As I mentioned in another thread, I donated a Mini to a local elementary school. We printed the octopus with the short length of HIPS included with the printer, then switched to PLA and printed a couple of disks with that. Both prints went well using the stock Lulzbot profiles for each filament. On the first attempt to print something with the a few days ago, the Mini went through it’s auto-leveling sequence (supposedly normally, but I was not there to see it). When it started to print, the print head moved into position then lowered the nozzle into the PEI surface and wiggled around a bit, melting a hole in the PEI. I came in about an hour later and fired things back up. There was some nasty thumping and vibration when we tried to run it through an auto-leveling sequence.
We seem to have two problems:
(1) What caused the nozzle to crash into and melt the PEI
(2) We can’t seem to get the filament out: We heated up to the print temperature for PLA. I tried to push a bit through to manually extrude some filament, but it wouldn’t budge. It also won’t let me pull it back out. We tried a variety of temperatures from 160 to 230˚C and it won’t move either in or out. I’m wondering if the heater or heat sensor (thermistor?) got damaged when the nozzle was melting the PEI and it’s now not getting up to the stated heat. The Cura screen shows it has reached the set point, but it sure isn’t acting like it. If a piece of the PEI melted into the nozzle I suppose that could be part of the problem, but I would think I could still pull the filament out.
I’ll be picking up the printer on Sunday, and hope to do a bit of trouble-shooting and clean out the old filament. I’ll put in a call to tech support once I have the printer in my hands. I thought I’d see if anyone here had any thoughts.
Any thoughts on helpful tools to have together when I contact tech support would be appreciated as well. (I’ll have the tools that came with the mini, as well as an 18mm and 7 mm wrench)
Having a point and shoot infrared thermometer on hand (automotive or tool supply, ~$35) might be useful to see if the hotend is actually heating.
As for what caused the nozzle to hit the bed, it could be a couple of things. if the metal washers came loose in shipping, and one of them deflects down when the auto leveling process hits it, the printer might think it is actually lower than it was and hit the bed. That or if someone screwed with the software Z offset value and set it to a value below Z0
Once it hit, it probably skewed the leadscrews and X carriage out of allignment on one side, which is probably causing the grinding noise. You will need to get the X axis back into sqare to get it to move properly. The easiest way I know of to do that is to measure from the top of the lower Z motor mount bearing to the bottom of the leadscrew nut. The distance should be identical on both sides. It is unlikely that you damaged the Z axis, they are pretty robust.
The filliament removal issue is odd. If it’s heating, the filliament should pull out. The only thing I can think of aside from it is no longer heating, is that there is somehow PEI resin jammed up the bore hole, and has glued the end of the filliament inside the extruder. Even then, if you get it hot enough it should still melt or burn out without doing too much damage to the hotend.
This may be a silly question, but when you are attempting to remove the filliament, are you undoing the idler arm retention latch? If you aren’t undoing that, and are trying to pull the filliament out, there may be a C shaped notch in it around the hobbed bolt actually locking it in place.
Have the printer go to the home position at te top and then turn off the printer and see if you have an equal distance from the top of the Z-Nut to the bearing at the top of the frame. You should be pretty even on both sides distance wise. This will roughly level the bed Z for you, and is fairly quick to do. If they are uneven adjust only the right one to the same distance from the bearing to match the left one, by turning the right Z rod up or down.
If it will not perform this without making noise after doing this you might have something broken like a motor shaft or a damaged belt.
Thanks for the tips, guys. I’ll give them a shot once I get the printer home. We did get it to go through an auto-leveling sequence finally, and all appeared normal - no bed or washer deflection when the nozzle touched down.
The Cura software SAID the head was at the set point (we tried several temps, up to a high of about 240). When we adjusted the set point, you could see the display temp going up or down as appropriate The head was putting out at least some heat. When Cura said it was at 200˚C, I touched it using a scrap of an old cotton chamois/flannel shirt and could feel some heat. I don’t have a feel for how hot it should feel (other than “hot enough that I know I don’t want to touch it bare handed”). I was surprised that I was not feeling more heat through the cotton, but have little to compare it to.
Piercet, I’m not at all offended that you asked about whether the idler arm was released. When helping someone troubleshoot, you have to ask the simple questions as well as the more complex ones. (And yes, my first gentle tug was with the arm still latched, but I figured that out pretty quickly.)
Do you have a recommendation on a point and shoot IR thermometer? I used to use one all the time at work. My recollection is that they were rather inaccurate on shiny surfaces like bare metal we had one that was adjustable for the emissivity of the object being measured - but often times we’d leave that adjustment at it’s default setting and just stick a piece of masking tape on the object to be measured and measure that. The other thing I remember was that the tightest the thing would focus (on our tightest one) was about a 1/4" or 3/8" circle (going from memory back 20 years here). If your part was smaller, the instrument would display an average of the part plus the background temperature of whatever was in its view. Is this an issue with some tools?
I have one of the Fluke units myself, but they are ridiculously expensive. For this purpose, pretty much any of them will be accurate enough. You want to know if your bed is getting up to the right temp, if the nozzle is heating, etc. A close enough reading +/- 5-10 degrees will get you the information you need for that. The better ones can read a spot about 3mm in diameter, but there is still some variance, especially on the cheaper ones.
If you want more detailed information about a part, thermal imaging cameras have dropped a huge amount recently into the range of actually affordable, I purchased one of the Seek Thermal Android Infrared cameras and that works pretty well. It doesn’t have an albedo calibration like the more expensive ones, but for $200 its not a bad device at all http://www.thermal.com/ They have a hand held standalone one now too. That comes in very handy for troubleshooting bed adhesion issues, as well as designing printer upgrades that affect heat
If you end up not being able to remove the PEI from the nozzle, you may want to consider getting a small pile of replacement spare parts on hand. The refurbished parts from Lulzbot all end up here: http://i-t-w.com/parts/ And you can get a few hotend heater blocks or even full barrels to have on hand pretty inexpensively. Having a couple extruders on hand just in case one fails, or to allow for experimental extruder designs (or flexystruders) might be useful. You can also pick up a spare bed assembly there if they have them in stock. Your printer should still definitely be under warranty, but I remember from my high school days that kids can be hard on CNC equipment, so the teacher might want to order some spares of things, and that’s the least expensive source I know of.
I ended up ordering them a complete spare print head. As you commented, school kids will find a way to break something. Also ordered a couple of spare nozzles and a spare PEI sheet. All from I-T-W. Unfortunately, they were out of reconditioned print heads.
The Z motor couplers can break occasionally, and it’s a good idea to have a few spare Idler arms and Gear sets for the extruder body around, especially the small gear that can wear over time. The idler arm can wear at the top notch, or crack at the pivot base. It is gluable with plastruct plastic weld, but printing a few good ones of those printed parts out of ABS should be a priority for the teacher. If you can source some of the tiny set screws for the motor pulleys (I think they are M1.5 thread) those do work themselves loose over time and occasionally fall out. The belts and pulleys themselves should last the life of the printer, but I have seen a belt shred itself before too. A GT2 timing belt 6mm wide (https://www.lulzbot.com/store/parts/gt2-1164mm-6mm-belt) might be useful to have in a crash kit, but its one of those parts you probably shouldn’t need to have on hand Nozzle wiper inserts, I don’t know how long they last, but you can make new ones from felt or red scotchguard pads, or they have a 4 pack or so in the lulzbot store. A couple of 608zz bearings for the Idler arm also might be a good idea. They don’t usually fail anymore since they moved over to the sealed bearings, but that used to be a problem. You might have to replace them once every 3 years with really heavy use now. The rod bearings should be fine for quite some time so i wouldn’t worry about keeping them on hand.
For maintenance supplies, you want some White Lithium grease to apply to the leadscrews (never the smooth rods) once every 6 months or so. Some of those white Magic Eraser cleaning pads work well for cleaning stuff off the bed without damaging it, especially in a school where isopropyl alcholol might be problematic.
You might also want to look into part removal tools. I use this: http://www.amazon.com/Calphalon-Gadgets-Dual-Cheese-Plane/dp/B000SDQGL6 with Heavy duty Gorilla Tape wrapped around the slicing part. The tip of that thing is the best part removal tool I have come across ever. Using a tool like that, or a putty knife or paint scraper to wedge the part up and away from the bed will prevent the air bubble forming under the PEI sheet problem that causes things not to stick to the bed anymore. This one also works well http://www.amazon.com/Cricut-29-0099-Spatula-Tool/dp/B000NP1ESO
Aside from that you might want some m3 and m5 heat set inserts and the Soldering Iron tip to insert them. Depending on the level and age of these particular students, having acess to the ability to put threaded inserts in their plastic parts can be useful, but a soldering iron is also a mobile death wand in the wrong hands. Either way, Mcmaster Carr has the inserts. Plus you can use those to make new Mini parts where needed. (frame, etc) You shouldn’t need any of those parts though unless something goes very wrong.
Thanks again, Piercet. I ordered the cheese slicer / part removal tool you recommended. We’ll get to printing some of those spare parts you recommended as soon as we get the printer dialed in. Is HIPS an acceptable substitute for ABS on any of these parts? If not, we’ll get a roll of the ABS. (I gave them a bunch of HIPS to get started with.)
Are these the Z motor couplers you mentioned? If so, I assume that’s something I can print also?
I found that Fastenal can get some of the heat-set inserts, though their selection is limited. I may take a look at some of theirs, since my business account will get me a 45% discount on fasteners from them. Here’s a link to what I found on Fastenal Inserts. (Go to the “Application” tab on the left and choose “plastic” to narrow the search. for some reason if I try to bookmark the narrowed search, the link no longer works in this post.) I wasn’t sure which style would work best. Their “Taper Double Vane” insert seems the closest to what is pictured on the McMaster Carr website.
HIPS is probably ok for the idler latch and the gears, I’m not sure how well it wears compared to ABS, but it will be better than a PLA set. ABS would be ideal, but worst case scenario if you have a spare HIPS part and it isn’t suitable for long term, it will give you time to print an ABS replacement.
That coupler is actually the old coupler for the AO-100 and AO-101 series printers. AN edited version of that design actually would work but that particular one is for an 8mm shaft and a 5mm shaft, you want a 5mm to 5mm one, specifically this one: http://i-t-w.com/parts/lulzbot-taz-45-mini-misumi-gsasl16-5-5-coupler which seems to be out of stock. but any 16mm long 5mm shaft to 5mm shaft helical coupler (ebay) will work fine. I would get pairs of them if you are ordering from another source just to ensure the materials match.
On Fastenal, those “TaperSingleVane Insert For Thermoplastic” look identical to the heat set inserts Mcmaster Carr sells. I think the Taperdoublevane ones are the longer style they also sell. Both are useful, the longer ones are better if you have the space, more surface area to mate with the plastic, though i tend to use alot of the narrow ones just because of where I am putting them. Those should work perfectly. It might end up being a “the Teacher Installs this for you” part, but they definitly give you a whole other layer of design flexibility
Piercet - Thanks for the coupler link. I’ll order a couple to have on hand. I’ll probably also get a roll of ABS so we can print the parts out of that. This teacher is big on letting the kids get in and get their hands on stuff. She has a well-stocker “makerspace” including several soldering irons, which some of the kids have used already (after some safety instruction, and under supervision). I’m pretty impressed for a grade 5/6 class. (Makes me wish I was in 5th grade again, though I’m sure my daughter would be mortified.)
Kmanley57 - yes, the Fastenal selection is limited. However, there is nothing that says the kids have to use metric when they are designing their own stuff. I may be able to use some of the SAE inserts that are close in size. I may have to go McMaster for their larger selection, but to give you an idea, the longer M5 inserts work out to be 31¢ from McMaster. THe list price from Fastenal is 37¢, but I can get them for about 20¢. They are cheap enough that it’s no big deal, but every little bit helps, and I won’t always be stocking the classroom out of my own pocket. It also helps that my wife works almost across the street from a Fastenal store, vs a 35 minute drive to the nearest McMaster-Carr.
Is there any reason to stick with the tapered vs the straight inserts? I’ve never used any of these before, so if anyone has a good description of the procedure, I’d appreciate it. Do you generally print the holes or drill them (or maybe print them undersize and drill them out)?
The noise has stopped. I still don’t know what it was. The Z-nut distances were as close as I could measure, though I didn’t have the best tool for measuring with me when I checked. How much vibration would there be when the hobbed bolt grinds against some PLA filament if the filament is jammed? This was LOUD, and the whole machine was vibrating.
The tapered ones go in easier. Typically for the M3 ones, you print a 5.2mm hole, which the 5.4mm wide top section sockets into, and the 5.3mm bottom section slides right into at temperature. The straight ones are more of a brute force insertion.
Where is the best place to find the files for the current parts for the Mini? I’ve come across them in a couple of places, but am having a hard time figuring out whether they are early development or current designs. Also, some of these places have a number of other files that I can’t figure out. I’m assuming I’m looking for STL files, correct?
Unless you have the springs tightened real hard the sound the bolt makes chewing filament is not loud nor machine shaking. I have a extruder right now that chews up filament fast and it makes a mild popping sound is all.
I guess that eliminates that possibility. If I’d thought of it at the time, I guess I could have just released the idler arm and seen if the vibration went away.
There are only a few differences between the two, and the extruder parts are identical. you want the herringbone gears, lg and sm, and Idler arm. An extra extruder body might not be a bad idea to print out either. Plus they look neat when they print.