My Tantillus Variant

So I love the Tantillus but ended up parting it out as you may have seen in for sale. I will probably build another one (or more) but have been thinking about building what I would change for my own personal one. I know reprap is filled with too many printer variations so let me know if this sounds like something that should just go in the trash.


LARGER PIC

This pretty much trashes a lot of what makes the Tantillus great, it is uglier, less enclosed, less reprappy etc..

Some of the things I am looking to accomplish:
- Size configurable, in the picture I have it roughly Tantillus size, but it can easily be scaled for larger beds / heights etc.
- Extrusion and laser cut panels should be easy and rigid without worrying about laser kerf on the tabs, acrylic thickness, bowing in the center etc.
- Easier to disassemble, remove rods.
- Re-use as many of the Tantillus parts as possible, I am not trying to re-invent anything here.
- Direct drive on the steppers, yes it is ugly but it gets rid of the gears and lets you use regular 50mm motors easily. This is a popular mod with Ultimakers.
- Replace the extruder with airtripper / planetary stepper variant.
- Possibly going mxl pulley / belts , lose accuracy but easy to setup.
- ??

Thoughts?

I apologize in advance to Sublime for this travesty!

Edited 5 time(s). Last edit at 03/08/2013 12:21AM by goopyplastic.

No worries I like that it's getting derived and changing.

The only thing I see in the rendering that made be say huh was the use of laser cut parts to hold the bearings. Why not use printed parts to hold them?

I have thought about the direct drive motors but did not like the space it takes up. But it does make me wonder if you could use a smaller round stepper in there place and get it closer to the size with the gears. Also note that I have added GT2 3mm pulleys to replace the gears for those people that do not like the gears or want higher resolution (1.25 times current resolution).

If I was not so concerned with the size of the machine I would have added tensioners to the X/Y ends which then requires drums on the shafts so you could square the axis by loosening the drum from the shaft. This would allow you to get the cables really tight and make the entire process easier.

Keep up the good work and keep us up to date with the progress.

Edited 1 time(s). Last edit at 03/08/2013 12:29AM by Sublime.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

With the extra clearance afforded by the open frame, you might be able to modify the XY ends to have built in tensioners. IMO the biggest pain in building a Tantillus is setting up the cables, and with the extra room to work stringing should be easier too.

That being said, I haven't had to retension since I replaced my last broken end in August, and I've put more than a few pounds of filament through my Tantillus.

Edited 1 time(s). Last edit at 03/08/2013 11:59AM by pokey9000.

Played around with the idea a little more and got away from lasercutting the parts and instead printing them like sublime recommended. (blue parts are printed parts).




Uses all of the Tantillus x/y/z parts and 8mm rods. The 5/16 rods need to be cut longer though, I am just using the same length for motor and idler rods then using collars / knobs to secure the rods in place.

The nice thing is that the frame scales up easily and you just have to change the extrusion and smooth rod lengths, granted a new beefier Z axis would probably be needed. Scaling it up also makes the 2020 look less overkill

Here is a 200x200mm printbed mockup



I like to the call the frame the tantillus tardigrade after the cute indestructible water bear.

Edited 1 time(s). Last edit at 03/11/2013 02:21PM by goopyplastic.

Looking good. What program are you designing in?

I hope you do not mind the feedback. Just a few ideas that popped into my head.

If you have access to a lathe it would be nice to center drill the rods so the motor shafts fit in them which would allow you to move the motors in and eliminate the coupler. You would then need a way of allowing the motor to flex but that could be as easy as having L shape slots cut out in a square around the motor mount bolts. This would allow all the rods to be the short rod length. It would also eliminate two bearings.

Also note that movement of the outer rods along there length is ok. The outer rods do not constrain the carriage/cross bars in the X/Y plane. The cables do that and the outer rods only constrain the carriage/cross bars in the Z axis. This is why Tantillus does not have any method of keeping the rods absolutely constrained along their length.

Could you not move the lower rear extrusion up to eliminate the extra bracket holding the lower Z bracket. (edit: Why not shorten the machine so you do not need the extra bracket?)

If you rotate the entire X/Y assembly 90 degrees you could then raise the upper rear extrusion to eliminate the extra bracket holding the upper Z bracket. You would then be able to remove the cross bar at the front of the machine (second from the top) and attach the front bearing holders to just the top and side extrusions. It would also make tightening the rear cable way easier since the tension screw would face upwards.

Edited 1 time(s). Last edit at 03/12/2013 04:28AM by Sublime.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Funnily enough I had an almost identical design penned, though my belt path is different.
Having said that I do like the idea of drilling the rods and mounting direct to the steppers, and it reduces the part count which is always a good thing.

Question for Sublime, where did you source your 8mm rods from the chrome plated ones I have lying around will not fit into a 608 bearing?

@Sublime
>What program are you designing in?

I am using Solidworks, overkill for sure. Thank goodness for edu licenses.

>If you have access to a lathe it would be nice to center drill the rods so the motor shafts fit in them which would allow you to move the motors in and eliminate the coupler. You would then need a way of allowing the motor to flex but that could be as easy as having L shape slots cut out in a square around the motor mount bolts. This would allow all the rods to be the short rod length. It would also eliminate two bearings.

It is a neat idea however chucking up a 5/16" rod that long might be difficult, plus I'm not sure other people could repeat it. It would be a lot of work and potential alignment issues, bearing load on the steppers etc.. My goal with the frame is to make something solid and easy that re-used the tantillus parts even though that means there are some concessions.

>Also note that movement of the outer rods along there length is ok. The outer rods do not constrain the carriage/cross bars in the X/Y plane. The cables do that and the outer rods only constrain the carriage/cross bars in the Z axis. This is why Tantillus does not have any method of keeping the rods absolutely constrained along their length.

Ya I hear you, I still like the idea of having some collar / knob, the upside is that you could attach a large collar / knob to give you something to manually jog the axis with since you don't have the herringbones to grab on to anymore. Not to say that you couldn't just build the frame with flush rods and not use the collars.

>Could you not move the lower rear extrusion up to eliminate the extra bracket holding the lower Z bracket. (edit: Why not shorten the machine so you do not need the extra bracket?)

The bracket is so I can use the original z bracket and mounting points for the Tantillus. It would be easier for sure to just have a new z top/bottom bracket that attached directly to the extrusion but these would be specialized parts (I realize the irony since the adapters are specialized parts). I was kind of looking to maintain as much of the internal tantillus as possible and have the varient be more of a frame. The upside of this is that when you release new z axis mods like the cable z you wouldn't have to wait for a redesigned part for the frame (if someone even got around to it).

@Polygon the Tantillus uses 5/16" rod so that it will fit inside the 608 bearing, good 8mm shaft won't really fit. Also what printer/s did you end up with?

Yeah I realized the 5/16 thing when I found the BOM.
I'm somewhat stalled on the HBot I was building, I'll get back to it, but I need to think through a few things.
I wanted an Ultimaker style machine to compare the geometry with my existing machines, and I refuse to import one.
My design is based around Open beam and uses bearing plates that look almost identical to yours, i.e. I use the same offset plate inverted for the axis between two rows of extrusion. The major difference is I have a separate stepper mounting plate and I intend to use belts. Though the belt thing might change if I buy drill rod for the 5/16 rod since I can relatively easily machine that, and the Chrome plated 8mm rods I have seem to be more or less impervious to almost anything.

I like the concept of integrating the steppers with the bearing plates and this is more appealing if i drill the rod-ends and lose the stepper offsets, I'll probably give it a shot because I can.

Edited 1 time(s). Last edit at 03/12/2013 06:58PM by Polygonhell.

I have some more time to work on this so hopefully will be updating this as time goes on. I only had an Afinia printer so printing the panels wasn't an option (too much warping in abs, and poor pla support). Got around to getting my i3 printing so now I have a printer to print parts for this on.

Did a test print of one of the panels just to check clearances and part strength etc..



Video of part printing if you are really bored

Edited 1 time(s). Last edit at 03/29/2013 07:23PM by goopyplastic.

Nice looking part. Is there anything holding in the bearing or is it press fit.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

There is a flange underneath it restricting the depth of the bearing in the hole. The exterior face will be kept in place by the little knob/collars on the shafts. Might be fun to make a snap fit lip on the top too, but that would be harder to replicate consistently if others were trying to print it.

I do think the rod movement I had in my first tantillus was due to rough bearings / rods. Having built some printers after it I have noticed that it can be much much smoother and probably wouldn't exhibit this problem.

I do kind of like the external knobs collars for some sick reason, having a press fit lip would make them optional.

The printed panel is quite strong, and I think that I might be able to get away with scaling back some of the extrusions, for example I don't know if the middle extrusions are all that necessary.One main benefit is that the interior of the machine should be more accessible. I was toying with one piece corners:



But I think printing those would be needlessly difficult and that separate pieces would probably be strong enough.

So here is what I am playing with now:




The upper Z rod holder extension is a bit of a placeholder for now.

Edited 4 time(s). Last edit at 03/31/2013 02:46AM by goopyplastic.

I'm liking this variant, and it looks relativity quick to bang together. Mind if I join the party and do a version?

All are welcome to join in the fun.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Made a GitHub:

https://github.com/goopyplastic/tslot-tantillus

BOM should be in the README as well, at least that is the intention.

I printed off the separate panel parts and I just don't like them as much, they float around in the t-slot a but before tightening making registration potentially an issue plus they weren't as rigid as I wanted. I had good results with a test print with a combined panel so I am going that route for now.

Hopefully I will more stuff to show later this week.

Edited 1 time(s). Last edit at 04/07/2013 04:04PM by goopyplastic.

Quote
goopyplastic
I printed off the separate panel parts and I just don't like them as much, they float around in the t-slot a but before tightening making registration potentially an issue plus they weren't as rigid as I wanted. I had good results with a test print with a combined panel so I am going that route for now.

Sounds good, and I think there is a third way.

I have created some printed parts that live in the corners of 2 T-Slot extrusions like a cobweb in the doorway. These parts have integrated tab that slips into the T slot. They must be slipped into the end of the extrusion.

It makes for a very rigid corner and fast assembly.

Nice work

Haven't had as much time to play around lately, but thought I would show a pic of the frame.

Looking nice Goopy!


Does the standard Tantillus Z axis work in there? Is that z-axis part that connects to the top rear extrusion rigid enough? I wonder if there is a slightly different Z azis design that isn't cantilevered that you could use? I think the cantalever might be a limiting factor in how big your varient could grow.


I should get some 20x20 so i can have a go with your design too.

It is meant to work with the standard tantillus Z axis. I am currently building it up with the cable Z axis (and converting some of it to metric), which has the benefit of not needing a separate motor mount. The top mount is a little flexible currently although the parts are 5mm thick so fairly rigid. Once I have the Z bolted together I can see if it needs more reinforcement or not.

I do agree, if you scaled up the design you would need to redesign the Z axis. I am going on vacation for awhile, but when I get back I expect to have more pics.

I have the extrusion on order, I even have some yellow PLA.

Finally something to update, it's definitely coming along and I am really happy with it so far.



More Pics on Flikr

Also I have made the github current

https://github.com/goopyplastic/tslot-tantillus/

Your build is looking fantastic.

What's the purpose of the gears at the ends of the xy shafts? Is that how you keep those rods from moving?

They look like knobs to me...

I suspect they are for moving the X and Y axis by hand when no power is applied to the stepper motors.

...OR - he just hasn't mounted the stepper motors yet!

Edited 1 time(s). Last edit at 05/16/2013 09:39AM by rhmorrison.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

Looking good. Shouldn't be too long before its up and running.

Any reason you went with the airtripper extruder? From the maths I have done it will limit the lowest layer height you can go.

1.75mm filament
0.5mm hotend

100 steps per mm = proper min layer of 0.119
200 steps per mm = proper min layer of 0.08

-------------------------

1.75mm filament
0.35mm hotend

100 steps per mm = proper min layer of 0.171
200 steps per mm = proper min layer of 0.108

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

The knobs I have for a couple of reasons:
- so I can easily manually jog the x/y
- they hold the 608's in place
- they keep the shafts from moving laterally
- they will look neat spinning

@ Sublime The airtripper may be temporary until I get it running and tuned. I could go with a geared airtripper with a phidgets or kysan 5:1 gearbox stepper as well. I also don't have a source of hobbed bolts at the moment so couldn't really use the tantillus extruder. I may want to modify the tantillus extruder to see how compact I can get it since I have less space constraints. I am open to ideas on the extruder though, the airtripper is just what I am used to, and what os1r1s used on his tantillus.

Also you were right, once I added the spectra the x/y was like butter.

I am not saying it will not work I was just curious. The limitation I am talking about only comes when printing high detail stuff where you need extrusions no longer then they are wide. If the model is made up of long segments then you can still go really low and not have any issues.

Making a hobbed bolt with two printed pillow blocks or a bolt hobbing device is really easy. But as for a shorter extruder I do not see any reason why it would not work fine. Or use RichRaps bowden cable variant of Gregs extruder. He is using it on his Rostock with PTC fittings. He also has a variant on the bowden clamp or carriage of Tantillus to use the same type of PTC fitting.

Tantillus's extruder is unusual with 2.7:1 gearing which gives decent torque while still allowing fast retractions. I believe most of the gears for Gregs are between 3:1 and 5.5:1

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

The good news, updated the github with a current and complete set of parts.

The bad news, my tantillus is starting to look like something playschool made! perhaps it is just the odd choice of colors.



It is now complete mechanically, need to start working on the electronics.

Looks like the Barbie 3d printer!
All the girls should have one...

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

It's looking great. And I really envy those corners after I snapped (as in snapped into pieces) four or five of the ones I printed for my build.

That is one of the finest looking printers I've ever seen excellent job! One question though if you were to expand the build platform to a 200 x 200 mm table for example why couldn't you just use 2 of the cable driven Z's on each side of the table with two steppers run in parallel? Wouldn't that let you keep the speed and "jerk" relatively the same to the smaller machine, and would'nt the 4 rods keep the larger table more stable than some of the other rep raps?