Adding Openrail Z axis to a Lulzbot Taz 4/5

Now that I have the X axis mounted and working on the printer and the Y axis project is nearly completed, it’s time to start looking at the Z axis. I want to use as many off the shelf parts as possible, so I am going to attempt to design it to use the Openbuilds Mini V wheels and the existing Mini V Wheel plate. I’m also going to try using a single 20mm x 20mm Vslot extrusion per side. I am concerned about the potential flex in a 20mm x20mm extrusion, so I plan to also include 2 or 3 backbone stiffener segments running perpendicular to the extrusion and tying into the existing vertical frame rails. The top plate and bottom motor mounts on both sides will receive fairly major surgery to make everything fit and will need to be replaced on both sides. I looked at going with a 20mmx40mm extrusion but I can’t fit it in with the mini v wheel plate and the existing control box, and if I turn it sideways, I’ll have to get a custom plate made. Plastic would have too much flex for that particular application. That’s a definite possibility, but I’m trying to keep this as accessible and as low cost as I can, and custom milling is definitely not low cost.

I’m also probably going to replace the Z endstop target with a micrometer barrel. I plan on going with automatic bed leveling soon, but at the same time I still want a fallback endstop just in case, and there may be others who want to stick with regular leveling for now, but want more precision. I may make 2 variants for people who want to stick with the existing piece since micrometer barrels can be somewhat expensive.
I’ve ordered the following hardware for the prototype:

4 x Aluminum Spacers (Size: 6mm)
4 x Eccentric Spacer (Length: 6mm - Mini Size Wheels)
2 x Black V-Slot™ Linear Rail (Size: 20mm x 20mm, Length: 500mm)
4 x Mini V Wheel Plate™
8 x Xtreme Mini V Wheel™ Kit
8x 25mm low profile M5 nuts

Hardware cost with shipping $129

This is still a very early stage project. If anyone has any ideas or requests, please let me know. This will also work with my Anti Z wobble nut (V 1.0 or the still as yet top secret V 2.0) or the stock nut mounting pieces. A future variant of this project may also incorporate Belt drive Z rather than Leadscrews

More to come

Some movement on this project, I got the parts from Openrail.

The plan is to use 2 of those square blocks, with 2 wheels each, mounted to a printed adaptor plate that will in turn mount to the stock 4/5 X axis end plates. The blue thingies are the anti Z wobble devices which I plan to mount at the same time.

The rail itself is going to mount behind where the existing smooth rod is located, closer in line with the other 2 vertical extrusions. The top and the bottom printed parts will both get a bit of a redesign to accommodate mounting. The top piece will be easy, the bottom one will not be as fun. I’m also not sure if I can print the square hole I need without requiring support on that part, which I would like to avoid if I can. Other options include a 2 piece lower motor mount section, which I would like to avoid since it would potentially introduce alignment issues.

There should be plenty of space to fit everything in. I’ll most likely get the adaptor plate and the mounts ready this weekend sometime

The mini vslot wheels do not roll as nice as the larger ones. An option that I may end up doing since it isn’t under a huge amount of constant motion like X or Y, would be to print a slightly larger V wheel adaptor plate and use those instead. Or make my own aluminum one. I want to keep this as much off the shelf as possible just to keep costs and printing time down.

First new part for the Z retrofit. It’s going to need a bit of support to print properly, But i’ve added that in. The 20mm extrusion replaces the existing Rod. My theory is you install the middle extrusion first, using the holes for the outer extrusion bolts to insert the hex wrench, then insert the outer bolts and install as normal. I’ll probably finish and attach the right side and see what the clearances look like with the new as yet to exist backplate that will couple the Z wheels to the Taz 4/5 former double bearing mount plate location. Everything is looking pretty good for clearance so far. Right top plate is up next, then the backplates. The backplates should be symmetrical so that should simplify things a bit.

I’m also debating extending the bearing down a little bit. With the Z endstop target, the existing double bearing holder can’t ever theoretically “bottom out” on the leadscrew lower bearing anyways, so I have potentially another 20mm to play with there. Not sure the benifit from the greater separation distance would be worth the tighter corners for the endstop target though. We’ll see. Anyone have any thoughts on that one way or the other?

Openrail_Z_Motor_Right_1_0_a.stl (303 KB)

Top Leadscrew mount plate. Not sure if all 3 screw holes will make the final part for the Openrail. I also have a need to accomodate some electrical wiring for the lights across one of those. I may end up making either a provision for the cables, or a cable attic of some sort for that as well, but I haven’t got that far yet. I’m also not 100 thrilled about the 5mm travel loss at the top, but I’m not sure of a better way to do it yet. On the upside, with this particular rail setup we could theoretically make the taz a bit taller without any degregation in accuracy.

openrail Z both top 1_0_a.stl (298 KB)

I also finished up the back brace. Since the openrail has a T-slot on the back, why not support it more? I’m planning on adding 2 or 3 of these per side. The scallop relief cut on the bed side is a precaution to guard against leadscrew misallignment impacts. Once could fit more of them in there, or make a wider plate of some sort. I may experiment with those options as well. This will require moving the Spindle arm to the outside of the frame if you wish it to still fold up. You could also consider this part optional. I may end up making a variant incorporating a spindle mount as well.

openrail Z Backbrace.stl (75.2 KB)

I ended up redoing the backbrace since it wasn’t clearing the bolts on the openrail carriage. Forgot they protrude that far out. Here is the updated piece.
openrail Z Backbrace.stl (92.2 KB)

I ran into a bit of a puzzle with the backplate. I basically have 7mm total to play with in terms of space if I keep the openrail exactly parallel to the existing Frame members. The metal X/Z interface plates can’t move, and the 4 M3 screws that would have held the Double bearing holder, and will now mount the backplate, exist in exactly the same location as the openrail blocks need to go. I don’t have enough width to safely bury the screw heads out of the way and still have a thick enough surface.

Neither of the options I have some up with so far are great. For the first, I move the rail back 4-5mm, redo the motor mounts, redo the back braces, and redo the top plate. Doable, but I have to add additional plastic to the motor mounts to make that work, and then the frame assembly and allignment process gets more complex.

The other option involves making the backplate itself be the openwheel mount as well. I can make that work from a design perspective, the part that worries me is rigidity. The resulting piece would be thicker than the aluminum part, but made out of plastic. That may be my only other real option though. The plate butts right up against the X to Z bracket, so I can’t really do much to reinforce there, and if i go that route I have to print up and out towards the Frame, which strictly limits the thickness I can make things. Adding an I-beam like outer perimiter feature is possible, but i also have to avoid the leadscrew interface area, which i completely forgot about in the one that won’t work that is pictured, so just ignore that part. That also means the option of using the aluminum ones won’t work unless I space them out much further, which would increase stability but decrease overall Z travel unless I make the frame bigger. On the other hand with that bearing system and the backplates, I really could make it quite a bit taller if I could find longer leadscrews without affecting the new theoretical greater accuracy I hope for.

Anyways, not sure how I am going to proceed there yet, but I’ll hopefully think of something less drastic than top secret plan “g”

Ok, i think I have a working backplate / openrail hybrid thingy now. Printing it at the moment, we’ll see how it goes. It’s reinforced thicker than the Aluminum piece, so I am hopefull that will compensate enough for ABS’s tendancy to flex. The scallop out of the bottom section allows clearance for the Z endstop target. There will also probably end up being a right and a left one ofthese, because I want to keep the Adjustment cams on the same side of the frame.

Taz_Openrail_Z_Backplate_2_0_a.stl (229 KB)

Part is printed, And my fears about too much flex turned out to be unfounded. It’s very solid as is. I do need to make some modifications to improve it. Right now my Anti Z wobble nut is too tall to fit in there properly, and the entire assembly printed out about 0.5mm thicker than I need it overall. Both of those matters are easily fixed though. For the rest of it, it certanly seems like it will work at this point. I’m also going to remove a small amount of material from the Top motor mount central structure. That will allow the carriage to reach near the top, restoring 4mm of the 5mm travel loss.

Just asking, why not exclusively use metal brackets like I did with Taz Mega?

Space constraints made it impossible for me to fit the premade stock mini v wheel blocks that I intended to use there in. Cutting an aluminum plate would be an option, but my CNC mill is still not finished, so making a pair of identical plates with sufficient precision would probably require outsourcing. My main focus of this series of projects has been creating bolt on additions for the existing TAZ that can be implemented by any 3d printer owner. Staying with a 3d printed part there widens the number of people potentially able to use the modification in theory anyways. Metal brackets are useful, and when they can be easily integrated and won’t add too much additional weight, they are definitely the way to go. This is one of those instances where a plastic piece can do as good a job as the metal one. Some of the structures would be difficult to mill as well. Near as I can tell, the Taz doesn’t seem to need much in the way of frame reinforcement. Adding the extra backbone pieces and the Y axis modifications and whatnot makes for a more solid machine anyways.

Here are a bunch of pictures of the mock up rail. Also, for the record I love the new picture upload thingy.

Some of the structures would be difficult to mill as well.
Actually, I think a lot of this stuff can be done with just flat plates and off-the-shelf brackets. Designing around 2D parts allows for easy drop-in upgrades later as required.

By the way, eMachineShop prices for T6 custom plates are highly competitive, comparable to blanks from MetalsDepot, with excellent customer service.

CNC mill is still not finished
Curious to hear more about that project.

That’s my Other Robot Toy. It’s a Blurry Customs Ironman series CNC mill. 24" x 16" x 8" cutter head travel. Gantry style, with a Gecko controller feeding a Paralellel to USB adaptor. 4 Axis capable, currently 3 Axis with a Bosch router. Since these photos were taken, I’ve also finished the air spray / mist cooling system, part of the tub, and the vaccume line. I still need to hook a few things up, print a pulley cover or two, and build the Z minimum stop adjustment bracket. This thing did not come with any endstops, so i’m having to figure out where to put them as I go, and alot of that work occured before I had a 3d printer. I actually bought the first 3d printer mainly to make parts to get this thing working. It’s at about 90% complete. most of the remaining tasks are wiring and a few minor design things.

Wow, that is a really nice machine, and it looks reay for work. Good luck firing it up!

I had an error in the top plates, Here are the corrected ones.
openrail Z top a.stl (139 KB)
openrail Z top b.stl (139 KB)

One side of the assembly is on. I want to see if I can notice a difference between print quality on one side of an object compared to the other with only half the unit installed. That and taking the side with the electronics box off is a pain in the ass.

Anyways, heres what it looks like so far. Seems to be very solid, no play at all now, and so far seems to move well. Trying the first print next. If it all works according to design, next up will be the other side motor mount, then it will be done. The motor mount with the Z endstop target will actually come in two flavors. A standard model, and one that will accept a Micrometer barel for a height adjustment. Mainly because I have a spare micrometer barrel, so why not?

Overall unit

The Back brace. Not strictly necessary, but it won’t hurt and could help.

Left side motor files. The micrometer barrel one is designed to be used with a micrometer with a 9.6mm barrel, a 14mm retaining ring, and a 5.5mm thick mounting base.
openrail Z_ left_motor_micrometer 1_0_a.stl (343 KB)
openrail Z_ left_motor_standard_1_0_a.stl (320 KB)

Here’s the final micrometer equipped left motor mount all printed. Also, lacking a better place to put them pictures of my new aluminum heat bed

This all looks very intriguing to me, as I’ve had persistent and constant issues with my Z axis ever since building my kittaz. Will this build help eliminate the wobble and binding that I’ve had to combat every few prints? I will absolutely build it if so.

The X,Y and Z retrofits will significantly reduce, but not eliminate entirely Z banding issues. To fully resolve Z banding, you will want my anti-z wobble nut project, which is currently in final prototype design and testing phase, and I hope to have finished version by next week. These projects will help more with overall part dimensional accuracy, layer thickness consistency, bed adhesion for larger parts (since they eliminate droop and the apparent 0.01mm “hump” in the middle of the bed due to rod flex) and positional stacking of layers. On my Taz with the three upgrades installed, I have less wobble than I started with, but it is still visible without the anti wobble. WIth the early version of the Anti wobble that is effective for X axis only, I see a 100% reduction in wobble in the X direction. My early Y prototype is only partially effective.

It may also help with your binding issue. Z binding is usually caused by the leadscrew heights being offset a turn or two from eachother. A rigid box structure keeps them better aligned, but you may want to use a pair of calipers and measure from the bearing to the underside of the leadscrew nut and make sure the distance is identical on both sides when you get a chance. You’ll also want to make sure both motor mount pieces are slid up all the way to ensure that the Leadscrew can’t move up or down at all on one side, which can also cause binding