Repstrap vertical axis coupling problems and printable options

Hello everyone !

While my repstrap is up and running i am still having regular issues with 'slanted" printing due to the vertical axis slipping :

I am currently using aquarium acryl tubings to attach the steppers to the M8 rods :

(oh and i am using a dual stepper vertical axis, but the problem does not stem from them) and after a few days of printing the axis start to slip.


As i can still print things i am considering following options for the couplings (rough designs)

Option one : "spider coupling":

Option two: pin coupling

Option three: hybrid



Before i try printing anything (my printer is quite slow), could anyone with experience with this type of problem have any feedback ?
or even any other method that would work, because frankly i am at my wits ends finding other , simpler solutions
Thanks in advance!

---

Toiling away on a repstrap : [www.kaosat.net]

It looks like either your X or Y axis is slipping, not your Z.

The spider approach would work, allowing the shafts to be off-center. Your X and Y axis probably don't use the threaded rod though?

---

--
I'm building it with Baling Wire

I don't know why you implicate the z-axis. It looks like X or Y is the problem.

---

[www.hydraraptor.blogspot.com]

Thanks for the quick answers guys !
My bad , i think i should have been more clear about the machine's structure:


-All of the axis use threaded rods.
-At first i was not certain it was due to the z axis, but after trying things out , this problem always turns up when the X axis' becomes "slanted" on the z axis' two threaded rods
-Each time i replace the coupling tubing and make sure the x axis is horizontal again it works ... for a while :

@jgilmore: is there no risk of "slipage" with the spider approach with a dual stepper system ?

---

Toiling away on a repstrap : [www.kaosat.net]

A couple of hose clamps on the tubing might help a lot; you could also try stronger tubing, perhaps air hose?

Also, you may want to swap in some acme-threaded rod for the x and y, just for the massive speed increase. Email Rob Gilson; he's been playing around with the stuff lately and can point you at a supplier and rave at you about how nice it is.

While you're at it, don't forget to put a parts list for Doboz in the wiki, eh?

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

Thanks for the replies!
In the meantime, after some additional testing it turns out it was a double problem .. not only did the Z axis tubings slip but also those on the X axis (first time this happened ... weird ) no wonder the offset was even worse in the prints after the one i pictured above

@Wade : you are completely right ! A hose clamp on each stepper axis and the threaded rod on the tubing should hopefully do the trick ! Thanks a lot, will try to find some tomorrow

@Sebastian : haha that funny, i was just looking at acme threaded rods for those axis yesterday (people getting a tenfold speed increase in some cases really is a motivation but all suppliers i found so far have maddeningly high prizes so if Rob Gilson can help, i would be very gratefull! - i just Pmed you for more details .

I haven't forgotten about the wiki , i'm preparing a full parts list & build instructions thingamajig

Two things:

A hose clamp will work (good suggestion, I'd forgotten that one even though I'm using similar on my machine), but so will baling wire (also called "tie wire" and sold by concrete suppliers) a roll of tie wire will cost about the same as 4 or 5 hose clamps, but will last MUCH longer. And will also take up less space, as you can simply cut off the excess. I was able to fit the baling wire in on my Z axis, no way would the hose clamps have fit in there.

One inexpensive supplier for both simple ACME threaded rod (and nuts) and also multistart stuff is www.roton.com Their Torque Spline stuff is also worth looking at as it's more-or-less designed for our sort of application.

---

--
I'm building it with Baling Wire

@jgilmore: thanks a lot for all that information, that seems to have solved most of the issues! i haven't been able to find small enough hose clamps locally, but there was bailing wired, and that helped a lot ! Thanks

I would really like to thank everyone for their feedback, since i made quite a bit of adjustements (redone heated bed and its electronics, used bailing wire on the axis as suggested, redone a part of electrical connections, and adjusted skeinforge settings) and these are the (much improved) results:

before/after:

Finally clean and sharp:



Also i looked at the Torqspline and its pretty much spot on ! Within the acceptable lead-screw diameter range i could adapt to my machine, i could easilly get 1 inch per rev so a around 25.4 speed increase (with the current screws its close to 1 mm per rev...) . Ideally though i would like to find european suppliers for similar lead-screws, to avoid switching from metric to imperial , but really great info nonetheless!
Which leads me to the question : does anyone know of european suppliers of multi start , lead-screws that also accepts individuals ordering in their shops and at acceptable prices ?

Damned ! After a couple of successful prints, my lateral offset problems started again, much worse this time !

Even after a lot more corrections i still get this kind of thing :

Always an offset to the left, on the X axis.

The thing that is driving me nuts is that this problem is ONLY present when printing round elements!!
Even on the slanted elements , the problem just dissapears when it prints the teeth for example:

And it is not present at all when i print square things :


These are all the things i changed / checked, to try to solve the problem

• Changed vertical axis rods for higher quality ones
• adjusted the X axis to reduce wobble and backlash
• changes all axis couplings
• added more bailing wire on each axis to secure axis to motor coupling
• checked for alignement problems with a laser pointer !!
  
  

  5225688847_8b89fe5cfd.jpg

I am frankly starting to think that this might be caused by something else altogether but i am completely at my wits end, i simply don't get it.
Has anyone got ideas ? Or even experienced something similar ?
Thanks a lot in advance!

Ps: i did not want to spam the forum with another topic, but it really is not vertical axis problem related anymore

... you can test with bows and segments of a circle or ellipse, so you can check if the malfunction is only present with full circles or only with specific angles, what could be a problem with the bresenham-algorhythm or with coordinate-transformations ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Thanks for the very fast reply , Viktor!
would those approximation problems not cause distortions within layers aswell ? Or would the error be small enough and cumulative to only show up on layering like this ?
As far as i can tell each layer is as rounded as can be , as i also tried to use the anti backlash within skeinforge, and that just distorts each layer, without solving the problem ...

... it seems someone is shifting the home-position or skipping steps when closing circles

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Thanks again Viktor!
Since the problem is still getting worse i am just going to have to go through a checklist of all other possible error sources i guess...
The step skipping or home position skipping would be weird since i am using a hardly modified 5D firmware but i will check all that aswell...

Hello all !
Since it might be of interest to people coming across similar problems, it turns out the problem came from the Pololu stepper driver!
Might have been the trimpot being set to low, or just a random glitch on the x axis driver, but switching it out solved it all !
I also have an article detailing all the steps i went through to find the problem.
Thanks to all those who tried to help out with this problem

---

Toiling away on a repstrap : [www.kaosat.net]

I had a similar Issue, and solved reducing the values in DEFAULT_XYJERK and DEFAULT_MAX_ACCELERATION (config.h) in the Sprinter config.

as read in : [forums.reprap.org]

Regards

Crispy