Exoslide Core XY with "trough" down the middle of the bed?
Posted by dlc60
|
Exoslide Core XY with "trough" down the middle of the bed? May 30, 2022 02:34PM |
I have an odd one. I built a custom CoreXY using a Duet Maestro and Exoslide "rail runners".
I built it using "standard" length extruded aluminum rails so I didn't need to do any custom cutting or finishing.
It uses a standard 200mm square aluminum hotbed with 3mm borosilicate glass for its surface and a Bowden E3Dv6 hot end. The bed clamped on the edges.
Now the weird part.
If I am printing single pieces or relatively large X/Y surface parts it works great.
If I print something with lots of islands, like the "cute octopus" for instance, at some point in the print, an island will take a hit and come dislodged.
I have checked my clearances, trammed the bed, made sure my surface doesn't cantilever, raised my Z-hop to 1mm, and even slowed the printer down to 40mm/s print, 80mm/s travel.
Still, I get an island hit and the print goes sideways. The PLA I am using sticks solidly to the glass, and I hear the "clink" when the strike happens.
I spent some time with my gap gauge at .1mm measuring the nozzle to plat height. At all four corners I will get a setting of 0.1mm and at the center of the bed, 0.15mm. Running with that, I find that I have a 0.5mm "ditch" that runs across the bed at around 100mm X and all along the Y axis.
Huh? The way that I designed this printer is that the hotend carriage travels along the rail in the Y axis and that rail is carried in the X axis.
I have not over constrained any of the Exoslide rollers, and I have very carefully tensioned the belts so that they run as close to perfectly parallel to each other as I can measure them.
So, why does it look like there is a trough down the middle of the plate at Y = (about) 100mm? It appears that when the nozzle leaves the center, shortly thereafter it is not as high as the controller thinks it is and it strikes a island. The strikes always occur along the X axis.
Has anyone ever seen an oddity like this? I'd love another idea of where to look. I am completely baffled and ready to scrap the whole thing or just relegate it to only certain jobs. Here is a picture of it.
Sorry, for some reason the picture got rotated...
oddly,
DLC
Edited 1 time(s). Last edit at 05/30/2022 02:36PM by dlc60.
Kits: Folgertech Kossel 2020 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Maker-Beam XL Micro Delta, 220x220CoreXY.
Actually bought: Bambu Labs X1 Carbon
I built it using "standard" length extruded aluminum rails so I didn't need to do any custom cutting or finishing.
It uses a standard 200mm square aluminum hotbed with 3mm borosilicate glass for its surface and a Bowden E3Dv6 hot end. The bed clamped on the edges.
Now the weird part.
If I am printing single pieces or relatively large X/Y surface parts it works great.
If I print something with lots of islands, like the "cute octopus" for instance, at some point in the print, an island will take a hit and come dislodged.
I have checked my clearances, trammed the bed, made sure my surface doesn't cantilever, raised my Z-hop to 1mm, and even slowed the printer down to 40mm/s print, 80mm/s travel.
Still, I get an island hit and the print goes sideways. The PLA I am using sticks solidly to the glass, and I hear the "clink" when the strike happens.
I spent some time with my gap gauge at .1mm measuring the nozzle to plat height. At all four corners I will get a setting of 0.1mm and at the center of the bed, 0.15mm. Running with that, I find that I have a 0.5mm "ditch" that runs across the bed at around 100mm X and all along the Y axis.
Huh? The way that I designed this printer is that the hotend carriage travels along the rail in the Y axis and that rail is carried in the X axis.
I have not over constrained any of the Exoslide rollers, and I have very carefully tensioned the belts so that they run as close to perfectly parallel to each other as I can measure them.
So, why does it look like there is a trough down the middle of the plate at Y = (about) 100mm? It appears that when the nozzle leaves the center, shortly thereafter it is not as high as the controller thinks it is and it strikes a island. The strikes always occur along the X axis.
Has anyone ever seen an oddity like this? I'd love another idea of where to look. I am completely baffled and ready to scrap the whole thing or just relegate it to only certain jobs. Here is a picture of it.
Sorry, for some reason the picture got rotated...
oddly,
DLC
Edited 1 time(s). Last edit at 05/30/2022 02:36PM by dlc60.
Kits: Folgertech Kossel 2020 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Maker-Beam XL Micro Delta, 220x220CoreXY.
Actually bought: Bambu Labs X1 Carbon
|
Re: Exoslide Core XY with "trough" down the middle of the bed? May 30, 2022 07:31PM |
My guess is that it's due to the bowden extruder. It can't be the X/Y motion system or the previous layers would follow the same curved shape and wouldn't hit.
If the 1/8" aluminum bed is mounted solidly to that 2020 extrusion carrier, heat expansion could cause it to bow. It needs to be mounted in a way that it can expand freely. But again, as long as the temperature remains stable throughout the print, it should only cause issue with the first few layers and then level out.
The bowden, however, is nothing but trouble. The time delay between motor movement and flow through the nozzle means that when printing an island, a little bit of extra plastic is extruded at the end before the nozzle moves away. And with Z hop, it will be pulled upward into an even taller obstacle.
Go check out rq3's threads in the tech-talk section and make a Schnekenstruder. It's the best
If the 1/8" aluminum bed is mounted solidly to that 2020 extrusion carrier, heat expansion could cause it to bow. It needs to be mounted in a way that it can expand freely. But again, as long as the temperature remains stable throughout the print, it should only cause issue with the first few layers and then level out.
The bowden, however, is nothing but trouble. The time delay between motor movement and flow through the nozzle means that when printing an island, a little bit of extra plastic is extruded at the end before the nozzle moves away. And with Z hop, it will be pulled upward into an even taller obstacle.
Go check out rq3's threads in the tech-talk section and make a Schnekenstruder. It's the best
Sorry, only registered users may post in this forum.