Geeetech Prusa i3 Autobed Level

I'm new to 3d printers, but I have been doing research for the last few weeks, and I am really impressed by the Prusa i3. I also like this geeetech model as it's cheap and comes with ramps 1.4 and mega 2560, the mk8 extruder, and supports printing with flexible PLA. One thing I would like to do is add an autobed level to it, and I've seen a few kits for other Prusa i3 models, and I'm wondering what I need to check as far as compatibility to make sure that they will be compatible with the geeetech version.

Here is the specific model I'm looking at. [www.geeetech.com]

I don't own that build but I've added a microswitch and servo probe to my prusa i3. You'll probably need a way to attach the probe, so you'll need to adapt or download an extruder or hotend holder. In the firmware you'll need to enable auto bed levelling. The are two types, three point autobed levelling and autobed levelling grid. I use the latter. It's also highly recommended to have an lcd to set the offset between the z height of probe vs nozzle.

forget auto bed level. It s gimmick that wont serve your print quality.

The easy way often has consequences ....never more apparent then here

bigfilsing,

I assume you are saying that based on you own experiences, but I can say that mine is the complete opposite.

Quote
ggherbaz
bigfilsing,

I assume you are saying that based on you own experiences, but I can say that mine is the complete opposite.

Of course and if it works for you great.
In my opinion by the time you take all the factors into consideration ...sensing unit accuracy ( micro switch) servo arm set up etc i dont think its that much of an advantage. If you then consider the "work load" it puts on the Z axis motion system you must also consider the anti back lash
mechanics / tension/ stiction and wear and tear in that area. Im confident it wont be long before people realise that simple threaded rod isnt up to the task of being continually driven up and down and that even with constant lubrication when stainless steel binds its almost as good as a weld.
Proper anti back lash ball screws are a different story, but still need good lubrication. Unfortunately we are still looking in the US$ 100 + for anything decent.

I don't use switches for Z I use a capacitive sensor. I agree with you that the servo and micro switch setting is quite unreliable (servo glitches, switch arm flexing if present, proper arm angles, etc) but new methods are more precise and reliable, I have few hundred hours of printing with my setting and no problems whatsoever And it didn't cost me but few dollars.

Quote
ggherbaz
bigfilsing,

I assume you are saying that based on you own experiences, but I can say that mine is the complete opposite.

I am interested in adding autoleveling to my printer. Can you post information or links on how to do it and the parts that you used.

Greatly appreciate it.

Ok people may say its a gimmick and people have told me a perfect level bed in 1st place then you will never need to use it, well my experience is different did the paper trick every single time to get correct nozzle height that was hit and miss to the point of frustration so i decided to use the autobed level system and now i dont have issues getting the 1st layer down, i do have a issue which was down to the m5 bolt not seated correctly in x idler, so it has save me the frustration, its each to the own if it does a job and works why not implement it

Edited 1 time(s). Last edit at 05/04/2015 06:38PM by chris33.

---

Check my rubbish blog for my prusa i3

up and running
[3dimetech.blogspot.co.uk]

Auto-bed leveling is a crutch. It is contrived to compensate for sloppy materials, design and build.

You are attempting to solve issues cause by variances in your system. Control / correct the variances, and you will not need "auto bed levelling" gimmicktry.

Case in point:

I have a Prusa i3 with the acrylic frame, back support "wings", and acrylic Y-axis bed mount. This is the first problem. The acrylic frame Y axis is crap. It bends too easily at the corners, and warps in the wind.

Problem #2, unsolid build surface. The PCB heater is not a build surface. Some newbs, myself included, will attempt to affix tape to the PCB and print on it. The pcb bends and warps during heating. Because that's what stuff does when it heats, it changes shape.

Problem #3, non-parallel interfaces. This means that the X axis rods need to be parallel to the bed across the X axis. It matters less if the Y axis is actually physically level to earth.

So do some tests. Get a line level, and place it on your top X axis bar. Use the two x axis threaded rods to move left/right side up or down until it's level.
Now, use little line level lengthwise and place it between your bed on the left edge under the X axis bottom bar. Move Z axis down until the bar touches the end of the level. You're just using the level as a measuring stick here...

Move the measuring stick to the right side. Does it touch? Too high? too low? Now adjust the bed (not the Z axis rods) until left and right are the same. Now your bed and bars are parallel.
Move the bed from Front to Back corners and recheck your measurement. Adjust the bed adjustment screws until it matches all four corners. As a measure of curiosity, you can use the line level on the front and back of the bed just to compare...But paralell is typically more important than actual levelness. The term "auto bed leveling" is a misnomer anyway.

Before worrying about servos and such, you should focus on a proper build platform (your foundation): stable/parallel Y rods (zip tied to the legs), tight base threaded rods (nylock nuts), premium precision linear Y axis rods (C60) are tons better than Ebay cheap drill rods; linear bearings with little or no side-side play; Aluminum Y plate; PCB heater; Aluminum/Glass Build surface; bed adjustment screws with springs and nylock nuts.

That makes a solid, stable surface, with minimal variables. With that, I suggest, you do not need the autoleveling gimmicks. I might add that the above will solve more problems for you than a servo alone.

Lastly, the end result of "auto bed leveling" is to have the Z axis move up and down to the contours of the bed during printing. This will impact X/Y movement speeds overall.

Cheers,
Mark

Edited 1 time(s). Last edit at 05/06/2015 01:35PM by thetazzbot.

thetazzbot,

All that sounds good, but by the time you do it all your printer goes from 400 to 4000 dollars. We all know that a 10 or 12mm linear rod is better than the cheap 8mm ones, we all know that an all metal frame is better than wood, acrylic or plastic, that lead screws are better than threaded rods, etc, etc.. But coming back to earth where people want to get a printer without needing a second job to pay for it, a cheap 400 dollars printer with auto bed leveling it is perfectly fine and does the job.

My little printer that uses 5/16 threaded rods and printed parts as frame, 8mm linear rods for all axis M5 screw for the Z axis, a cheap Mk2 with glass on top, a geeetech PEEK j head with Chinese mega and ramps board that in total probably cost me 300 bucks to make print as good and as reliably as a 2000 dollars Lulzbot taz 5.

With my CTC replicator without auto bed leveling, every time I change the glass plate or change material I have to calibrate my bed. Yes it's simple and doesn't take too long but it is tedious job that ABL eliminates, with my little printer, I just hit print and that's it, have never lost a print due to first layer detachment or delamination.

Edited 1 time(s). Last edit at 05/06/2015 02:16PM by ggherbaz.

Well, I was actually referring to the weakest points. The Y axis bed thing that connects to the bearings; I changed mine (acrylix) to an $18 aluminum. I changed the cheap drill rods to premium linear rods pulled from junk printers ($5 each). So, no, I didn't suggest drastically changing the cost. I suggested addressing the weakest, easiest variables.

why would you have to calibrate the bed when you change material? ABS or PLA or Nylon, shouldn't affect your bed level... Not sure why you'd change the glass that often. But anyway, The servo was fun to tinker around with (for me) but it sucks because the cheap servos are well, cheap and I just had too much headaches with it not goin down all the way, or the stupid printer draaaagggginng it across the bed without raising Z... ugh. yeah, calibration sucks.

Edited 1 time(s). Last edit at 05/06/2015 05:53PM by thetazzbot.