EZ Extruder skips

[geneb]

How does the resistance feel when you're manually pushing filament from the extruder end of the bowden tube? You may have hit a case where there's a bit of a narrow spot in the bowden tube, but that would be pretty surprising.

g.

[osr]

There's significantly more force required to push filament through the tube than to just push it through the extruder, but it's not hard to do. When the printhead is up high and the tube has a loop in it, there seems to be more friction. Should I lubricate the inside of the tube? What should I use?

[626Pilot]

Here's what I would do. COMPLETELY disassemble the hot end, including unscrewing the PEEK from everything else and carefully inspecting it and the PTFE liners. You may find a thin layer of plastic hanging around on the outside of the inner or outer liners (or both.) You can remove that pretty easy as it will be thin and won't adhere well to the PTFE. (Yay for Teflon, I guess.) Make sure that the ends of the liner tubes are completely flat, of equal length, exactly long enough to reach from the top of the inside of the hot end to the nozzle when screwed in, and that there are no burrs. Inspect the inside of the metal and PEEK components and ensure no plastic is hanging out in there either. What happens is that this plastic will swell and increase the feed pressure, and then the cussing starts.

IF:

• The inside of the hot end and both PTFE liners are clean and free of burrs;
• The hot end is assembled tightly, with no room for plastic to get out;
• The nozzle is not clogged and free of burrs;
• The temperature is correct (might want to verify with a thermocouple, since the thermistor can't be sworn in);
• The PEEK section is cooled with a fan and shroud;
• The fan shroud is insulated w/ Kapton tape to prevent it from leaking air down onto the hot end;
• Your PID loop is tuned well*;
• Your extruder steps/mm is calibrated correctly; and
• Your heat is correct for the filament...

...then you should be good to go.

The asterisk for the PID loop is because I don't use the SeeMe hot end anymore. Someone who does can tell you where the duty cycle (% of time turned on) should be in Repetier. That's the lower of the two curves in the Temperature Curve tab. I have an E3D hot end, and when I've had some issue the duty cycle has often been 50-80%. The usual duty cycle is 25-35%. When it shoots up too far above normal, it means something's wrong and you're likely to have critical melt creep sooner or later. The automated PID tuning you get with Repetier is okay but not that great. It produces a duty cycle that looks like an earthquake on a seizmograph. For even heating, it should look more like the profile of a gentle mountain range (see attachment).

I got a lot better at PID tuning after I read this: http://www.pcbheaven.com/wikipages/PID_Theory/ (way more useful than the Wikipedia article)

Repetier also has a "dead time" heat manager that you can elect by changing the heat manager in EEPROM from 1 to 3. You have to be using a newer version to get that feature. Information is here. I don't like it as much as PID tuning beacuse the duty cycle curve can be made a lot flatter with PID.

If all that doesn't work, I recommend upgrading to an E3D all-metal hot end. It can withstand much higher temperatures and has fewer things to go wrong with it. There's a huge thread here if you're interested.

[foshon]

IMHO, the wades in any variation is far from a better solution. First, the drive gears are printed. Any flaw in the eccentricity of the printed gears is transferred to your print. Second, these gears wear; meaning the first revolution after you set them your settings have changed, a very tiny amount sure but they have changed. I have a fully functioning Wades, I even used it on my MAX at one point. I keep it just in case I need to print an emergency body for my micro.

[626Pilot]

Is the Wade's extruder the one they were shipping before the EZStruder? The gears on mine look injection molded or something similar.

[osr]

I spent some more time trying to get my Rostock max to work properly today. I tried most of your suggestions, 626pilot. I wish I could say I found a chunk of plastic in the hot end or something, but there wasn't anything obvious. I replaced the teflon tube again, carefully shaving away slices until the pneumatic fitting just touched it as the o-ring compressed. I could still see through it, but the inside of my 0.5mm nozzle looked a little dirty, so I installed a new 0.7mm nozzle. I don't have a thermocouple, but the temp couldn't be too far off could it? I've tried everything between 190c and 220c, so even if it was a few degrees off I must have been at the perfect temp at some point. I have a peek fan and shroud already, but I used the tape as you suggested. My temp curve looks almost as nice as yours, never varying more than a degree. I also cleaned my extruder gear again. I printed an object at 30mm/s (infill, perimeters, etc...) and I would say that the skipping went from constant to every so often. I'm not sure which of the things I did caused the improvement, but it wasn't enough. Increasing the speed to 40mm/s gave me constant skipping again. Squeezing the bearing against the filament by hand puts a stop to it, and I held it that way for a few minutes to see if the bowden tube would pull out of its fitting. It didn't. I can't say that my hot end is set up perfectly, but there doesn't seem to be anything wrong with it. I can easily feed filament by hand, and if I pause the print and try again when the skipping is at its worst, there doesn't seem to be any extra resistance in the hot end. Maybe the teeth on the extruder gear are too sharp. Maybe I'm expecting more speed than it can handle. It doesn't really matter, because I'm not interested in fighting with the EZ extruder any longer. I'm going to have to replace it with something that will securely clamp down on the filament, but I'm not sure what. The easiest thing I can think of is the airtripper design, since I can mount it to the existing bracket. I might try that first. The Greg's Wade reloaded is what I have on my Prusas, and it works very well. If I hold the spool above the printer so it can't turn, it has a good enough grip to lift the entire printer off the table and climb the filament. This is the extruder I'm most familiar with, but I'd love to hear everyone's recommendations.

IMHO, the wades in any variation is far from a better solution. First, the drive gears are printed. Any flaw in the eccentricity of the printed gears is transferred to your print. Second, these gears wear; meaning the first revolution after you set them your settings have changed, a very tiny amount sure but they have changed.

I would agree with you if my EZ extruder worked the way it's supposed to, but I'd rather have any other extruder at this point. You're right about the printed gears. They do have to be perfectly flat and round to get a nice smooth extrusion, but that's not too difficult a thing to do when you have a 3D printer. Backlash can be a problem too, but I haven't seen much since I switched to herringbone gears. I haven't noticed any wear on my PLA gears. Are yours ABS? I wish SeemeCNC would injection mold some. Are you using the EZ extruder on your Rostock? How fast can it go?

Is the Wade's extruder the one they were shipping before the EZStruder? The gears on mine look injection molded or something similar.

The old SeemeCNC design was called Steve's extruder and it had 6 gears on it. I would guess that it had a lot of play in all those gears, but I would rather have one of those than this EZ extruder. At least it looks like there was a way to tighten down the filament.

I have a fully functioning Wades, I even used it on my MAX at one point. I keep it just in case I need to print an emergency body for my micro.

What's a micro?

[626Pilot]

You can't feed filament through a 0.7mm nozzle with an EZ-Struder? That just doesn't sound right at all. I can push 0.25mm with mine. 0.7mm should be ridiculously easy to pump. What are your motor current and retraction settings? If you try feeding the filament through by hand RIGHT after it starts skipping, is it easy or does it fight? If you unscrew the nozzle, heat up the block, and push filament through, is there much resistance? How much resistance do you get just pushing filament by hand through the Bowden tube by itself, unconnected to the hot end?

[osr]

The 0.7mm is working much better than the 0.5mm. I can print at 30mm/s now, with just a few random skips, and now that I've lowered the current, they aren't as noticeable as before. I am starting to suspect that the diameter of my 0.5mm nozzle might be off, because the print quality is much better with the 0.7mm nozzle. The beads melt into each other now, but before they were just next to each other. I entered a lower than actual filament diameter to compensate for it before, but it's no longer necessary with the 0.7mm nozzle. After I've cleaned the 0.5mm nozzle, I'll measure the filament extruded from it. It's soaking in alcohol now. I've never turned on the hot end without the nozzle attached, but I slid a piece of filament through while I was assembling it, and there wasn't any friction. I did check my bowden tube for resistance before, and it seemed to catch slightly where the pnuematic fittings had dug into it, so I cut the tips off of both ends. Now it slides right through. I still can't go much faster than 30mm/s without constant skipping though. I think I was just unlucky and got a bad gear or a weak spring or something. How fast can you print with your EZ-struder?

I went ahead and modified the airtripper design to bolt to the EZ-studer mount. Actually, I started with the BI version since I only have 608 bearings on hand. Here's the link to that mod.

http://bootsindustries.com/portfolio-it ... i-edition/

Here's mine.

And here's the stl, in case anyone wants to try it out.

airtripper_mod_2.stl

(146.37 KiB) Downloaded 372 times

I didn't change the bearing holder at all. I am printing it now, and I'm hoping it will let me go faster than 30mm/s.

[cope413]

Something else is wrong. It's not the extruder. It really seems like your temp isn't right based on the symptoms you've described. You've gone through all the hot end calibration and pid auto tuning? What is your extruder steps/mm setting in EEPROM?

[osr]

I don't think my pid loop is perfectly calibrated yet, but I don't think the temperature has had anything to do with the skipping. It stays very close to the target temp during printing, and I have tried printing at many different temps. I ran the pid autotune again just to see if it would improve, and got the seismograph style duty cycle that 626pilot mentioned and slightly less accuracy. I reverted to my previous settings that stay within a degree of the target with a smoother duty cycle, but I don't think they are the optimum either. As the hotend is warming up before a print, the temp rises quickly to a few degrees below the target and then very slowly rises the last bit. It's much faster to manually set the temp above my target and then start the print. I'll probably tune it again soon using the method described in 626pilot's link. That's a really good write up. http://www.pcbheaven.com/wikipages/PID_Theory/ I also double checked my extrusion steps. 50mm of extrusion measures 50mm on the ruler. My steps/mm is set to 92.4. As I was reading about pid tuning, I came across this, and it made me think that my skipping is not out of the ordinary.

From http://reprap.org/wiki/Rostock_MAX

.2mm layer height is a good general thickness for all around printing. If you slice at thicker layer heights, you may need to slow down, to around 30-35mm/s for .3 and slower for thicker than .3 layer heights. This is mainly due to there being a lot more filament being extruded at that layer height.

I usually print at a 0.4mm layer height, and I haven't been able to go any faster than 30mm/s. I tried a layer height of 0.2mm and was able to do 40mm/s. I guess the EZ-struder is working as it's designed to, but it's just far to weak for the speeds I want to print at. The lack of a tension adjustment makes it much worse than it could have been. My printed extruder is already outperforming the EZ-struder. I brought the stepper current back up to 195, tightened down the filament, and printed a solid 3cm cube at 40mm/s with a 0.5mm layer height. No skips. Now that I've added an adjustable tensioner, the speed limit should be determined by the torque of the stepper motor. I'll have to do some more testing to find out what that limit is. I've switched back to the freshly cleaned 0.5mm nozzle now, so I can get some detail back into my prints. I measured the extruded plastic at 0.45mm. I know it should shrink to be less than 0.5, but does that seem low to anyone? I'm going to set the nozzle size at 0.49mm and see if I get better results.

Here's my printed extruder.

I had to make a few changes to my first design. The mounting holes were a little too low. The gear that the airtripper uses is a smaller diameter than the mk7, so the filament path was too close to the gear. I had to slide everything over so the filament wouldn't curve around the gear. A few things could still use a little tweaking, like the size of the nut captures, but this version is working well for me.

airtripper_mod_v4.stl

(146.18 KiB) Downloaded 403 times

[geneb]

The speed limit is mostly based on how fast the hot end can melt filament. Pushing harder isn't the answer.

Never print a layer height that's higher than the diameter of your hot end. If you want to print .5mm layers, use a .5mm nozzle.

g.

[cope413]

Again, I'll say what I said before, your extruder wasn't the issue. I print much higher than those speeds with layers higher than that. I've got over 800 hours on my machine and have printed with virtually every kind of filament out there and never had an issue with the extruder itself.

Your "fix" is just a bandaid.

[626Pilot]

I recommend ordering some spare PTF connectors. If the new extruder works by brute force it will eventually cause one of them to fail. If the PTF connector on your hot end goes, it may drop tiny metallic shards into the hot end, so be ready to clean it out. That happened to me a few times when I was messing with speeds over 100mm/sec, but also when I was troubleshooting a J-head that wouldn't stop jamming.

I probably said this before but you might consider getting an E3D hot end. I've tried the SeeMeCNC, J-head, Budaschnozzle, and E3D hot ends and the E3D is the most trouble-free experience I've had. It's all metal, and can handle hi-temp stuff like nylon and T-glase that can melt a SeeMe hot end. There are still a few gotchas you have to learn, but the time investment is worth it. I'm sure Cope will tell you the same thing. There is probably some way to fix your SeeMe hot end, like there was probably some way to fix mine, but at some point you have to ask yourself, "how much more time do I want to spend on this before I try something else?" What you have now may work but the price is that you will keep blowing PTF connectors.

[osr]

The 0.49 nozzle size setting over extruded so I put it back. I think it was just the skipping thinning out my beads, because it looks fine now. I wish I could find some problem with the hot end. When I push filament through by hand, it just gushes out. I can manually extrude plastic much faster than the extruder can. I do think there was something wrong with my EZ-struder, because the skipping used to happen at lower speeds too, just not as often. With the printed one, there's no skipping until I reach a certain point. I can now walk away from a print and not worry about coming back to a big mess. That was the main problem I had. I do have some extra pneumatic fittings in case they can't take it, but I didn't tighten the tensioner very much so as not to stress the stepper bearings. I can't print too much faster than before, but I can do it much more reliably. It's good to hear that higher speeds are possible, and I guess the hot end is the only other thing that could be in the way, but I have no idea what to do to it next. I could redo the foil and silicone, or maybe use some kind of thermal paste instead. I'll probably get a thermocouple and see if the temp is off. In my last temperature test, I started a print at 190c, at the speed where it was just starting to skip. I raised the temp in 5c increments up to 220c during the print to see if there was any less skipping, but there was no difference at all. Not knowing if the thermistor reading was accurate, I didn't go up to the 240c limit in case it's reading low. The idea of disassembling and reassembling that thing yet again is really unappealing, so I might just order an E3D instead.

The speed limit is mostly based on how fast the hot end can melt filament. Pushing harder isn't the answer.

Never print a layer height that's higher than the diameter of your hot end. If you want to print .5mm layers, use a .5mm nozzle.

g.

I was using the 0.7mm nozzle with that layer height.

If it wasn't for the bowden tube, I think the force of a greg's wade would be a great way to solve this problem. There must be a great deal of pressure inside my extruder mounted jhead. That pressure would surely deform the PTFE tube inside the SeemeCNC hot end, even if it didn't break the pneumatic fittings first. For now though, I'm happy to finally be printing things instead of working on the printer.

[626Pilot]

I wish I could find some problem with the hot end. When I push filament through by hand, it just gushes out.

What happens if you heat it up, feed it through by hand, then leave it alone for two minutes and try to feed some more through? Does it require noticeably more force? Also, what happens if you turn off retraction? I've been flabbergasted about how easy it was to feed filament through a hot end, but as soon as it's left to sit still for a short while it gets plugged up like it's been to a chili cookoff. The retraction thing is because sometimes the filament will run through fine, but as soon as it's retracted it'll find a tiny crack somewhere, expand into it, and then it jams.

[osr]

I was rereading this thread and noticed this post.

Here's what I would do. COMPLETELY disassemble the hot end, including unscrewing the PEEK from everything else and carefully inspecting it and the PTFE liners. You may find a thin layer of plastic hanging around on the outside of the inner or outer liners (or both.) You can remove that pretty easy as it will be thin and won't adhere well to the PTFE. (Yay for Teflon, I guess.) Make sure that the ends of the liner tubes are completely flat, of equal length, exactly long enough to reach from the top of the inside of the hot end to the nozzle when screwed in, and that there are no burrs. Inspect the inside of the metal and PEEK components and ensure no plastic is hanging out in there either. What happens is that this plastic will swell and increase the feed pressure, and then the cussing starts.

There are supposed to be two liners? My hot end only came with one PTFE tube inside of it, the smaller one that is the same size as the bowden tube. It looked kind of funny to me the first time I took it apart, but I assumed that was how it was supposed to be. The PLA must have been expanding in the tube and creating too much friction. I put a second tube in, a 4mm ID 6mm OD tube from SeemeCNC, but there was still plenty of room between the outer PTFE tube and the inside of the hot end. Is this the right size tube for the outer liner? I was not able to make the tubes the same length. The smaller tube goes from the pneumatic fitting to the bottom of the recess in the nozzle, and the larger tube is a little shorter, resting on the top of the nozzle. Anyway, I tried printing with it for a while and was able to go a little faster. I played with temps yesterday and got up to 50mm/s before the pressure was too much for the stepper motor.

Today though, my E3D arrived and I was happy to install it. I used your mount, 626Pilot. It's a perfect fit. I was worried about the PLA warping from the heat of the hot end, but the top of the E3D isn't even warm when the nozzle is at 300C. The prints have been great, up to 70mm/s without even a hint of skipping. Next, I bolted the EZstruder back up to see if it would work. It did not. There was random skipping at all speeds. I think I've finally found the reason. The small bearing on the red lever has quite a bit of play in it. While I watched it print, the filament seemed to wander back and forth across the drive gear. By hand, I can move the outer race maybe 0.5mm relative to the inner. I'm just guessing at the amount, but it has way more play than a used skate bearing. My printed extruder is working fine, but I do like the idea of a lever for faster filament changes, so at some point I'll probably make something similar to the QUBD extruder with a 608 bearing, adjustable spring pressure, and a bowden tube mount. Thanks to everyone here who helped me get this thing up and running, especially 626Pilot for pointing out the two PTFE tubes. Actually I'm glad I missed that post the first time around, or I might not have ordered the E3D. It only took 3 days to arrive since I ordered it from Filastruder, their new US supplier. I would still like to get the SeemeCNC hot end working to test another printer I'm building, so if anyone can tell me how it's supposed to be assembled that would be helpful.

[Eaglezsoar]

A cad drawing of the hotend is here: http://forum.seemecnc.com/viewtopic.php?f=54&t=2254