X and Y belt problems
Posted by threedyprinter
|
X and Y belt problems July 11, 2013 12:27PM |
Registered: 10 years ago Posts: 90 |
Chris,
I chipped in with neptunier's X axis problems yesterday [forums.reprap.org] saying my X clamps had failed.
I've been trying to fix that problem today and in so doing I've discovered there's a related problem with the Y axis.
To recap:
When I built the machine, both X and Y carriages moved smoothly with no tightspots at all. The belts twanged well, and ran dead centre in their idlers and pulleys.
I started getting small X errors. I lived with it for a day, then started getting occasional 2,5mm errors. In fixing that problem the X motor front clamps failed. The X belt was slightly damaged. Looking at the profile through a loup I can see the step in the teeth walls where the tight spot occurred are slightly rounded where it climbed up onto the pulley teeth. I put all this down to the clamp failure and the belt not running centrally in the pulley.
Today I discovered exactly the same problem has developed with the Y belt/pulley only this time the belt still runs perfectly true in the centre of both idler and pulley. It still twangs nicely too. As the bed moves to the front just past Y0, say Y=-20, the belt starts riding the pulley. Everywhere else is smooth.
I cannot see obvious damage to the Y belt or pulley (but there's a fair amount of aluminium black on the belt). It's almost as though someone had suddenly changed the belt pitch.
I suspect that the same problem has occurred in both axes. In X, the extra tension caused the front clamp to slip, making the belt run out of true. In Y, there's only the rear idler to slip (which it hasn't yet) so the belt will just keep riding the pulley.
I'm a bit stuck; I don't want to waste cash on replacement belts because I don't really think they're the problem. I'd rather really understand what's going on first.
Edited 1 time(s). Last edit at 07/11/2013 12:27PM by threedyprinter.
I chipped in with neptunier's X axis problems yesterday [forums.reprap.org] saying my X clamps had failed.
I've been trying to fix that problem today and in so doing I've discovered there's a related problem with the Y axis.
To recap:
When I built the machine, both X and Y carriages moved smoothly with no tightspots at all. The belts twanged well, and ran dead centre in their idlers and pulleys.
I started getting small X errors. I lived with it for a day, then started getting occasional 2,5mm errors. In fixing that problem the X motor front clamps failed. The X belt was slightly damaged. Looking at the profile through a loup I can see the step in the teeth walls where the tight spot occurred are slightly rounded where it climbed up onto the pulley teeth. I put all this down to the clamp failure and the belt not running centrally in the pulley.
Today I discovered exactly the same problem has developed with the Y belt/pulley only this time the belt still runs perfectly true in the centre of both idler and pulley. It still twangs nicely too. As the bed moves to the front just past Y0, say Y=-20, the belt starts riding the pulley. Everywhere else is smooth.
I cannot see obvious damage to the Y belt or pulley (but there's a fair amount of aluminium black on the belt). It's almost as though someone had suddenly changed the belt pitch.
I suspect that the same problem has occurred in both axes. In X, the extra tension caused the front clamp to slip, making the belt run out of true. In Y, there's only the rear idler to slip (which it hasn't yet) so the belt will just keep riding the pulley.
I'm a bit stuck; I don't want to waste cash on replacement belts because I don't really think they're the problem. I'd rather really understand what's going on first.
Edited 1 time(s). Last edit at 07/11/2013 12:27PM by threedyprinter.
|
Re: X and Y belt problems July 11, 2013 06:28PM |
Admin Registered: 17 years ago Posts: 7,879 |
Does the belt ride the pulley because the axis jams at one spot so there is a lot of force on it, or is there a duff patch on the belt?
With the power off does the belt ride up when you move the carriage? If it does that implies the belt has a bad batch. If it only does it when the motor is driving that implies the axis has a point where it binds.
[www.hydraraptor.blogspot.com]
With the power off does the belt ride up when you move the carriage? If it does that implies the belt has a bad batch. If it only does it when the motor is driving that implies the axis has a point where it binds.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 11, 2013 07:49PM |
Registered: 10 years ago Posts: 90 |
>With the power off does the belt ride up when you move the carriage?
Yes.
All the belt is duff!
I've just looked at page 153 in the Brecoflex pdf you mentioned to neptunier and realised that the sealed edge is missing on one side of both X and Y belts.
One edge is smooth and connects all the teeth nicely. On the other side there is no smooth edge at all for the whole run of the belt. No tooth is connected to its neighbour or the belt itself. I thought that was the way it was supposed to be, not having played with such things before, but now I look it's obviously a manufacturing fault - as though a trimming blade was set too close perhaps.
Certainly explains the problems I've had IMHO.
Sorry I can't get a decent close-up of it to post.
Edited 1 time(s). Last edit at 07/11/2013 08:05PM by threedyprinter.
Yes.
All the belt is duff!
I've just looked at page 153 in the Brecoflex pdf you mentioned to neptunier and realised that the sealed edge is missing on one side of both X and Y belts.
One edge is smooth and connects all the teeth nicely. On the other side there is no smooth edge at all for the whole run of the belt. No tooth is connected to its neighbour or the belt itself. I thought that was the way it was supposed to be, not having played with such things before, but now I look it's obviously a manufacturing fault - as though a trimming blade was set too close perhaps.
Certainly explains the problems I've had IMHO.
Sorry I can't get a decent close-up of it to post.
Edited 1 time(s). Last edit at 07/11/2013 08:05PM by threedyprinter.
|
Re: X and Y belt problems July 11, 2013 08:14PM |
Admin Registered: 17 years ago Posts: 7,879 |
I am not sure what you mean by the edge missing.Do you mean the wire is exposed?
The belt is the solid all the way across the profile except where it is punctuated by the wires. Even if the edge was missing it should just behave like a 5mm belt. The teeth are connected to each other all the way across. There is nothing special about the edges, they are just the same as the gaps between each wire. Brecoflex seems to be different in this respect. It seems to have edge sections where there are no wires.
[www.hydraraptor.blogspot.com]
The belt is the solid all the way across the profile except where it is punctuated by the wires. Even if the edge was missing it should just behave like a 5mm belt. The teeth are connected to each other all the way across. There is nothing special about the edges, they are just the same as the gaps between each wire. Brecoflex seems to be different in this respect. It seems to have edge sections where there are no wires.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 11, 2013 09:13PM |
Registered: 10 years ago Posts: 90 |
I wish I could take a photo of it!
Imagine looking at the belt in profile. You have the belt at the bottom and the teeth on top. On the good side you see one continuous profile.
On the bad side you see the belt at the bottom, then a gap, then the teeth. You can get your nails in the gap a mm or so and wobble the edges of the teeth around a bit.
All the teeth are firmly attached to the belt (except at the bad edge). There are four rows of wires but they are not central in the belt (which I thought strange?). They are evenly spaced but closer to the good edge than the bad.
It looks to me as though there must be more tension on the good edge of the belt than on the bad which is definitely not desirable.
I'm not sure if that makes things any clearer Chris!
Imagine looking at the belt in profile. You have the belt at the bottom and the teeth on top. On the good side you see one continuous profile.
On the bad side you see the belt at the bottom, then a gap, then the teeth. You can get your nails in the gap a mm or so and wobble the edges of the teeth around a bit.
All the teeth are firmly attached to the belt (except at the bad edge). There are four rows of wires but they are not central in the belt (which I thought strange?). They are evenly spaced but closer to the good edge than the bad.
It looks to me as though there must be more tension on the good edge of the belt than on the bad which is definitely not desirable.
I'm not sure if that makes things any clearer Chris!
|
Re: X and Y belt problems July 11, 2013 09:26PM |
Admin Registered: 17 years ago Posts: 7,879 |
Yes I can see that on the belt here but I don't see how it makes any difference. I think it is just random where the wires are. In the early days of reprap we used to slice wider belts up with a belt splitting jig because it was cheaper. The wires came out the side then but they still worked fine. I don't see why the wires being offset would cause the belt to climb out of the pulley. When it is under tension it is hard to imaging it climbing out unless there is a gross pitch mismatch.
[www.hydraraptor.blogspot.com]
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 11, 2013 09:39PM |
Registered: 10 years ago Posts: 90 |
It's impossible for me to see exactly where/when the climb out starts. My X axis is in pieces, on the Y axis the motor gets in the way one side of the pulley and the flange of the pulley obscures the other - I have to use an eye glass to see anything useful so it's a bit difficult. Added to which, I've pulled and twisted the belt around so much today that it's just started behaving itself properly again. I have no idea why, the tight spot/climb out just vanished. That also happened last week when the X axis started misbehaving.
|
Re: X and Y belt problems July 11, 2013 10:21PM |
Admin Registered: 17 years ago Posts: 7,879 |
Well all I can do is send you a new pair of belts and X ends tomorrow (actually today now!) but the belt is the same in that it has one rough edge and one smooth. I am not sure if it is the same 100m spool, or the one before.
If simply sliding the carriage makes it skip teeth then either there is next to no tension on it or there is something very wrong but I can't think what. A slight pitch mismatch would not cause it because only 8 teeth are engaged at one time so you would need an error that was significant over only 8 teeth. If the belt had a major defect like that why would it suddenly get better.
Another possibility is it climbs up the rim of the pulley for some reason. You could maybe loosen the Y motor bracket and angle it a bit to try to get it to run more central to the pulley. If it is caused by the rough side of the belt against the rim you could try reversing it.
[www.hydraraptor.blogspot.com]
If simply sliding the carriage makes it skip teeth then either there is next to no tension on it or there is something very wrong but I can't think what. A slight pitch mismatch would not cause it because only 8 teeth are engaged at one time so you would need an error that was significant over only 8 teeth. If the belt had a major defect like that why would it suddenly get better.
Another possibility is it climbs up the rim of the pulley for some reason. You could maybe loosen the Y motor bracket and angle it a bit to try to get it to run more central to the pulley. If it is caused by the rough side of the belt against the rim you could try reversing it.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 12, 2013 05:49AM |
Admin Registered: 17 years ago Posts: 7,879 |
My supplier says that it is inevitable to have a rough edge for part of a reel because the wires are actually a spiral when the belt is made and then cut up. It must be a massive machine to make 100m belts in a cylinder!
I think I will switch to Brecoflex as it is a much nicer product but it still doesn't explain why you have a problem with this. Please you can return the belts you have so I can try to work out what the actual problem is.
[www.hydraraptor.blogspot.com]
I think I will switch to Brecoflex as it is a much nicer product but it still doesn't explain why you have a problem with this. Please you can return the belts you have so I can try to work out what the actual problem is.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 12, 2013 07:49AM |
Registered: 10 years ago Posts: 90 |
>If simply sliding the carriage makes it skip teeth then either there is next to no tension on it or there is something very wrong but I can't think what.
Exactly! I spent a long time getting the axes lined up correctly. On the Y I found it quite challenging to position the idler with enough tension, the belt smack in the middle and not have it wandering side to side in use. I did manage it though. It ran in the middle of the motor pulley too - also with negligible wander.
The tension was double bass rather than violin. I increased it slightly later to control the degree of vertical belt displacement during fast infills. There's absolutely no way the belt could climb out - unless the tension is uneven across the width of the belt.
>Another possibility is it climbs up the rim of the pulley for some reason.
Honest guv - it never went near the flange. I knew that would cause problems which is why I spent much time getting it central in the pulley and idler.
I'm thinking this is what happened:
The bad, discontinuous, belt edge must be under fractionally less tension than the continuous edge.
This allows enough movement for the corners of the loose edges of the teeth to gain purchase on the rough milling of the pulley teeth - fractionally higher than they should (as opposed to sitting neatly in the valley floor if you see what I mean). (In my terms there's a constant and cumulative phase lag between tooth engagement on the bad side and the good side.) This would start the climb out.
> If the belt had a major defect like that why would it suddenly get better.
I really don't know. I've not adjusted anything.
Straw clutching: Maybe my sweaty hands lubricated the belt teeth a bit so they settle in the pulley teeth better.
>Well all I can do is send you a new pair of belts and X ends tomorrow (actually today now!)
Thank you. And you work far too late; but it's much appreciated!
>Please you can return the belts you have so I can try to work out what the actual problem is.
Certainly, would you let me have your address please?
Exactly! I spent a long time getting the axes lined up correctly. On the Y I found it quite challenging to position the idler with enough tension, the belt smack in the middle and not have it wandering side to side in use. I did manage it though. It ran in the middle of the motor pulley too - also with negligible wander.
The tension was double bass rather than violin. I increased it slightly later to control the degree of vertical belt displacement during fast infills. There's absolutely no way the belt could climb out - unless the tension is uneven across the width of the belt.
>Another possibility is it climbs up the rim of the pulley for some reason.
Honest guv - it never went near the flange. I knew that would cause problems which is why I spent much time getting it central in the pulley and idler.
I'm thinking this is what happened:
The bad, discontinuous, belt edge must be under fractionally less tension than the continuous edge.
This allows enough movement for the corners of the loose edges of the teeth to gain purchase on the rough milling of the pulley teeth - fractionally higher than they should (as opposed to sitting neatly in the valley floor if you see what I mean). (In my terms there's a constant and cumulative phase lag between tooth engagement on the bad side and the good side.) This would start the climb out.
> If the belt had a major defect like that why would it suddenly get better.
I really don't know. I've not adjusted anything.
Straw clutching: Maybe my sweaty hands lubricated the belt teeth a bit so they settle in the pulley teeth better.
>Well all I can do is send you a new pair of belts and X ends tomorrow (actually today now!)
Thank you. And you work far too late; but it's much appreciated!
>Please you can return the belts you have so I can try to work out what the actual problem is.
Certainly, would you let me have your address please?
|
Re: X and Y belt problems July 12, 2013 08:31AM |
Admin Registered: 17 years ago Posts: 7,879 |
Quote
I'm thinking this is what happened:
The bad, discontinuous, belt edge must be under fractionally less tension than the continuous edge.
But this is normal for this type of belt. All the machines I have run myself and all the kits I have sold have this type of spiral reinforced belt. Brecoflex advertise having clean edges as an advantage over their competitors but that doesn't imply their's is the only belt that works.
The way I see it is the effective part of the belt that takes the tension is the wires. Having a bit of extra polyurethane flapping about on one edge should not make any difference unless it rubs against something.
So far there are only three people with problems with the belt, which is about 1%. They are all recent though. I have started using pulleys from a different source so the problem could lie with them.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 12, 2013 09:25AM |
Registered: 10 years ago Posts: 90 |
>The way I see it is the effective part of the belt that takes the tension is the wires.
Sadly, I have to agree with you!
>pulleys
Were the belt to run against the end flange that would definitely have given problems. I had to scrape out a few bits of swarf sandwiched between the flange and the teeth. Running in the middle though ought to be fine. I can see the milling lines on the teeth but they're not that bad to my untrained eye.
The belts are ready for return. Please email your address.
Sadly, I have to agree with you!
>pulleys
Were the belt to run against the end flange that would definitely have given problems. I had to scrape out a few bits of swarf sandwiched between the flange and the teeth. Running in the middle though ought to be fine. I can see the milling lines on the teeth but they're not that bad to my untrained eye.
The belts are ready for return. Please email your address.
|
Re: X and Y belt problems July 12, 2013 09:31AM |
Admin Registered: 17 years ago Posts: 7,879 |
|
Re: X and Y belt problems July 12, 2013 12:47PM |
Registered: 10 years ago Posts: 90 |
|
Re: X and Y belt problems July 12, 2013 01:34PM |
Admin Registered: 17 years ago Posts: 7,879 |
Thanks.
We posted a couple of X ends and a pair of Brecoflex belts. I haven't tried it myself yet but it is much more flexible. There are more wires but they are thinner and stay away from the edges. If it still doesn't work then it must be something other than the belt and if it does now work it must be something to do with the old belt.
I have had a small roll in a drawer for a year or so. Having not unrolled it before I didn't realise there could be so much difference in T2.5 steel cored belts. It is a lot more expensive though.
[www.hydraraptor.blogspot.com]
We posted a couple of X ends and a pair of Brecoflex belts. I haven't tried it myself yet but it is much more flexible. There are more wires but they are thinner and stay away from the edges. If it still doesn't work then it must be something other than the belt and if it does now work it must be something to do with the old belt.
I have had a small roll in a drawer for a year or so. Having not unrolled it before I didn't realise there could be so much difference in T2.5 steel cored belts. It is a lot more expensive though.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 12, 2013 07:20PM |
Registered: 10 years ago Posts: 90 |
|
Re: X and Y belt problems July 14, 2013 07:13AM |
Registered: 10 years ago Posts: 90 |
I received and fitted the ends and belts yesterday. (Thanks for superb service!)
The difference in belts is surprising; let's face it, they're chalk and cheese!
I noticed:
Tensioning X was a breeze; just a turn on the tensioner got the belt twanging really nicely. Three clearly audible notes and the belt is nicely tangential to the pulley. (I had to put much more tension on with the old belt, and even then it did not twang so nicely.) It runs centrally in the idler but is too close to the pulley flange; not touching but too close. I shall fix this.
Tensioning Y was similarly easy. I have the idler at it's tightest position (end of the slot) and it could do with a tiny bit more but it runs so true and smooth that I'm not going to change it.
There's no aluminium black on either belt yet - after running for ~3hrs. Whether it's significant I don't know. I only noticed the black on the old belts after trouble started (but that's not to say it wasn't there earlier).
All in all, the whole machine seems much happier now; a quick test print was very, very good. It's never produced such consistent perimeters before. In fact the X,Y positioning is so good that I now have to revisit slicer settings to correct errors that weren't noticeable previously.
Do let me know your thoughts when the old belts arrive.
The difference in belts is surprising; let's face it, they're chalk and cheese!
I noticed:
Tensioning X was a breeze; just a turn on the tensioner got the belt twanging really nicely. Three clearly audible notes and the belt is nicely tangential to the pulley. (I had to put much more tension on with the old belt, and even then it did not twang so nicely.) It runs centrally in the idler but is too close to the pulley flange; not touching but too close. I shall fix this.
Tensioning Y was similarly easy. I have the idler at it's tightest position (end of the slot) and it could do with a tiny bit more but it runs so true and smooth that I'm not going to change it.
There's no aluminium black on either belt yet - after running for ~3hrs. Whether it's significant I don't know. I only noticed the black on the old belts after trouble started (but that's not to say it wasn't there earlier).
All in all, the whole machine seems much happier now; a quick test print was very, very good. It's never produced such consistent perimeters before. In fact the X,Y positioning is so good that I now have to revisit slicer settings to correct errors that weren't noticeable previously.
Do let me know your thoughts when the old belts arrive.
|
Re: X and Y belt problems July 14, 2013 08:16AM |
Admin Registered: 17 years ago Posts: 7,879 |
I got the old belts yesterday thanks. Just running them by hand over a pulley with no flange mounted on a shorted out stepper motor I can't find anything wrong with them. The teeth don't ride out provided there is enough tension. The points you marked are both near the end, which doesn't get as much wear as the middle. It is almost as if the tension drops as the axis moves but that should not be the case as the belt paths should be perfectly straight on Mendel90.
They do seem quite stiff but it is the same belt I have been using since I switched to T2.5 in December. The first 20 kits used T5 with kevlar, which is very flexible.
I have a machine running here with the same T2.5 belts supplied in the kits and it works well.
I have never worried about the belt riding against the rim of the pulley as it is rotating with the belt, so doesn't add any friction, unlike the idler washers. Most of my machines run like that.
I do get black from the aluminium over time. I presume it oxidises and the belt picks up the oxide causing it to oxidise further. I expect anodized pulleys would be better in this respect.
So it's still a mystery why they didn't work for you. I will have to build them into a machine to be sure but it may be some time before I can do that.
[www.hydraraptor.blogspot.com]
They do seem quite stiff but it is the same belt I have been using since I switched to T2.5 in December. The first 20 kits used T5 with kevlar, which is very flexible.
I have a machine running here with the same T2.5 belts supplied in the kits and it works well.
I have never worried about the belt riding against the rim of the pulley as it is rotating with the belt, so doesn't add any friction, unlike the idler washers. Most of my machines run like that.
I do get black from the aluminium over time. I presume it oxidises and the belt picks up the oxide causing it to oxidise further. I expect anodized pulleys would be better in this respect.
So it's still a mystery why they didn't work for you. I will have to build them into a machine to be sure but it may be some time before I can do that.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems July 22, 2013 07:38AM |
Registered: 10 years ago Posts: 90 |
Just a quick update now that I've been running the Brecoflex belts for a few days.
Every print has been spot on. Belts are running straight and true, not a hint of climbing out of the pulleys and no blacking evident on the belts at all.
I've noticed one thing; buzzing. At certain speeds there is a metallic "bzzzzt" sound (guess about 15-20Hz, very sawtoothy!). I thought it was just a loose fixing or the foil heat reflector resonating with the motor; but everything is held down firmly. Looking at the machine you would swear it was coming from the Y motor but, sticking a screwdriver on the motor with the handle against my ear, that doesn't seem to be the case. (It's difficult to tell with all the other normal noises going on.)
It's not a big deal, the prints don't suffer at all but for the life of me I can't track the source down, it certainly wasn't there before changing the belts. Might the more compliant Brecoflex belts be causing resonance? If so, can you send me a box of poles and zeros to tune it out?
Every print has been spot on. Belts are running straight and true, not a hint of climbing out of the pulleys and no blacking evident on the belts at all.
I've noticed one thing; buzzing. At certain speeds there is a metallic "bzzzzt" sound (guess about 15-20Hz, very sawtoothy!). I thought it was just a loose fixing or the foil heat reflector resonating with the motor; but everything is held down firmly. Looking at the machine you would swear it was coming from the Y motor but, sticking a screwdriver on the motor with the handle against my ear, that doesn't seem to be the case. (It's difficult to tell with all the other normal noises going on.)
It's not a big deal, the prints don't suffer at all but for the life of me I can't track the source down, it certainly wasn't there before changing the belts. Might the more compliant Brecoflex belts be causing resonance? If so, can you send me a box of poles and zeros to tune it out?
|
Re: X and Y belt problems July 22, 2013 08:09AM |
Admin Registered: 17 years ago Posts: 7,879 |
Quote
Every print has been spot on. Belts are running straight and true, not a hint of climbing out of the pulleys and no blacking evident on the belts at all.
Great but that is what I get with the old bed, apart from the blacking. I think that is simply because your old belt cleaned up your pulleys.
I am waiting for a delivery of rods. When I get that I will build a machine with your belts.
Yes rattles can be very hard to track down. Handles on the bed clips can be a source. Also the strain relief clip on the extruder connector. I also think the foil can be a source. I think it probably isn't needed. I suspect just filling the gap with corrugated card will work just as well. I also have machines with Dibond carriages and no heatshield that work fine.
[www.hydraraptor.blogspot.com]
|
Re: X and Y belt problems December 31, 2013 11:40AM |
Registered: 10 years ago Posts: 1 |
I have exactly same kind of problems with my x and y belts. At first there was no problems, but after I had totally printed about 400 grams of filament the x-belt started toget 2.5mm errors and soon after that also y-belt. I tried to adjust belt tightening with no success. I looks like the belts are streched. See video: [www.youtube.com]
Where can I buy new belts?
BR Jukka
Where can I buy new belts?
BR Jukka
Sorry, only registered users may post in this forum.