T slot extrusion tolerances
Posted: Wed Dec 31, 2014 6:50 pm
Hi,
I'm just wondering if anyone on this forum knows a lot about T slot aluminum extrusion tolerances. It's difficult to find data on the internet about this kind of stuff, but I think it's important for our printers.
There are a bunch of things that can be measured;
Length of the extrusion
Cross-section width (I measured this to be 0.996 on several of the rostock towers)
Slot width
Straightness
Twist
Warp
The cross section widths I've measured seem to be very consistent across extrusions, they're always about 0.996 inches wide for a 1 inch extrusion.
I have no way to measure the length of the extrusions precisely. The surface finish on the cut ends looks very smooth (almost fly cut), but repeatability is the most important thing if a 3d printer relies on having precise alignment based on extrusion length.
Slot width: important if rollers are used like on the Rostock Max.
Straightness: (sometimes called "bow") is how much the rod is bent over its length. This is very important too.
Twist: is how many degrees it rotates from end to end, this can be reduced greatly be clamping the ends in a rigid frame. Unfortunately the Rostock Max doesn't do this.
Warp: I'm not sure about this, but I think it's similar to straightness.
I've seen quite a few delta printers that use cylindrical linear shaft instead of rolling carriages on the T slots. There are also a few printers now that use special ground linear rail bolted to the T slot extrusions. I want to know if these are necessary for precision before I finalize the design of my own large delta printer. T slots seem convenient and rigid, but if they have bad tolerances that affect print quality, it may be worth it to use precision linear rail in addition to the T slot extrusions.
Also, does the size of the extrusion affect its tolerances? My hypothesis is that larger extrusions will be more rigid than smaller ones, so they will bend less during manufacture. Maybe I'm wrong about that though.
I hope this doesn't come across as a ramble, I've just been thinking about it a lot because it's a very important part of any linear motion machine.
I'm just wondering if anyone on this forum knows a lot about T slot aluminum extrusion tolerances. It's difficult to find data on the internet about this kind of stuff, but I think it's important for our printers.
There are a bunch of things that can be measured;
Length of the extrusion
Cross-section width (I measured this to be 0.996 on several of the rostock towers)
Slot width
Straightness
Twist
Warp
The cross section widths I've measured seem to be very consistent across extrusions, they're always about 0.996 inches wide for a 1 inch extrusion.
I have no way to measure the length of the extrusions precisely. The surface finish on the cut ends looks very smooth (almost fly cut), but repeatability is the most important thing if a 3d printer relies on having precise alignment based on extrusion length.
Slot width: important if rollers are used like on the Rostock Max.
Straightness: (sometimes called "bow") is how much the rod is bent over its length. This is very important too.
Twist: is how many degrees it rotates from end to end, this can be reduced greatly be clamping the ends in a rigid frame. Unfortunately the Rostock Max doesn't do this.
Warp: I'm not sure about this, but I think it's similar to straightness.
I've seen quite a few delta printers that use cylindrical linear shaft instead of rolling carriages on the T slots. There are also a few printers now that use special ground linear rail bolted to the T slot extrusions. I want to know if these are necessary for precision before I finalize the design of my own large delta printer. T slots seem convenient and rigid, but if they have bad tolerances that affect print quality, it may be worth it to use precision linear rail in addition to the T slot extrusions.
Also, does the size of the extrusion affect its tolerances? My hypothesis is that larger extrusions will be more rigid than smaller ones, so they will bend less during manufacture. Maybe I'm wrong about that though.
I hope this doesn't come across as a ramble, I've just been thinking about it a lot because it's a very important part of any linear motion machine.