wasn't linearly flow.. it was... like a wave... slow... slower... stop... slow.. faster.... normal.... slower... stop..... the big gear moved normal all the time though... so the problem... MUST BE! INSIDE! THE CHEWING THING! IT! IT HAS TO BE THERE! I KNOW IT!

When I look at the filament that has been chewed I see this pattern..


___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨___...---***¨¨¨

it's not just.. steps.. it's like... waves..

Edited 2 time(s). Last edit at 07/20/2012 03:45PM by injisera.

OK, get a couple of 13mm wrenches and pull the big gear off the machine. That bolt that holds the gear on should have a section of grooves on it. Take a video of the bolt and the opened up extruder post it. It sounds like something is definitely wrong either inside the extruder or in the firmware, or both.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

[www.youtube.com]

There's one... kind of.. flat "side" on the extruding-screw... but I'll put it back and see how many revolutions the big gear will do and I'll calculate the distance the chewing would've "chewed" to see if it's even near 100mm or not.


4.55 revolutions (roughly)... the chewer's diameter is roughly 6mm. So the circumference is then roughly 6*3.14159265 = 18.84, this times 4.55 should be close to 100mm. It is 85.76 mm. So either I measured the chewer's diameter wrong (I didn't get to touch it with the measuring tool.. it was mostly eye-sight-then-measure)

(100/4.55)/3.14159265 = chewer's diameter, 6,99 mm... 7mm. I was 1 mm off I guess .


So... then... by noticing how the "wave" pattern was acting... and how the chewing thingy is actually chewing the correct value.. then... there's either crap in the extruder-nozzle... or there's something blocking the way down there... great...

but either way, it's nowhere close to 15.2mm as it only chewed when I ordered it to chew 100mm.

Edited 2 time(s). Last edit at 07/20/2012 04:17PM by injisera.

I have had blockage inside of the J-extruder head, the "hot-end". But that blockage was near the nozzle... I now believe I have a blockage inside of the hot-end but.. a little bit further up... high enough so the flow goes straight down from the nozzle.. but enough to make the chewing just chew chew chew... I ended up with a divot...

Disclaimer, I haven't even finished the frame on my mendelmax yet so I have ZERO experience with reprap printing. But I am a mechanical engineer so that's gotta count for something, right?

if you're seeing a wave pattern, then it sounds like something (hobbed bolt?) is not concentric. Either the mounting was loose and there's a set screw or something forcing it off center or the bolt has been bent and it's camming as it comes around (bites filament too much at one end of the cycle and lets go entirely one half cycle from that and chewing in between the extremes as it regains purchase). Try taking out the hobbed bolt and rolling it on a flat surface like the edge of the table (with the head off the table, you're checking the shaft for straightness). The amount of light coming under the screw should be uniform. Alternatively, you could put some grease on it and roll it on a piece of paper and see if it leaves a solid smear or dashes. OR if you have the equipment, chuck it up in a drill press/lathe and put an indicator on it while you slowly rotate it.

no idea how the hobbed bolt would get bent over time, maybe will after I start playing with my extruder...sounds like a euphemism

The inner diameter of the hobbed bolt that engages and drives the filament doesn't really matter. What really matters is that the extruder needs to be calibrated. This is a bit of a process. Check this and this. That should help. You need the arduino IDE to get into your board. The version of the IDE depends on the board that you have.

From there it's fairly straightforward.

Also, do you have a pair of digital calipers? If not don't bother until you do - they are a necessity.

To take the J-head apart you need to get a little bit of heat into it. Set it for PLA and wait for it to get to around 80C then turn it off. Let it soak for a few minutes and then disassemble.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

My point of the post was that my guess was fairly close to the correct value, 6mm -> 7mm, and also that it did rotate the correct rotation = the calibration is not needed, when it did rotate the filament got through the J-head, but... alot less than predicted, so there has to be a blocking inside of the J-head, and I'm going to get a 0.35mm drill to fix it.

I'll leave an update on how it goes to help other people with similar problem.

just to add my 2 cents on that video with the lipstick. When you move your extruder that slow only one way it has probably going right. Problem with printing is that you are loading unloading filament fast and much. so probably during print it will come of some time, the grubscrew is pretty abligated in this design....

also that looks like not the main reason like others tell you aswell. The hobbed bolt is that pretty flat or does it has a nice hole were the filament supposed to go. i heared some storys about filament wondering of from the path of the hobbed bolt. i have made a centring slot in my bearing to let it stay at its place.

looks a bit like this (sorry no photo)
My bearing in the extruder idler
________ ________
| | | |
) | | (
|________| |_______|

lines make it difficult to draw

If you can't extrude it down, try to retract for X-cm and see check if X-cm comes out ? At least you can get the Steps per mm correct. It's something hard to calculate - the chewing action and bolt diameter measurement together make the math difficult.

It turned out there was some "crap" inside of the J-head, I cleaned it out with a drill (0.35mm), but the difficult part was to take it off while it was still warm... but I managed.. and shorted my power supply unit... I bought a new one, and now I'm having this redicilously difficult problem with putting it back in the PEEK plastic. I did put it back.. when the J-head was at 80 degrees C, But I couldn't get it far enough..

But I did manage to print 1 layer correctly.. then the other 4 layers got F:d up because the nozzle got bent and pushed everything downwards which in turns made the stepper motors on the X / Y mode to lose some steps which in turn made the model shift to the side.. it looked nasty...

Gah... this is so frustrating...

The images are quite self explainatory... I used force.. too much of it... then I put it together with 2x nails, works like a charm!