Last modified: 2024-11-03 22:30:26
< 2024-11-01 2024-11-04 >The Bambu wiki has some instructions for drying filament: https://wiki.bambulab.com/en/general/bambu-filament-drying-cover
Basically you want to put the bed at the bottom of the chamber, put the spool on the bed, and cover it with a cardboard box or something, then heat the bed up to:
They suggest leaving it for 12 hours, but I'm not going to leave it that long, I'll just leave it for a while. I also don't have a suitable cardboard box so I'm not covering it.
I want to dry out PC, TPU and PAHT-CF, so I'll set it to 100 deg. C. and put the TPU at the top, further from the bed.
With the bed at 100 deg. C the chamber is only getting up to 36 deg. C, I'm not sure this is going to dry it out at all.
Looking at the notes from 2024-10-31.
The gcode-cat M2 issue was caused by having a comment on the same line as the M2, I've made peck-drill
not write those comments now.
I put the counterweight back to 40mm long, and also made it 18mm wide instead of 17mm.
The stand was cut out in the wrong place because while the actual Body was in the right place in CAD, the Model in the Path job was in the wrong place. Also fixed that now.
Since the heavy cuts are a bit much for this machine I want to back off on the feeds and speeds. The calculator suggests that if I want to be feeding at 500 mm/min then I want about 6750 rpm.
So I've made sure all of the 6mm end mill paths are at 500 mm/min and I'll try to remember to use 6750 rpm, which works out to 112.5 Hz on the VFD.
I've updated all of the gcode.
Things to remember:
If I want to get the anniversary clock escapement working, what do I need to do?
I last worked on it on 2024-09-26, and I concluded the clamps were slipping on the wire.
Ideas include:
I don't really want to modify the wire if I can help it, because that then commits me to those locations for the clamps.
Lining the clamps with sandpaper could work, but seems fiddly. Easiest has got to be to reprint then in PAHT-CF.
I think the chamber is not going to get hot enough to really dry out this filament, so I'll just take these stools out and put them in the AMS now, and see how I get on. Will retry with drying in the oven probably.
The printed parts look pretty reasonable, although the fork is over-extruded to the point that it doesn't fit around the lever on the anchor. I think dimensional accuracy is better with PLA.
I put it back together, this mechanism is super annoying. There's just a knife edge between fluttering if you give it too much drive torque and stopping if you don't give it enough.
I should possibly abandon this design and come up with something simpler. Maybe more like a normal balance wheel, or a "deadbeat verge and foliot" with balance spring.
It's also not obvious to me that the PAHT-CF threads don't strip just as easily as the PLA ones.
I'm going to see if I can get it to run satisfactorily this evening just by adjusting the fork. If not, I'll think about other designs.
I can't get it to work. The wire is slipping inside the clamp for the fork. I wanted to put nuts on the back of the bolts to clamp it tighter but there isn't room. And even then you've still got the problem of finding the exact right drive torque to make it run. I want something a bit more agricultural, that will just run no matter what. Something reliable.
The thing that originally appealed about the anniversary-style escapement is that you can get very large amplitudes, which means you can put lots of energy (and angular velocity) in the balance without a correspondingly high tick frequency. The other way to put lots of energy in the balance is to make it have a really high moment of inertia.
This is quite nice: https://www.youtube.com/watch?v=-veELEu1wXk
And this: https://www.youtube.com/watch?v=TPMPCyz68to
We could put ball race bearings on the pallets.
I'm going to try a gingerbread man in Polymaker PC-Max, even though it's probably not dried satisfactorily.
Apparently default generic PC settings are suitable.
It looks to have printed really nicely! Way better than I expected considering the age of the filament.
Layer adhesion is pretty poor, as I remembered. I can tear it apart by hand.
I thought this was unusually bad compared to PLA, but I just tried to tear the PLA one apart and that one came apart as well.
The generic PC profile uses the cooling fan quite extensively, I could reduce that. Or just live with it and design stronger parts.
< 2024-11-01 2024-11-04 >