Linear Z rail for AO-10x printers

[piercet]

This is another one of my hairbrained printer upgrade schemes, but since the last couple seem to have worked pretty well, lets give this a go!

Objective: replace the existing round linear slide on a Lulzbot AO-10x series printer with a two carriage THK 14.5" long Linear slide, which will in theory allow for a much more rigid Z axis motion and constrain the motion to the rails themselves, and to do this for as little money possible with as little modification to the machine as possible.

To that end, I have aquired a pair of THK Linear Slides 369 mm long with four 25 x 30mm carriages, and two 380mm long sections of frame rail to back them. I also have on order 8 45 degree frame pieces and 8 45 degree frame joining plates (could be printed, but I want rigidity)

The basic design I am going for is a triangular structure built around the lower motor mount from the existing long lower frame rails to the new bearing frame post, 4 45 degree frame sections per side, 2 per rail level. I’ll also be modifying the Z axis motor mounts to accomodate and lock the bearing frame post in place.

On the upper side of the machine is where things get a bit murkey in my mind. I have two options: 1. leave the A frame as is, and build a new upper crosspiece that has provisions for attaching the top of the bearing frame post, or 2. convert the entire A-frame structure to a box shape (similar to the TAZ) and use a short piece of 20x20mm frame tubing as the upper bridge in place of the existing plastic piece. I haven’t worked out which will be easier and stronger to implement as of yet, though I am leaning towards the box design. Any thoughts on that though would be welcome.

The other piece of engineering that needs to happen is the motor mount X end clamps. I’ll need to put the motor either 30mm further out than it is, or flip it around. The idler side should be usable as is with minimal modifications to the AO-TAZ design. The question I keep running into on that design though, is if I am going to tinker with the X ends at all, why not convert them to a linear rail at the same time? Or at least a vertical carriage like the TAZ uses? And I think eventually I will do exactly that, but for now in the interests of simplicity I’m thinking it will be best to just leave that aspect alone.

Still in the initial phases, still waiting on initial parts before I see how everything fits together and determine what I actually have to make, but if this works we’ll have a Linear rail system as a bolt on option for the older style printers, and be one step closer to making the most accurate desktop printer that we can make, which is neat!

this will be a fairly long duration project given the hollidays and demands on my printer for printing things, but i’ll keep you all posted

[1013]

how much where the linear rails? I know they can get expensive fast

I personally think the rectangular liner rails are WAY overkill.
I would suggest using the supported rods and slotted bearings instead…should be much cheaper, and still very stiff.

Like these:
http://www.probotix.com/index.php?view=category&path=34_40
http://www.probotix.com/index.php?view=category&path=34_37

those sizes may be a little large, but there should be smaller ones…
http://www.mcmaster.com/#open-style-linear-bearings/=pmv3zx
http://www.mcmaster.com/#open-style-linear-bearings/=pmv438

[piercet]

I picked them both up for $25 each, which was admittedly a good deal. It ended up being about $60 total for the pair after shipping. the smaller ones don’t go for as much as the larger sizes needed for full sized CNC units. 4 nice and small THK RSR12VMA carriages and two 14.5" long rails.

I looked at the supported rod rails, and the main reason I don’t want to go with them is that I would have to push the motor out 40 or 50mm to clear them instead of the 30mm that is already kind of pushing it without a major redesign of the X end. The other option there would be to make a support brace for the motor that boxes it in and ties it back to the X end around the rail itself, or just beefing up the mount plate itself another 9mm in thickness and runnign longer mounting screws.

I agree that the boxed rails are probably really really overkill and the rails you linked would be fine. the main improvement that needs made to these printers is support for whatever rail system they use in the middle of the rail. The approach i plan on taking could be easily modified to accomodate rails such as that though. the frame length on the stock AO-10x printers might make getting the 45 degree braces on the bottom in their problematic if I switch to a two channel extrusion. I’ll have to look at the positioning when the angle braces get here next week.

[piercet]

Here are a couple of pictures showing one of the rails on a piece of leftover 20mm extrusion from the printer widening project. The whole thing is fairly compact. if it weren’t for the motor it would be an easy retrofit, just a new backplate for the AO-TAZ x ends. as it is, I shouldn’t have to do much to the idler side at all. The Z axis height adjustment bracket will be relocated to the new bearing back post. The new Z motor mount will be created based off the lowered Z axis mounts that I don’t currently have installed. if I can make it work I might just make a bolt on spacer adaptor and tie the structure in off the two existing mount bolt holes for the existing rod mount.

I may end up chopping the rails shorter for clearance, or just shifting the extra length up for an eventual height increase.

[piercet]

got the 45 degree pieces in today. They are a bit long, so i’ll end up trimming about 15mm off the ends. that won’t be a problem because I also have 45 degree angle plates on the way so it will still be plenty strong. I was originally planning to have 4 angle pieces per side, but the motor mounts are going to be in the way if I go that route even if I modify them. I may change my mind on that later, but for now it will be two on the bottom, and a top cross tie. I’m thinking i’ll be able to use the existing plastic top piece and design a top pocket to mount to the existing bearing rod top mount and the two upper spar ends easily enough. The motor mount will get a similar treatment

Also for no apparent reason I swapped out the belt tensioner for the Y axis.

[piercet]

First of the new printed parts:

The top spar mount is designed to capture the upper end of the bearing support strut spar thingy and secure it to the top gantry. As this is a retrofit item I specifically did not want to make an entirely new top mount, so this is designed to bolt up to a machine that has already been upgraded to the additional 60mm print height modification stickied in this forum. These mounts will not fit a stock AO-10x printer.

Next up will be the modified AO-taz style end backplates to accept the new bearing carriages. After that will be the extended motor mount piece, and the lower z motor mount spar tie pieces. As usual these are all subject to change multiple times as I figure out what I did wrong the first time.
top spar mount 1_0.dwg (98.6 KB)
top Spar mount 1_0.stp (294 KB)
top spar mount 1_0.stl (107 KB)

[piercet]

The new backplate:

I did discover one additional clearance issue while building this. The lower X end clamps will also need some surgery to accomodate the lower bearing block. Just a minor 3mm removed from the inside backplate channel on either side, but without that it won’t fit. I’ll likely just cut that on my existing ones, but i’ll try and remember to draw up some corrected new ones for anyone else attempting this mod.

edit, I forgot 4 holes. the ones where the old bearings were, which will now just be additional mount points for the backplate. I’ll add those in later.
backplate_linear_z_1_0_a.stp (228 KB)
backplate_linear_z_1_0_a.stl (246 KB)
backplate_linear_z_1_0_a.dwg (86.8 KB)

[piercet]

Here’s the one with the corrected holes. I also added positioning guides for the bearing block itself

backplate_linear_z_1_1_a.stl (304 KB)

[piercet]

There wasn’t a good way to get the lower Z motor mount to capture the lower rail using the two existing rod mount holes. It just wasn’t fiting well, and there aren’t additional bolts like there are for the top plates to compensate. Since I’ve always intended to someday swap out to the lower motor mounts anyways, I decided to just go ahead and make a back rail spar capture derivitive. The main lower structure will actually come from the angle extrusion pieces, but this shouldn’t hurt either.

i’m going to dispense with uploading step files for now unless someone actually wants them.

Still need to make a longer X motor mount arm and the lower clamp ends with the clearance, but that will wait until I get a few of these things printed to check the fit.
lower_z_motor_mount_rail_1_0.stl (132 KB)
lower_z_motor_mount_rail_1_0.dwg (218 KB)

[piercet]

The new backplate print test fitted on one of the older AO-taz prototype idlers. Everything lines up well, though printing this does show me one major issue i forgot about. I can’t take that chunk out of the lower idler, because that’s right where the rear inner clamp screw goes. My only option is to extend down 7mm. This won’t be the problem it might have been, because its onlt the rail and the backplate that needs to go down, not the spring assembly or the threaded rod where the coupler might have been an issue. Also, the Z axis height adjust piece is going to get mounted on the extrusion back spar anyways so it will be completely out of the way. Remind me I need to make that piece before I try installing parts heh.

So, there will be another backplate revision, but the existing AO-taz lower x clamp piece will remain untouched. I think this will work. We’ll see!

[piercet]

Aaaand here’s the fixed one.

This will likely be the final arrangement for the backplate unless I notice somehting else off.

So that leaves the X motor mount clamp, and the Z height adjust piece assuming the top spar mount doesn’t need a redo too.
backplate_linear_z_1_2_a.stl (299 KB)
backplate_linear_z_1_2_a.dwg (93.4 KB)

[piercet]

a bunch of pictures of the prototype along with some of the dimensionally accurate but rejected AO-taz project bits to see how things line up. Printing the second side frame pieces now.

The assembled Z gantry

The new backplate

The top spar mount

The lower Z motor mount rebuild

A couple of the old AO-taz prototype pieces, with an idler filling in for the Lower clamp . You can see just how far out the motor is going to need to go to clear everything. I am worried about motor flex, so I will be reinforcing the crap out of the mount when it gets redone next.

At this point it might just be easier to flip it around TAZ style. but then it would hit the gantry and all that. Maybe next revision.

I also have a small micrometer barrel section on the way which will become the base for the new Z axis level adjustment thingy. It’s not here yet though

[piercet]

Here’s the new motor mount. I “think” it’s going to be solid enough. Printing it now, really won’t know until it is done. My fallback plan if it isn’t stiff enough will be a temporary aluminum brace across the top until i can print a thicker one.

xmotor_linear_z_2_0_a .stl (197 KB)
xmoto2r.dwg (366 KB)

[piercet]

After printing, i discovered there was still flex in the X motor, though not in the reinforced section, but at the curved join segment. I re-engineered that to eliminate the flex and the new one has less flex than my existing shorter mount, so I’m happy wiht it.

Just the Z height adjustment left now, then I’m waiting on parts.
xmotor_linear_z_2_1_a.stl (199 KB)

[piercet]

Here’s “a” Z endstop. I’ll be using one that incorporates a small inexpensive micrometer I’m about to cut up, which I will also post when I get it done, but this one should be in about the right spot. I’m not going to test this particular piece, but the measurements look right.
Z endstop Rail.stl (81.7 KB)
Z endstop rail 1.dwg (79.3 KB)

[piercet]

Here’s the endstop target I will be using with my inexpensive micrometer. I’ll likely also print a micrometer “hat” to widen the target area of the top threaded bit since I am not so much concerned with accuracy of the scale as I am with makeing sure I hit the limit switch

It is designed to fit this micrometer exactly once you cut the upper part of the “C” off. The endstop target mount will also be drilled along with the micrometer to accept at least one M3 bolt. maybe 2
http://www.ebay.com/itm/360702635282?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649
Z_endstop_Micrometer.stl (84.2 KB)
Z_endstop_linear_Micrometer.dwg (65.3 KB)

[piercet]

The newly printed, and not yet cleaned up motor mount. I’m happy with this one, no flex at all, seems to fit fine on the mock up arrangement with the spare prototype idler filling in for the lower clamp. Assuming the Z endstop print works, that’s all of the printed parts. I need to trim the aluminum bits next, then it’s just waiting for the piles of bolts and T nuts to actually show up heh. I swear I am cursed on shipping. Anyways, I expect to have this phase wrapped up soon. The deck plate proect is also waiting on that same pile of bolts to arrive so it’s getting close as well.

[piercet]

Printed the Z stop and attached my micrometer to it. Turns out it hits the backplate just slightly, and it doesn’t really have enough room for clearance on the bottom, so I’ll need to redo it a bit. it’s lined up pretty much exactly where I want it though, so I’ll just need to put an elbow in the arm and rotate the dovetail of the former micrometer part without moving the head of it which clears fine, and I’ll be set.

[piercet]

Here’s the redo of the endstop target for Micrometer. It’s a little rough on the top of the one support section, but I figure I’ll likely be the only person using this particular part, so I’m ok with just trimming it.
Z_endstop_Micrometer_Linear_Rail_2_0a.stl (101 KB)

[piercet]

Test fit on my mockup with the rail secured and all printed parts in place showed the backplate needed to be 0.88 mm thicker to fit properly. I took the time to clean the part up a little and also chamfer the corners of the raised bolt section that was in contact with the origional Z endstop just to get a bit more clearance. First layer height on this part is going to be somewhat important, though there is some room for adjustibility on the Z motor mount and the clamps themselves, so everything can conform to the top spar mount. I may need to add some adjustment slots to that as well, but I’m going to hope for the best for that for now.

backplate_linear_z_1_2_a.stl (253 KB)