Prusa i3 - alignment issues

Hello Everyone,

I've been very busy the last few days and I finally have some time to work on Reprap stuff!

So now I'm trying to understand how to align the Y axis on a Prusa i3. Mine has a wooden box frame, but the issue would crop up in a similar form with the flat panel frame. At the moment my main concern is just getting the Y axis perpendicular to the Z. I find that once you fasten the Y axis threaded rods, you are basically stuck that way permanently. The Mendel or Prusa i1/i2 has some adjustment options, but not the i3.



BTW, does anyone know how to properly pronouce Josef Prusa? Maybe "Yosef Prusha"?

Edited 1 time(s). Last edit at 07/21/2013 10:38PM by Yvan.

---

Yvan

Singularity Machine

Whew, I can hear the silence in here...

Seriously, how is everyone accurately aligning thier Prusa i3?

My lame solution is to slip card stock or some such shim material under specific mounting points. Not much room for correction there, so I'm trying to keep the frame as accurate as possible to begin with.

And does no one anywhere know how to pronouce Josef Prusa? What is his phone number??

---

Yvan

Singularity Machine

well.

I can write to you a Bible about everything what is wrong with today 3d printers.

but lets go in some order.

put machine on straight, leveled table, on place where this machine will be whole time, if you carry machine from one place to another then steps below must be done each time you move machine...


about aligning Y axis...
well you need to fix (hm, whats word in english...)that steel right angle device lol on your bed, best directly to Y axis, not to heated bed, its best to level heated bed to good calibrated maschine.

then you take hotend off, put on x axis meassuring gauge and meassure whats wrong, then is everything up to you...you need to improvisate to make it straight, and there is various ways, one is to support box somehow to make it straight or another is to add washers , sheets of paper etc under theZ printed parts and get right value....

i have single sheet i3, and will add threaded rods for setting up/support for yz




and when you start to calibrating everything with precise meassuring gauge then you will see, that all machines on this forum/world can be 200%
better if were used thicker smooth rods, and more rigid chassis, because then you can see how much whole stuff plays when it works.



ofcourse, if you print vases, bracelets, then it isnt so important , just print

Edited 2 time(s). Last edit at 07/23/2013 01:07AM by woo.

I hope you own a good set of vernier calipers. This is probably the number one tool for RepRappers. + a right angle square with leveling bubble.

I'm currently building several Prusa i3s ( single steel frame ) steel framed Prusa are relatively easy to square up as the smooth rods simply pop into
the plastic end pieces etc.. but still some measuring is required.

Edited 1 time(s). Last edit at 07/23/2013 08:36AM by Mickman.

I'm converting my i2 into an i3. I'm using the lasercut hardboard frame that has the large support braces. I think these will help get and keep the Z axis square. However, I'm going to also apply somthing I learned from my i2 and mount the thing on a 3/4" MDF board as large as the foot print of the printer. I'm screwing the Y feet to the board and going to use little right angle brackets to mount the frame and braces to the MDF board as well. I went with a thick MDF board to make sure it wouldn't flex. I also have left room to mount the PSU on the board.

woo, well yes that is very true! LOL

When Adrian Boyer started his RepRap stuff, it had to be made from low cost easily found components. So the machines are not very rigid...

Mickman, yes, I have some calipers, and I bought a digital inclinometre when I assembled my Prusa i1. That helped. I want to buy a basic set of machinists squares soon.

appdev, funny, I was thinking the same thing... I have extra 3/4 inch MDF I might use for the base. That might solve one of the numerous alignment issues.

---

Yvan

Singularity Machine

simple fact, i set my heated bed tolerances inside 0.03mm with meassuring gauge..

then i just EASLY lean hand on bed and it goes down by 0.2 mm....

that bothers me alot...but ok, we all learn on mistakes...

i have @ home 10 mm smooth rods and linear bearings from another project , so right now adapting whole i3 for all 10 mm rods so that i dont spend not much money for this project..


i know, its hard to say, and harder to accept the fact, we are all doin it wrong....


machine must be rigid, strong, i know we all wish more speed etc, but difference in bearing weight isnt so big...

never again thinner then 12 mm for me...

Edited 3 time(s). Last edit at 07/24/2013 04:41AM by woo.

woo, funny that you are saying this, but that is why I opted for 12 mm smooth rods for my next machine! On my Prusa i1 I beefed up the supports and springs holding the heated bed in place. It helps a little, but as you say, the entire machine is too flimsy.

Actually, I clamp my Prusa at the apex near the motor to the cabinet it is in, to reduce X axis wobble!

---

Yvan

Singularity Machine

I do not use squares for machinery. I have a bunch of squares, the only one which is remotely square is a compact Brown and Sharp that is at least 70 years old, it's companion, a 12" starrett is worn and no longer square. I bought a 24" Empire level which is not a true machinist level, but for under 30 bucks promises accuracy to 1/50,000 inch. So first you level the top, and then plum the verticals in at least 2 directions, then use a plumb bob. This is how I aligned my frame before welding it. Consider this, if you are using anything other than linear or machine ground components for your frame, a square, even some $300 square is going to be of little help, if the components are not perfectly straight in the first place. It may or not even get you close. My frame is structural steel. Rolled steel always has a slight arch to it, so I don't bother aligning the frame, so much as the holes. The frame is just there for stiffness, the holes do all the work, thus these are really what must be aligned. And you can only do this with a level. For a prusa, you would level the top and work to the sides. This will be an iterative process but you will eventually get there with a level.

This is the biggest issue with the Prusa line it seems - flimsy alignment. I'm amazed that nobody has come up with something better yet. Something like a couple of 45 degree angle beams added under the bed, along with another two shorter ones attaching to the frame would keep it level theoretically.

Maxx, you are making me re-think my approach. I start with squares for the basic assembly, but of the 5 combination squares I have, none are accurate. This is what happens when your elders pass away and all the tools get funneled to you! If I had one good machinists square I could at least file the other squares back into shape.

After basic assembly I also have an inexpensive digital inclinometer. It is the size of a stepper motor and fits around a RepRap much better than the blue 24 inch Empire level that I also have. I can see how you use your level around your big machine, makes sense!

The final step is usually testing things with maybe a pencil instead of a hot end, and tracing lines on paper taped to the bed glass. I have drafting squares that are not too bad for that final step, but they don't deal with the vertical dimension, or Z axis, only tracings on paper.



Rick, yes with the Prusa, and the Mendel which is its direct ancestor. I think if a trapazoid had been chosen for the footprint shape instead of a square or rectangle that would help with some issues. The latest versions of the RepRapPro use a diagonal brace at the base which helps a bit.

But the original idea was replication, not just making plastic parts. I believe it is a quantity vs quality thing. We are now past the first big wave with this RepRap approach, and people are trying to find the next step up.

---

Yvan

Singularity Machine

I tried to align it the best i could and then would move one side using the two cross threaded rods till the part was dimensionally accurate

Just a note, I aligned my Y axis by drawing a line down the middle of the build plate. Then heated the extruder until I could slip about 1/2" of wire into the nozzle about half way. After letting it cool I ran the bed forward and back checking that the point of the wire ran true with the line. Amazingly it was perfect. A miracle considering the plexiglass construction.