RE: Tension pulley broken
Support told me to stick to the old procedure 85x85hz, as of yesterday. The online app has been prematurely updated.
Interesting news. That might explain why there was no formal announcement of the new approved tension, and why the native phone app hasn't been updated to match. Do you think a sudden influx of requests for replacement tensioner pulley parts has given them pause for thought?
Posted by: @print_fandango-2
Second, I would also like to reprint these brackets, only have polymaker PCCF, and I am not sure if this is adequate filament... Here in Canada, prusament filament is too expensive.
I don't see why it wouldn't be suitable. The requirement is for a material that doesn't 'creep' under tension, and is strong enough not to break under tension.
RE:
Due to the small size of the parts, I'd recommend using *many* perimeter as well as 100% infill. It was also noted in another thread that the internal design of the tensioners may have been altered by Prusa sometime after release to create more support around the square nut slot. They appear to have added a tiny void that forces *internal* perimeters. None of this is necessary if you print the part solid. I find that more perimeters on parts like this produce more continuous extrusions, and likely more part strength than using fewer perimeters, even if printing solid parts. I'd use something like 8 perimeters combined with solid infill. The amount of material used per part will still be minimal.
It's important to note that prt strength makes little difference if the screw and square nut lock together. To prevent this I decided to remove my tensioners, carefully replace the square nuts with smoothly turning new ones, and then apply anti-seize to the screw threads before installing. Wipe up the extra before installing the screw because that stuff is messy and is hard to clean up after the fact.
-J
RE: Tension pulley broken
Support said that more news will be posted soon but to not follow the online app.
Just not sure what Polymakers PCCF temp tolerance is good, and since it is quite annoying to go through belt calibrations, I just was not sure...
RE: Tension pulley broken
I can't find any information about Polymaker PCCF - are you sure that's what it is?
RE: Tension pulley broken
you are right, no idea where I got this from...I was dreaming. I just checked the box. It is actually polymax (polymaker) PC. No carbon fiber on this one.
RE: Tension pulley broken
I'm no expert. I'd guess you'd have to compare the various properties - Polymaker have a useful tool for that on their web page. My guess is that the 'CF' is important in this application, for strength, creep and heat resistance.
Prusa PC-CF is expensive everywhere! On the bright side, these little pulley holders don't use many grams!
RE: Tension pulley broken
I'm no expert. I'd guess you'd have to compare the various properties - Polymaker have a useful tool for that on their web page. My guess is that the 'CF' is important in this application, for strength, creep and heat resistance.
Prusa PC-CF is expensive everywhere! On the bright side, these little pulley holders don't use many grams!
Support advised any PC or Nylon to me…although they might’ve just been trying to avoid sending me out any PCCF parts.
RE: Tension pulley broken
Nylon? I am surprised. It is strong but maybe too flexible?
RE: Tension pulley broken
Not sure mate.
I don’t have any Nylon or PC so I suggested we ‘meet in the middle’ and I’ll print the part if they send me roll of Prusa PCCF, but they declined and agreed to send the part.
It was worth a try 😂
RE: Tension pulley broken
Haha - nice try!
RE:
This is my analysis - I've posted it also on Reddit.
I will be attempting to redesign the tensioner pulley as it doesn't fit a normal square nut and it also cams out of tensioner rail.
This is a description that somehow wouldn't upload...Hopefully now it'll be clear.
I just spent whole Sunday battling the issue that should never occur, and I would greatly appreciate your thoughts and suggestions on the matter.TLDR: Core one belt tensioning is garbage, let's try to fix it.
In my 8 years of using Prusa printers I've never seen anything like it, so I've decided to write this rantalisys.
The issue:
My Core One (Mk4 upgrade) had issues with calibration and homing since the beginning. I've managed to fix them raising the belt tension to the level that was appropriate in my opinion (around 120Hz). It helped (no errors, no VFA), but I was certain that it will decrease the longevity of the CoreXY printed parts.I was however surprised by how little time it took for the belt tensioning mechanism to experience an RUD.What was even more surprising, it weren't the printed parts that failed.What failed is the M3x30 DIN912 screw and DIN 562 nut.And this is where the issue lies: the screw is too small, and DIN562 nut is a "low" one, which means it only has a few turns of the thread.
The failure mode(s):
When I experienced the failure, I did what every reasonable person would do: I've attempted to rebuild the part using my enormous stash of screws and filaments. Fortunately, I have many 3D printers and I was able to print the damaged parts in PC-CF (I keep a roll or two for such occasions).Unfortunately, my standard nozzle for filled materials (0.6mm) is TOO BIG to print those parts.The space for the square nut was much too tight and even precise usage of ultrasonic knife didn't solve the issue.
I was forced to use a regular 0.4mm brass nozzle to make the prints (again).In all, I've printed the pulleys six times. One set on 0.6 (tolerances off), one set on 0.4mm (tolerances off), one set on 0.4mm but this time I've used a part with looser tolerances I've found on Printables: https://www.printables.com/model/1348184-belt-tensioner-pulley-r1-more-tolerance/files
This is how it fails:
Five of those parts failed in various ways and in various moments.I've separated those failures into two groups:
1. part failure
2. metal part failure
Ad.1
Due to the DIN562 M3 square nut being tiny, and the forces that it sees being relatively large it has a tendency to spin in the pulley socket. This is something that Prusa evidently knew, because the socket is designed to be very tight, what I've noticed in the original parts.- Second failure mode is damage to the thin wall of the pulley part- only 2mm thick in this place.
I've experienced this failure three times.
Ad.2
The standard M3 thread has a pitch of only 0.7mm which makes it prone to crossthreading, especially if the nut doesn't sit perfectly perpendicular to the screw line. This casues seizing of the screw and nut, and then there are two failure modes: first one is a failure of the plastic part, the second one is screw being destroyed by the shearing force from the twisting momentum.
I've experienced three failures in this mode.
Conclusion:
The parts as they are now (E2 and R1 versions) are not designed in accordance to best design practices for mass production 3D printing. There are several ways this assembly can be improved. Here are few ideas from me (I will apply them as soon as possible, because I need my Core One).
DIN912 M3 screw and DIN562 M3 nut (low) can be replaced with readily available high carbon steel variants. Unfortunately only the screw is availabe in high steel grades (10.9, 12.9) off the shelf. It would be perfect if they were both in 10.9 or even 12.9 grade. That should increase their resistance to cross threading and breaking quite aubstantially.The second way is to make the belt tensioners and belt tensioner pulleys from steel. And this is probably the approach I will take next. Why not aluminum? Frequent belt tensioning regulation will damage the threads in aluminum so fast, that it wouldn't be practical. Also, I don't like the idea of aluminum parts in a chamber that is heated, even passively.The third way is to redesign the belt tensioner completely. Not something that I'd like to take on at this moment, due to the lack of time (I'm actively working on the prototype of an industrial CoreXY printer I plan to sell in some time).
RE: Tension pulley broken
I'm not convinced that the nut and screw are cross-threading. This would normally happen from the very end of the screw, where it first enters the nut, but on the CORE One tensioner it typically seems to happen when the nut is some distance along the thread already. I believe the problem is due to galling - some pickup of the surface of the nut or bolt that effectively 'welds' the nut to the bolt. As noted, some lubrication between the nut and the bolt may help this, as may the use of a dissimilar metal for either the nut or the bolt - a brass insert instead of the nut is a good choice because brass has some minor self-lubricating properties.
I think you are right about the small number of threads in the square nut - the tension in the belt has to be carried by these few threads, so the force on each thread is high. Your use of 120Hz to help with VFAs sounds interesting, but that means that the tensile force is twice that of the original setting (85Hz), and this will increase the tendency for the stainless steel to gall. This is another way that a brass insert can help, since the thread length is roughly three times as long, so the force is distributed over more turns of thread.
A CNC machined replacement sounds interesting. I'd also be concerned about using aluminium, since that's another material that can gall.
RE: Tension pulley broken
I was forced to use a regular 0.4mm brass nozzle to make the prints (again).In all, I've printed the pulleys six times. One set on 0.6 (tolerances off), one set on 0.4mm (tolerances off), one set on 0.4mm but this time I've used a part with looser tolerances I've found on Printables: https://www.printables.com/model/1348184-belt-tensioner-pulley-r1-more-tolerance/files
Did you try reducing the extrusion multiplier for the Prusa PCCF? For some reason the default profile has an EM value of 1.04, which seems to produce oversized parts. I used 1.00 with some success, and could even have gone lower.
RE: Tension pulley broken
I also am skeptical on cross-threading, except I've seen some screws out of this failure that absolutely looked like that's what happened. It's likely many potential failure modes.
I think this should be a larger diameter, and also agree on more thread in the nut being a good way to go especially for m3. McMaster has brass square inputs in standard (no metric) but same thin profile. Still, I'm tempted to try 6-32 (3.5mm) or even 8-32 if I think I can make it fit. Or just get around to having the tools for inserts...
On the other hand, anti-seize should also work well.
Aluminum threads I'd stay away from, that can do worse than gall with stainless... Maybe machined brass though, threaded through.
RE:
Aluminum threads I'd stay away from, that can do worse than gall with stainless....
Agreed. Just ask anyone who's tried using an aluminum seat post in a steel or titanium bike frame. 😲
RE: Tension pulley broken
I suggested we ‘meet in the middle’ and I’ll print the part if they send me roll of Prusa PCCF, but they declined and agreed to send the part.
It was worth a try 😂
Nice try! But given the fact that a full roll of PCCF costs much more than the little parts, and is more expensive to ship as well, the chances were slim. 🙂
Maybe you could have argued that the roll would cover you for the next rounds of broken tensioners as well, until Prusa finally comes up with an improved design...
RE:
I also am skeptical on cross-threading, except I've seen some screws out of this failure that absolutely looked like that's what happened. It's likely many potential failure modes.
I think this should be a larger diameter, and also agree on more thread in the nut being a good way to go especially for m3. McMaster has brass square inputs in standard (no metric) but same thin profile. Still, I'm tempted to try 6-32 (3.5mm) or even 8-32 if I think I can make it fit. Or just get around to having the tools for inserts...
On the other hand, anti-seize should also work well.
Aluminum threads I'd stay away from, that can do worse than gall with stainless... Maybe machined brass though, threaded through.
I think a longer thread might help. The part geometry of the tensioner block, with the idler and belt displaced sideways from the nut and screw, puts a lot of lateral and tilting force on the nut and its threads.
The brass heat-inserts which some users have installed are a promising solution in my opinion: A longer thread to cope with the tilting force, and a material which is less likely to gall.
RE: Tension pulley broken
The pulley holder itself seems to hold up fine. Especially the new variant with the small cavities that force the area around the square but to be solid. I don't see a reason to make the pulley holder from a different material.
I agree with the thread size and thread length, though. Those should be increased. Easiest step would be to use a longer M3 thread, either a brass insert or a different, longer, nut. Add in some grease on the threads and I suspect you will have a sufficiently working assembly.
RE: Tension pulley broken
I think a longer thread might help. The part geometry of the tensioner block, with the idler and belt displaced sideways from the nut and screw, puts a lot of lateral and tilting force on the nut and its threads.
This is my feeling too. The actual movement is constrained by the length of screw passing through the main mounting block, but there will still be a lateral force between the screw and nut, which can only increase the friction force between parts of the nut and screw.
Posted by: @jurgen-7
The brass heat-inserts which some users have installed are a promising solution in my opinion: A longer thread to cope with the tilting force, and a material which is less likely to gall.
I used a 5.7mm threaded insert in my model, which is more than 3x the thread length of the square nut. My feeling is that this is ample. You may be able to squeeze another mm or so in, but any longer and the insert will either encroach into the pulley space or reduce the shoulder depth. I guess the whole part could be elongated to take a longer insert, but I don't believe it's necessary.
RE:
Galling is the word I was missing, thank you! In my opinion that's the mechanism I was trying describe.
That's main reason, why I suggested CNC milling those parts from steel, not from aluminum.
Another idea is using steel for tensioner and phosphor bronze for the pulley. This will create a good self lubricating effect.
I will redesign the part so it'd take inserts well (we're doing thousands of those in SLS and FDM).
I also have Siraya Tech Fast Mecha and even some POM filament lying around... Maybe I will give those a go?
If the parts will be milled out of metal, it would allow for some dovetail that will make camming the pulley out from the tensioner less likely.
I did tolerance test before printing the tensioner, and 0.05 was very tight and 0.0 doesn't fit, so when I saw "out of tolerance" dimensions on the pulley, I was convinced it's by design. I will give your suggestion a try.