My turn getting a Core 1+ working
I upgraded my MK4S to a Core 1+ these past three days. Trying to get through the calibration routines, and am failing on Homing. I passed the previous tests. It's not obvious what to do to fix it. My belts were loose, I tightened them using the app. The gantry was slightly off, and I tweaked it. But I'm still stuck not homing.
What's the next thing to check?
RE: My turn getting a Core 1+ working
I don't know what I did exactly, but somehow got it to work. (Passed all the calibrations.) Can't say it made a good print, but it printed. I used all defaults for the sample PLA, and it wasn't impressive. Looked like over extruded first layer, but under extruded for other layers. The threads came out poor on the male and female parts. In PETG on the MK4S they came out perfect. Looks like there's another learning curve here. First layer is quite uneven on the bottom. I'll show pictures in the morning.
RE: My turn getting a Core 1+ working
I didn‘t know how fixed the belts have to be. They have to be very tight. The app gave me now good advice. So I tightened them with the function from the printer itself and after that with the app. That was me way for homing. the sound from the belt must be a very high “ping” instead of an “bam/blam”. Maybe that is useful for you.
RE: My turn getting a Core 1+ working
Well, it was a slicing issue, at least for the print. I re-sliced using 0.1mm layers, so the 0.8mm threads would print. (Minimum feature height of an ISO thread is P/8, where P is the pitch.). The print in PLA came out good, and the female cap screwed on the male thread. That's encouraging. My gcode last night didn't have the settings embedded in file name. This morning, the newly re-sliced file did have the nozzle size and layer information in the gcode file name.
What's not good is the Core 1 has issues with homing. It said collision detected while homing and asked if I wanted to re-home again. So I retried, and after 5 minutes it was successful. So out of a 25 minute print, there was at least 10 minutes additional where the printer was having difficulty finding home. It could be this printer is out of square, or something like that. I assembled it on our dining room table, which probably is neither flat nor level. While printing, it was visibly rocking due to movement. I will move the printer to it's normal area, hopefully that will be a more stable platform. I will find a square and tape measure and do a sanity check on squareness.
Anyone know where I can get a plastic rivet pulling tool? Just in case I need to pull them? Can they be reused?
The behavior is different for the Core 1. Is the filament wipe supposed to be on the right side? Yes or no? Used to always be on the left. (MK3 - MK4S).
RE: My turn getting a Core 1+ working
What's not good is the Core 1 has issues with homing. It said collision detected while homing and asked if I wanted to re-home again. So I retried, and after 5 minutes it was successful. So out of a 25 minute print, there was at least 10 minutes additional where the printer was having difficulty finding home. It could be this printer is out of square, or something like that. I assembled it on our dining room table, which probably is neither flat nor level. While printing, it was visibly rocking due to movement. I will move the printer to it's normal area, hopefully that will be a more stable platform. I will find a square and tape measure and do a sanity check on squareness.
Anyone know where I can get a plastic rivet pulling tool? Just in case I need to pull them? Can they be reused?
The behavior is different for the Core 1. Is the filament wipe supposed to be on the right side? Yes or no? Used to always be on the left. (MK3 - MK4S).
Yes, the filament purge occurs on the right side. It doesn't really matter where it happens. You can change it in PrusaSlicer (in Expert mode) under Printers -> Custom G-code -> Start G-code. It is in this section. For example change X from X119 - X235 to X9 - X45. However, since you have a C1+ I wouldn't put the purge line there since you could run into the vent latch. Suggest just leaving it where it is.
As for homing not being consistent, that usually due to either a skewed gantry or the belts are too loose.
;
; Extrude purge line
;
G92 E0 ; reset extruder position
G1 E{(filament_type[0] == "FLEX" ? 4 : 2)} F2400 ; deretraction after the initial one
G0 E5 X235 Z0.2 F500 ; purge
G0 X225 E4 F500 ; purge
G0 X215 E4 F650 ; purge
G0 X205 E4 F800 ; purge
G0 X202 Z0.05 F8000 ; wipe, move close to the bed
G0 X199 Z0.2 F8000 ; wipe, move quickly away from the bed
RE: My turn getting a Core 1+ working
Anyone know where I can get a plastic rivet pulling tool? Just in case I need to pull them? Can they be reused?
Most rivets can be easily reached from the inside of the enclosure. I recommend pushing out the pin from the inside, then it is easy to grab and pull with your fingers from the outside.
You can leave the little collars in place; they will not prevent you from taking off the side panel. But before reinstalling the panel, I find it easier to take the collars out as well, then re-insert them through both parts at once and secure with the pin. You may find that you have to squeeze the two "prongs" of the collar together a bit when installing a used part, because they get widened during use. But if you do that, you can re-use the rivets many times.
RE: My turn getting a Core 1+ working
What's not good is the Core 1 has issues with homing. It said collision detected while homing and asked if I wanted to re-home again. So I retried, and after 5 minutes it was successful. So out of a 25 minute print, there was at least 10 minutes additional where the printer was having difficulty finding home. It could be this printer is out of square, or something like that. I assembled it on our dining room table, which probably is neither flat nor level. While printing, it was visibly rocking due to movement. I will move the printer to it's normal area, hopefully that will be a more stable platform. I will find a square and tape measure and do a sanity check on squareness.
Anyone know where I can get a plastic rivet pulling tool? Just in case I need to pull them? Can they be reused?
The behavior is different for the Core 1. Is the filament wipe supposed to be on the right side? Yes or no? Used to always be on the left. (MK3 - MK4S).
Yes, the filament purge occurs on the right side. It doesn't really matter where it happens. You can change it in PrusaSlicer (in Expert mode) under Printers -> Custom G-code -> Start G-code. It is in this section. For example change X from X119 - X235 to X9 - X45. However, since you have a C1+ I wouldn't put the purge line there since you could run into the vent latch. Suggest just leaving it where it is.
As for homing not being consistent, that usually due to either a skewed gantry or the belts are too loose.
; ; Extrude purge line ; G92 E0 ; reset extruder position G1 E{(filament_type[0] == "FLEX" ? 4 : 2)} F2400 ; deretraction after the initial one G0 E5 X235 Z0.2 F500 ; purge G0 X225 E4 F500 ; purge G0 X215 E4 F650 ; purge G0 X205 E4 F800 ; purge G0 X202 Z0.05 F8000 ; wipe, move close to the bed G0 X199 Z0.2 F8000 ; wipe, move quickly away from the bed
I did tune the belts using the Prusa belt tuner. I don't understand how to deskew the gantry properly. As I understand it, you loosen the belts, then tweak the x cross bar so that the ends arrive at front stop evenly? What's the tolerance? How much gap is allowed on both sides? Or am I misunderstanding this procedure? Then I need to tighten the belts again? When I do the belts, do I need to go back and forth between them? They are interactive? The belt tuner doesn't say to do this at all...
RE: My turn getting a Core 1+ working
I did tune the belts using the Prusa belt tuner. I don't understand how to deskew the gantry properly. As I understand it, you loosen the belts, then tweak the x cross bar so that the ends arrive at front stop evenly? What's the tolerance? How much gap is allowed on both sides? Or am I misunderstanding this procedure? Then I need to tighten the belts again? When I do the belts, do I need to go back and forth between them? They are interactive? The belt tuner doesn't say to do this at all...
Follow this guide: Adjusting belt tension pay special attention to the "Quick Gantry Alignment" section. Short answer is the left and right screws should never be turned independently of each other. They should be tightened or loosened by the exact same amount. Never use the belt tension screws to correct gantry skew. That has to be done by manually bending the gantry.
RE: My turn getting a Core 1+ working
I did tune the belts using the Prusa belt tuner. I don't understand how to deskew the gantry properly. As I understand it, you loosen the belts, then tweak the x cross bar so that the ends arrive at front stop evenly? What's the tolerance? How much gap is allowed on both sides? Or am I misunderstanding this procedure? Then I need to tighten the belts again? When I do the belts, do I need to go back and forth between them? They are interactive? The belt tuner doesn't say to do this at all...
Follow this guide: Adjusting belt tension pay special attention to the "Quick Gantry Alignment" section. Short answer is the left and right screws should never be turned independently of each other. They should be tightened or loosened by the exact same amount. Never use the belt tension screws to correct gantry skew. That has to be done by manually bending the gantry.
I don't understand the Quick Gantry Alignment section. I don't understand specifically the manual bending of the gantry. The words confuse me, and I can interpret them several ways. What needs to be done?
RE: My turn getting a Core 1+ working
Try to understand the mechanics behind it, then it gets clearer. It's about the two sheet metal parts at the ends of the gantry that form 90-degree angles.
- ultimately, left and right endstops for the gantry need to be reached at exactly the same time or calibration will fail (and distortion would result on a print).
- without belt tension, this boils down to having ideally exact 90 degree angles on both sides (or less ideally, both sides off by the same amount so the errors cancel out in sum, but putting permanent stress on the linear bearings)
- the need to bend the sheet metal parts comes from the manufacturing process tolerances being too loose. So those parts require "postprocessing" during assembly.
- When you have the gentry perfectly aligned without belt tension (e.g. pulleys detached), you need to get them tightened evenly to nominal tension (belt tuner resonance frequency)
- as the two belts pull in different directions and the procedure is very sensitive (1/8 turn on the belt tension screws makes a measurable difference), the trick is to tighten them up without pulling the gantry out of alignment. If under tension there is a gap e.g. at the right end stop, it means the right tensioner screw is too loose and the left one too tight. At the end of the day, both end stops should be reached at the same time first without belts, then again after belt tuning.
RE: My turn getting a Core 1+ working
If under tension there is a gap e.g. at the right end stop, it means the right tensioner screw is too loose and the left one too tight. At the end of the day, both end stops should be reached at the same time first without belts, then again after belt tuning.
Thank you for that gem. Let me ask this question, can the system be tuned by judicious belt tensioning with this in mind? Can I achieve this without out doing the next paragraph?
Or do I need to de-tension first, then tweak the cross piece to roughly touching the front equally on both sides, and then tightening the belts. Do I need to tension both belts simultaneously? (Use two wrenches, and two hands?)
Once I understand this, it won't be tough to do. But I hadn't understood the big picture, which made it difficult to proceed intelligently.
RE:
You need to bend the metal first (gantry perpendicular - no error in end stops) then tighten belts to exactly the same tension (belt forces in balance - the gantry remains perpendicular).
What you should absolutely NOT do is use belt imbalance to compensate for a bent gantry (and I'm carefully picking my words to not convey that message). In sum it seems correct but there will be multiple mechanical forces fighting each other. It's soft steel, those parts bend easily, the task is not difficult.
You can tighten with a single wrench, just use common sense e.g. advance in quarter turns. People have managed to bend the gantry via belt imbalance (what comes next is ripping out the tensioner screws) but this seems very unlikely after some thought about the mechanics behind it.
RE: My turn getting a Core 1+ working
What you should absolutely NOT do is use belt imbalance to compensate for a bent gantry
Thanks for the clarification. I know what I need to do now (I hope).
Had to move the printer to it's "real" location, NOT on the dining room table. Things may have been "adjusted" by my carrying it upstairs and lifting it up onto a 114cm (45") high dresser. I needed assistance for that, lifting the base nearly 4 feet off the ground was quite difficult. The Core 1+ is heavier than the old MK4S!
RE:
my process for "bending" the gantry:
- completely loosen and detach the tensioning screws. If you like, use painter's tape, wire or whatever to keep belts and idlers roughly where they are
- pull one side forwards, push the other backwards. The force I remember might have been 5 to 10 kg ("kgf") that is, the force of the weight of 5..10 one-liter milk boxes (for example). But, start iterating up until the gap changes. As said, it's not particularly difficult.
Before gantry alignments, please make sure the printer is on a truly flat surface and feet are attached!
RE: My turn getting a Core 1+ working
Well, I think I tweaked it in. Belts are ok. Might need another iteration, but it is better, seems like both sides are at the stops now. Yes, the feet are attached! I did follow the assembly instructions. 😊
As for truly flat surface, well that's not likely, since you are talking to a guy who machines (and knows about flatness). My dresser top is not what I'd call flat, nor even carpenter level, but it's ok enough for a 3D printer. (It's somewhat level.) Only truly flat surface I have is a small 22.5 x 30.5 cm granite surface plate. I used the surface plate to calibrate my machinist level. It sits on three legs which needed to be adjusted, via fine screws, just to get within the leveling range of the level. I used a carpenter's level to get vaguely close, then fooled with the feet to avoid pegging my sensitive level. (0.0002"/10" or 20um/1m)
RE:
Well, take my "truly" flat surface with a grain of salt, no need to correct for the curvature of the earth. What I had in mind is, it should feel equally solid on the ground shaking it along both diagonal axes.
Yes, the feet are attached! I did follow the assembly instructions. 😊
Right, but those feet have an annoying habit of wandering off on their own thanks to the stock adhesive tape... I'm quite sure there's a fair number of three-feeted core ones out there and it doesn't exactly help with the headbanging during homing.
RE: My turn getting a Core 1+ working
Right, but those feet have an annoying habit of wandering off on their own thanks to the stock adhesive tape... I'm quite sure there's a fair number of three-feeted core ones out there and it doesn't exactly help with the headbanging during homing.
Fortunately, the feet seem to be sticking quite well. I did make it a point to clean the surface with an alcohol wipe and let it dry prior to installing the rubber feet. This printer is somewhat heavy, so I'd bet people try to push it rather than lift it, and that would tend to de-adhere the glue. I think the rubber feet are stickier than the pads on the MK4S, and they are much larger in area. Haven't had a chance to print today, been busy with trying to set up the machining of some little brass orifices for a steam vent. Don't think a printed one would last very long. 22 to 100C every heat cycle.
The printer is sitting solid on the dresser, so that's good. I'll do a quick print to see if it's better behaved now.
RE: My turn getting a Core 1+ working
Fortunately, the feet seem to be sticking quite well. I did make it a point to clean the surface with an alcohol wipe and let it dry prior to installing the rubber feet.
The adhesive does not detach from the metal frame -- it detaches from the rubber part which it came pre-applied to! Maybe Prusa have switched to a different brand or batch in the meantime; fingers crossed.
RE: My turn getting a Core 1+ working
No problems with my test print! Yes, there was some thumping a couple times, but it was happy quickly. I'm printing off OctoPi. Seems OctoPi thinks there's an issue with chamber temperature control, saying that it's not supported. I may not have enabled it. There was a delay of nearly two minutes after that message, before the printer started doing anything, even though it could have done something. I don't know if that's an OctoPi issue, or Core 1+.
The dresser is much more stable than the dining room table, no swaying or obvious movement. So far so good.
For the Buddy camera, where is it installed? I don't have one, but instead would like to install an RPI camera. Just looking for a spot to install it. I have one running with OctoPi, just need to find a spot to hang it.
RE: My turn getting a Core 1+ working
Wouldn't let me edit. It was an OctoPi printer configuration issue, I had failed to check the correct box. Fixed that.
Anyways, the print came out great. No complaints from the printer not being able to find home. Too much banging, but much better than before.