RE: Tension pulley broken
And this is why official answers are not fired off every time an angry internet someone or other insists they're taking too long to respond. Though in theory they don't ask an AI whose purpose is to make the user satisfied, not answer correctly...
If you properly tap the plastic (vs forcing a screw through) it'll pull exactly the way you want till it starts to melt and everything goes sideways... but we're even further into tools we all possibly should have but might not.
RE: Tension pulley broken
And this is why official answers are not fired off every time an angry internet someone or other insists they're taking too long to respond. Though in theory they don't ask an AI whose purpose is to make the user satisfied, not answer correctly...
I told it what I thought of its pathetically inaccurate estimate of the linear mass density. I think it’s learned its lesson. Well, one of us has 🤣
I still don’t see how a tapped hole is going to work in conjunction with the insert though. The threads are likely to be out of phase at least, between the plastic and the insert, even if they are concentric. What am I missing?
RE: Tension pulley broken
I think you'd have to pull it in fully with a screw so it's aligned. But you're right, it's far enough into the weeds and very possibly impossible for anyone to consistently apply at home, which makes having a printed part here pointless. Even the insert may be out of reach for a lot.
On the topic of stress - I imagine the fact there is a bit of moving mass changing direction applied to the problem, a static analysis is likely insufficient as well (not to mention a static test). Need to leave this to someone with the formal engineering chops who wants to dive into the math - I hope my assumption that you're not that person isn't offensive, but the chatgpt use hints in that direction 🙂 I know I certainly don't qualify. This does feel like the kind of classroom exercise I might have once been assigned but that's all long forgotten.
And this is why official answers are not fired off every time an angry internet someone or other insists they're taking too long to respond. Though in theory they don't ask an AI whose purpose is to make the user satisfied, not answer correctly...
I told it what I thought of its pathetically inaccurate estimate of the linear mass density. I think it’s learned its lesson. Well, one of us has 🤣
I still don’t see how a tapped hole is going to work in conjunction with the insert though. The threads are likely to be out of phase at least, between the plastic and the insert, even if they are concentric. What am I missing?
RE: Tension pulley broken
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. 🙂
If you don't buy a ticket, you won't win the raffle 🤣
It does seem this tensioner design is poor...my replacement from Prusa finally got shipped today, I wonder if it's still the same tensioner they're providing or if they've made improvements?
RE: Tension pulley broken
I'll be the first to admit that ChatGPT is often caught 'bullshitting with confidence', but I didn't see anything that stood out as plain wrong.
Blimey if they've finally mastered this them I'm almost definitely out of a job 😂
RE: Tension pulley broken
Blimey if they've finally mastered this them I'm almost definitely out of a job 😂
🤣🤣🤣
RE:
None taken! 😀
Posted by: @cjd
I think you'd have to pull it in fully with a screw so it's aligned.
So that implies a clearance hole, doesn't it? Not a threaded hole?
Posted by: @cjd
On the topic of stress - I imagine the fact there is a bit of moving mass changing direction applied to the problem, a static analysis is likely insufficient as well (not to mention a static test).
Yeah, the dynamic loading changes things, but whilst my education indeed isn't in mechanical engineering, I'm going to guess that a static load test will get quite close to an estimate of the pull-out tension, particularly if, as I suspect, there's a significant factor of safety. But I'm not going to bother with another static test. Although the cross sectional area of the insert is about half that of the square nut, the square nut is not gripped by the surrounding plastic - it relies only on the face pulling against the equivalent of the shoulder. My gut feeling is that the full length of the insert being gripped by the surrounding melted plastic should make up for the reduced cross-sectional area.
RE: Tension pulley broken
Where's that facepalm emoji when I need it... you're right.
So that implies a clearance hole, doesn't it? Not a threaded hole?
RE: Tension pulley broken
🤦♂️
Here you go. 🙂
RE: Tension pulley broken
I've been wondering if the unstable belt tension is somehow related to this issue - purely speculation, but I'm going to swap both of my pulley holders with ones with heat-staked screw inserts. It doesn't take long to try and if that results in stable belt tension that's a data point.
I don't recall any times with my MK3 or MK4 where I had to periodically re-tune the belts, so the idea that the belts are "stretching" seems off. That being said, it's true that these are longer belts with higher tension. Swapping these pulley holders out is simple enough and worth a try.
RE: Tension pulley broken- exactly the same problem
RE: Tension pulley broken
I've been wondering if the unstable belt tension is somehow related to this issue - purely speculation, but I'm going to swap both of my pulley holders with ones with heat-staked screw inserts. It doesn't take long to try and if that results in stable belt tension that's a data point.
I don't recall any times with my MK3 or MK4 where I had to periodically re-tune the belts, so the idea that the belts are "stretching" seems off. That being said, it's true that these are longer belts with higher tension. Swapping these pulley holders out is simple enough and worth a try.
Do you have to repeatedly retune, or was it just once after a bit of a shakedown? It'll be interesting to see what you find with the inserts. I guess it's possible that the nut-in-a-slot could be settling into place, since it's quite a tight fit, whereas the heat-set insert will likely stay where it started.
I'm not really convinced that belt tension makes a great deal of difference to VFAs - I saw a tiny improvement in one speed band with a higher tension, on my GT1.5 setup, but I reckon that if you tune it to spec, then re-check it after a few prints, a small amount of slackening after that isn't going to make much practical difference.
RE: Tension pulley broken
I'm not really convinced that belt tension makes a great deal of difference to VFAs
I agree, and this is what makes me a bit sceptical about the entire notion that tuning the belts in a certain way to a certain frequency will magically solve the issue.
The way that some machines are affected much more than others just points towards a dodgy batch of hardware imo, whether that be pulleys, idlers, belts, or (*shudders) motors I don't know.
I was really hoping after Josef's public statements on the matter we would know something more definite by now.
All we've had so far are the (now not recommended) tighter belts that has led to the destruction of some tensioners 😂
RE: Tension pulley broken
Maybe this seeming "recall" of the higher belt tension recommendation means that they've either found or will at least be more likely to consider a more substantive hardware solution like we're all hoping for.
-J
RE:
I am new to 3d printing so this may be way off but it occurs to me that this belt tensioner is placed at the most stressed position possible where belts turn 180 degrees, belts being stiff with a minimum of stretch all stress ends up on this little threaded nut.
Comparing this to timing belts on engines tensioning timing belts is usually done by pushing the side of the belt, turning points are rigid. Would it maybe be a better design to have this turning point on the printer as one rigid piece and add a tensioner pushing the side of the belt elsewhere? Maybe towards the mid rear?
Shortening the distances of free running belts should also shift resonance frequencies which may have other benefits, but that is for another thread.
/Anders
RE:
This is the lowest parts count and simplest solution. So many things would be easier with an automatic tensioner, but as an open ended belt (and a kit) they'd still need a way to take excess slack out of the system like this setup.
This might be the best/easiest way to apply a damper to the belts though.
RE:
Do you have to repeatedly retune, or was it just once after a bit of a shakedown? It'll be interesting to see what you find with the inserts. I guess it's possible that the nut-in-a-slot could be settling into place, since it's quite a tight fit, whereas the heat-set insert will likely stay where it started.
I'm not really convinced that belt tension makes a great deal of difference to VFAs - I saw a tiny improvement in one speed band with a higher tension, on my GT1.5 setup, but I reckon that if you tune it to spec, then re-check it after a few prints, a small amount of slackening after that isn't going to make much practical difference.
I found that after printing for a while, the belt tension would shift, perhaps by small amounts. But if you've tuned your belts to *exactly* 85 Hz on both sides, you find that there's very small adjustments necessary near the end (small fractions of a turn per side, alternating until both are correct.)
Now that I have pulley holders with the inserts, tuned to spot on 85 Hz, I ran a print last night that ran over two hours, covering the diagonal across almost all of the bed. No change in the tension after that. I'll keep monitoring, but I think your updated pulley holders are potentially worth using.
Watching that print as it ran lower-left to upper-right laying down the first layer with only one of the belts moving was quite fun to watch. Then the other belt when laying down the next layer 90 degrees shifted.
Also: yes, I agree that the belt tension isn't going to be a solution to the VFA issues.
RE: Tension pulley broken
I've updated my 'threaded insert' model for the pulley carrier to include a model suitable for a screw-in (or 'self-tapping') threaded insert. This type of insert doesn't seem as plentiful on Amazon and eBay as the heat-set variety, but it can be inserted without a soldering iron or any other special tools.
https://www.printables.com/model/1360563-core-one-belt-tensioner-pulley-to-take-a-threaded
RE: Tension pulley broken
I'm 8 days into the first of two MK4S->CoreOne upgrades. It sounds like a lot of folks are having the same issue I am.
In trying to do the alignment the first time the the left side had a good deal of play while the right was flush. Of course the order in the build is do all of this AFTER you've put the extruder on, the door sensor, and make the door sensor nearly impossible to get at to remove.
First attempt at getting rid of the play in the left side was tightening up the tensioning screw. This finally did remove the play in the process pulled the belt out of the buckle part the belts are threaded into. This of course required the extruder to be removed to remedy. I'm not sure if I didn't tighten down the buckle part enough or what. So I took the extruder off rethreaded the belt through the buckle and put things back in place. After getting things back together and trying to tighten the left tesioner it appeared to be cross-threaded(what folks in this thread clarified as galling) in the middle of the screw and turning it did nothing. so I decided to print a new tesioner and pulley. The only thing I had on hand was prusament galaxy black petg. Of course to remove the left tensioner after the printer is assembled you also have to remove the door sensor.... I found an allen key that would fit in there and was able to unscrew it. I installed the new part, put the extruder back together and tried it again. Fail. In monkeying with it the right side tensioner screw did the same thing so I had to print another tensioner and pully for that side. I have printed the right side twice now.
What I've observed.
In order to get the left side to have no play, i.e. make it flush with the front, you have to tighten the tensioner screw to to the point the pulley is roughly in the middle of the tensioner channel. But this makes the tension 125hz+
The right side is flush when the pulley is nearly at the end of the channel(towards the back of the printer) and is already tight at that point.
So basically I'm unable to get the alignment AND the frequency at the same time. I have specifically checked that there are 4-5 teeth showing when threaded through the buckle part before attaching to the linear rail. I almost feel like the Y belt needs to be slightly longer so it's more loose at the outset.
Thankfully I have the XL and my other MK4S to work on while this one is on the operating table.
So is PCCF a requirement for these parts?
I have another new upgrade kit in the box that I'm now wondering if I'm going to even use given the difficulties on this one.
RE:
This sounds like your gantry wasn't square before you started the belt tensioning process. Squaring it with belt tension alone is a bit of a kludge that is bound to result in improper tension.
With both belts completely loose you need to adjust the gantry so that there is no play on either side. Bring it to the front, and if there is a gap on one side, place a spacer item, slightly bigger than that gap, between the touching points on the OTHER side, then pull on the side that has the gap so as to bend the brackets that hold the x-rail to the y-rods. Remove the spacer item, recheck, repeat as necessary until both sides touch at the same time. Only then should you begin tensioning the belts.
Gantry to the front, print head to the right, tighten both belts in synchronisation, a little on one side then a little on the other, then measure their frequency. Aim to keep 6Hz difference in the measured frequency as you go, because eventually you want 98Hz on the top one and 92Hz on the bottom one. If you get to the point where a gap opens on one side you'll probably start to see that tightening one side either increases the tension on both belts, or neither. In that case it's simplest just to slacken them both off and start again, and this time go slower!
If you're in this thread I hope you've seen my Printables model for a tensioner pulley holder that uses a brass threaded insert in preference to the stainless steel square nut, to prevent the bolt seizing in the first place.