Tips look fantastic but diameter is too big for reentry
I've been messing with this MMU since it was one of the first MMU2s. It seems the longer I use it, the worse it gets. I know the problem I have is exactly this:
When a filament is pulled, the tips are wider than the filament. Here's an extreme example. Filament diameter is 1.80mm, the tip is 2.60mm. Typically, the tip diameter is around 1.95mm or so. I've read as many of these forum posts on extruder tuning, tried everything. I sometimes get slightly narrower tips when I put a new bowden tube in the extruder. The wider tips will pass through the extruder gears, but won't fit into the heat chamber. The gears start chewing the filament, and that's the end of hands off printing. It's now happening on every single 'removed' filament tip. My current solution is to arrange the filaments so that the selector is far enough away from the just removed filament that I can push it out, clip the end off, tuck it back in while the next filament is in use. In other words, I have to have my eye on the printer 100% of the print time.
I'm to the point where I'd just like to put a vacuum or drain in the MMU tray to prevent overflows and just cut the wide tips off by default, but I see all these posts with success stories. The slim chance of proper operation keeps me trying to find the real problem and eliminate it.
Has anyone else had this problem? The solution is NOT to reduce the spring tension in the Extruder Gears, because the wider tip passes through the gears just fine. It is too wide to get back into the heat chamber, which is odd, because it was able to come out from that chamber effortlessly.
RE: Tips look fantastic but diameter is too big for reentry
I did some late night web browsing, and read that ramming too much will create the problem I'm having. During this time, there was a lightning strike, so now I need to get a new display board for the printer. Once that happens, I'll start over again with the tip tuning. I guess I was so concerned about stringing I didn't realize I had drifted to fat tips.
I'll keep a log of what I do, because it might help others.
RE: Tips look fantastic but diameter is too big for reentry
I've got an MMU3 fitted to my CORE One, and I think 'fat tips' is the issue I'm suffering from. My MMU3 was much more reliable on my MK4S, so I've moved it back to that machine, but now I've added an MMU3 clone (the Blurolls kit built up into the Ultimulti printed parts), and I'm still having the same problems. Time to try and get to the bottom of them.
I'm mostly printing ABS (which the CORE One handles brilliantly). If I bypass the MMU, filaments load and unload entirely reliably, without me having to snip the ends off before manually reloading. With the MMU the tips seem to be fatter than the filament, as the OP described above, such that they often (usually) jam on the way into the top of the extruder. Sometimes they're fat enough that they don't even unload.
Given that the basic CORE One (minus the MMU) now does a bit of ramming after a print anyway, and pulls the filament clear of the hot end (so it can be unloaded without reheating, and so that it doesn't ooze during heating - both great features), and that the tips created by this process seem a lot better at reloading than the MMU-formed tips, I think I'd like to recreate the same process in the MMU configuration. ABS in general doesn't seem to string like PETG for instance, so I'm pretty sure that a tip created by the standard unload process would load to the MMU quite reliably.
But I don't really know how to go about this. I see there is a filament-specific 'Toolchange parameters with single extruder MM printers' section. There is a 'Ramming settings' button that brings ups a configuration screen, but it's not clear to me how I should configure that screen, or the rest of the toolchange parameters, to try and get the same tips as the stand-alone printer.
So does anyone have any tips for my tips?
RE:
I had a blind stab at changing the filament change settings for my ABS, and my first impression is it seems to have improved things. But since I didn't do this in a very methodical way I don't really know which setting was responsible. Maybe I'll work on that, or maybe I'll just stick with what's working until the next time it doesn't!
1) I changed the number of cooling moves from the default of 3 to1. Even if that hasn't helped with the 'fat tips' issue, it has at least speeded up each filament change.
2) In the 'Ramming settings' configuration page I reduced the 'Total ramming time' from the default of 1.25s to 0.5s.
The tips are definitely not as clean as before, with a slight wisp of stringing, but neither are they as fat as they were, which was my objective. I think it's probably the ramming time that's made the difference, and if I increase the number of cooling moves again I may get rid of that slight wisp of stringing, but I don't think it's going to be a problem.
In the process of figuring out where the jamming was happening I noticed that the selector on the MMU has a couple of pinch points. The M5-4 fitting has a small hole where the filament feeds through the centre. Not a problem for freshly cut filament, but if the tip is slightly oversize or crooked then it's likely to get stuck here.
And in keeping with this fitting, the selector body has a similarly snug filament hole through it. The rest of the filament path seems to have no further constrictions smaller than the inside diameter of the PTFE tube.
So I made a new selector plate to take an ECAS fitting, which allows the PTFE tube to pass all the way through, and I modified the Ultimulti B19 selector to have a bigger filament hole - still slightly smaller than the PTFE ID though, to make sure there's no step to block the filament loading path.
If these parts work ok over the next few days I'll post them as remixes of the Ultimulti model.
RE:
50 flawless unattended filament swaps. Confidence slowly returning!
And with my rack of Filamentalist rewinders also working flawlessly, all is serene once again 🙂
RE:
There seems to be some pattern/logic to the settings, and it seems like I accidentally hit on my target with my first guess.
Here's the tip that's formed by the non-MMU standard unload sequence (I'm using Overture ABS with the Prusa ABS profile, because there doesn't seem to be a Generic ABS). Ignore the yellow contamination, that's just the previously loaded filament.
Compare that to the MMU default that uses 3 cooling moves and 1.25s of ramming. The tip is very clean, but the wider compressed section is longer.
Here's the tip with my first stab in the dark - 1 cooling move and 0.5s ramming. It looks a lot like the default non-MMU result.
Finally I wanted to see if there was any benefit in adding another cooling move. Here's the result with 2 cooling moves and 0.5s ramming. The tip seems slightly better.
My guess is that the length of the fatter compressed section is as much an issue for reloading as the width, and a reduced ramming time shortens that section. I tried an even shorter ramming time, but when I set 0.25s it changed to 0s, and 0s ramming makes a mess of the wipe tower. SO I'm going to settle on 1 cooling move and 0.5s ramming for this Overture ABS.