I had a reel of PLA do this to me after I over-heated it in my oven (I had thrown a bunch of PETG in there to dry them out and grabbed the PLA By accident)
I couldn’t figure out the problem until I saw a video of a guy snapping filament like dry spaghetti, and sure enough…I’d hit the glass temp and pooched the whole reel.
Did I mention that this machine is a time suck? 12 Benchy’s in and it’s still not dialed up (there’s only 11 in this pic because the first one was so bad that it went right into the recycle bin LOL).
Fighting part cooling vs overheating the filament vs printing speed. If I print too slow it’s unhappy, if I print too fast it’s unhappy … several different fan layouts (that all seem to cool down the hotend too fast). Found a few interesting bits in the bowels of Klipper that I also had to tweak, this hotend doesn’t seem to live up to Klipper’s expectations.
Still not feeling the love.
Can I ask what is the process you’re using to try and find the right temperature/flow rate?
As a first step, wouldn’t printing something like a temperature tower be better than a bunch of Benchies?
I did print a temperature tower and it printed fine across a pretty wide range so that wasn’t a ton of help 
It also does a pretty decent job with calibration cubes, and the sizing is pretty good, even on most of the benchy areas.
I’ve been using mostly the info on the 3dbenchy.com site as well as a couple of videos that walk you through it. It’s actually a fantastic benchmark, there’s a lot of indicators with it. The Teaching Tech youtube channel is a good video about it: How a 3DBenchy can tell where your 3D prints need improvement - YouTube
Before I went and changed it up with this new extruder/hot end combo I had it dialed up reasonably well with the stock hotend (other issues like bed movement aside). The only thing that’s been consistent so far with the next extruder/hotend is the poor quality 
(actually that’s not true, I have made good progress on some bits, but some other bits just don’t want to cooperate). Part cooling vs temp vs speed vs flow is what I’m trying to juggle. I’ve started using pre-built BIQU BX profiles (should have been at least close given it’s the same working end), I’ve built from scratch, I’ve built by hand in Cura and Prusa Slicer. Once I found the Klipper setting that allowed me to tweak the panic attacks it was having for hotend changing too quickly sometimes it has been better.
I’ve pid calibrated, changed up nozzle sizes, changed fan mounts (single vs dual vs other layout). Faster with more heat, slower with less heat, more fan, less fan. Has been feeling like a bit of a moving target.
Do you have any experience in “Design of Experiments”? It’s a discipline that is used to understand sources of manufacturing variability to find the critical parameters.
The most popular approaches are Taguchi, TRIS and Shainin. Personally, I would recommend the Shainin technique as it allows for fairly simple experiments with (much) fewer test runs that Taguchi and doesn’t do a deep dive into the design of the printer mechanism like you would be expected to do with TRIS.
Unfortunately, I can’t find a good and simple explanation on line. The best I’ve found is: WHAT IS THE SHAININ SYSTEM™?
In a nutshell, the process is to select a number of reasons for your quality problem (in your case it would be things like nozzle temperature, speed, print angle, etc.) and then run experiments varying just these parameters to see what has the biggest impact on the quality of the prints (called the “Red X”). From there, you work on the Red X to see which values give you the best quality.
If you’re doing a dozen prints, then you’re probably doing as many as you would with a simple Shainin experiment. You just need to methodically identify your variables and then run tests changing one at a time.
Thanks, I’ll take that into consideration. I’ve been doing small changes that I think will make a difference and re-printing. Not too many changes at once, a lot like the Shainin system info you linked to. Not trying to solve every problem at once … just trying to solve the biggest issue and then working my way in from there, if that makes sense.
I think you’re missing the point - in Shainin, you’re changing one parameter at a time, not multiples. Once you’ve tried all your parameters you look at your data and see which parameter caused the biggest impact and work from there. The number of experiments is proportional to the number of parameters you’ve identified.
In Taguchi you’re changing multiple parameters at the same time and the number of experiments you have to run is factorial which means you can run into the thousands.
That’s pretty close to what I’ve been doing. I’ve changed more than one param at a time, but only two at most that work together. Had I done only one param at a time I would have even more benchys~!
Why don’t you try just changing one (say temperature) and see what happens?
I will do that (I have done that with some of these ones). But not tonight, I’ve given it enough of my time today!
1 Like
Oh man that looks frustrating. The sad part is a benchy isn’t difficult, I am sorry you are struggling.
If you are just flipping around with temp and fan settings and not seing a massive difference I have a different suggestion. Change the nozzle.
I found they wear out quickly (brass ones) and the do odd extrusion things and blob and string often. It looks like a possibility to me.
I like the clearance castle from makers Muse Can you Conquer the Clearance Castle? - YouTube
and
Sunshine engine (its free)
1 Like
So if everything was groovy before the extruder/hot end upgrade…
At your age, a bum nozzle is nothing to be ashamed of.
LOL
The reason for the upgrade in the first place was to be able to print more materials (like flexible stuff), plus I was adding a CR touch due to the wonky bed issues, I was at the point where I had to manually level the bed every other day or things were not so groovy.
Also I have changed the nozzle as well. I’ve been trying to dial it up with a 0.8 hardened steel nozzle (which is what I use on my other printer with great success). I had actually even changed to a new one of the same size. last night I changed it to a 0.6 steel nozzle and did one print (of a different torture test just because I was tired of looking at the benchy). It did ok but not amazing, same sort of deal as I was having with the benchys.
Ok so maybe I got to the root of the issue here … it seems like the new thermistor that came with the BIQU H2 is wonky. I pulled out my laser temperature reader and the temps that the heater block was hitting were extremely inconsistent with what the thermistor has been reporting (even after multiple pid calibrations). That would potentially explain why I’ve been pulling out my hair and that changes didn’t always reflect accordingly in the prints. At least I know I’m not losing my mind. Also I suspect that I’ve been trying to push it a bit too hard in terms of speed and flow for the larger nozzle sizes I’ve been attempting to use …
I still have the original thermistor which I’m going to reinstall. I actually damaged the other one last night trying to remove it, whoooops. Given that it was not reporting properly not really a big loss.
I also just pulled the trigger and ordered a brand new Prusa i3 MK3s+ kit. Lead time is supposed to be 3-4 weeks … so I will have some more quality time to spend with this 3Max before I get to build out a new Prusa. I’m really really hoping that the thermistor will solve this maddening issue. I’m running out of things to replace in the hotend.
1 Like
It does sound to me is a core issue, Thirmostor, Heater or something in the “core” of the printer. My mind works very much like dividing, test theory, and moving on.
Let us know if we can help
Congrats on purchasing a Prusa MK3S+. My recommendations for assembling the kit is to use the online manual instead of the hardcopy that comes in the kit. Also, take your time building it and look at the assembling pictures for proper orientation and read the comments. There are a few of us here that own Prusa printers and can certainly help out if you need it.
BTW, there are conflicting opinions on whether to clean and lubricate the Misumi LMU8 linear bearings or go with the manual that indicates the bearings are already lubricated. I did not as the manual indicated but had scored x-rods within the year of use. I got the rods and the bearings replaced on warranty. I cleaned the replacement bearings with 99% IPA and greased them with Superlube. I still got a small scratch on the top x-rod a few months later. I feel this is just a consumable item and will need to be replaced at some point. The rods can be rotated so the bearings don’t ride on the scratch. Needless to say, I went straight to cleaning the bearings in the new kit and lubed them before installing. Not to say you should do what I did, just thought I would mention there are two opinions on this, but if this is your first printer, follow the manual and watch your rods for the first year.
Thanks for the tips. I was planning on cleaning and lubing for sure. Have a can of superlube handy and not afraid to use it
I’ve assembled quite a few other printers (and other similar-ish devices like a couple of CNCs), so I’m just going to take my time and get 'er done well. Have already seen a couple of great (recent) videos covering some of the more common failures as well that I’ve already watched.
Also on the replacement parts that’s something I’m going to look forward to, as they have a full catalog of parts for just about anything they’ve built from the looks of it. The nice thing is when it comes time to replace something I’m relatively confident I’ll still be able to get a proper part direct from them.
Interesting. I followed the instructions and had no issues. One of my cleaning moments I tightened the bolts holding the head to rod more than I did previously and scored the rods. I wounder if it is more pressure than lube?
I wondered that myself. Similar to your line of thinking, I feel when I started printing ASA and ABS at the higher temperatures, the extruder backplate may have deformed, possibly due to what you mention and the tie wraps being too tight?
I have a roll of the Prusament PC CF that I haven’t tried yet. I might attempt to print the back plate with it to make it more durable.