Rostock off centered

[m4r1n5]

Ive been having this issue since I put this together. But with the bad board taking me a while to figure out, I wasn't able to deal with it til now. The issue is my X-Axis is off by about 18mm compared to Y&Z. Here are some numbers to help picture this out.
1) From the heat bed to all 3 endstops is 670mm (+/- 1mm)
2) From the base to the top "plate" is 702 (+/- 1mm)
3) "G0 Z0 F3500"= 18mm off the center of the heat bed.

This is what My X-Azis looks like at G28:
[img]http://i1326.photobucket.com/albums/u65 ... 446ff2.jpg[/img]

This is my Y&Z Axis on the endstops at G28:
[img]http://i1326.photobucket.com/albums/u65 ... 23bb24.jpg[/img]

At present, I am calibrating per genebs instructions:

1. EEPROM 0
2. Upload
3. Change printer radius
4. Adjust end stops
5. Test - if not correct, goto 3
6. EEPROM 1
7. Upload.

g.

Need help on this. My brain is at a blank....

[geneb]

Replace the screws you've got installed with the ones that came with the kit as a first step. I have no idea why you installed those absurdly long screws, but don't think I'm going out on a limb when I label them a "contributing factor".

The only way your X axis could be 18mm off is if the head of that silly screw is 18mm offset from the position of the other two end stop screw heads.

g.

[m4r1n5]

I put those silly screws in BECAUSE of the massive offset I was getting. The factory screws are 2inches and I was having to adjust with having 1to 2threads left while my x axis was screwed all the way in. I had to get those screws because they're 3inches and I needed all of it. They are not the problem, they were my only solution until one better comes along.

If u need me to video how I calibrate, I'd be more than happy to so I can get past this.

[Eaglezsoar]

I see that the little bump on the switch is hitting on the edge of the screw that can't be good for repeatability.
I personally would flatten out that "bump" so that the screw's head has full contact with the switch lever.
I am not telling you to do this, I am simply saying that it is what I would do to my printer.

[m4r1n5]

Thanks, but would that cause what im experiencing?

[geneb]

The issue isn't the switch.

There's something very mechanically wrong with your build if you've got an 18mm offset when you compare the X to the Y & Z (only 2 pics appeared in your original post, so I don't know what all three look like).

I've personally built 3 of these things and the 2" screw is not only adequate, but works well when threaded nearly all the way through the mount.

Did you happen to install a CheapSkate upside down? That's the only rational explanation I can think of for the offset you're seeing. Another reason could be that one set of arms are too long. Did you measure all three sets to verify they're all the same length?

g.

[Nylocke]

His cheapskates appear to be mounted correctly, are all of your aluminum extrusions the same length, And are you sure that from the base of each tower to the top is the same? I really highly doubt it in seeing this.... Did you follow gene's second edition build guide?

[geneb]

Good point Nylocke. If the top of the machine is tilted, that would give the error he's seeing as well. I've only seen pictures of two of the three cheapskates, so I don't know for sure if they're all ok.

g.

[Nylocke]

I'm assuming he can recognize basic patterns and isn't a drunk monkey, and seeing he got it right twice, I think he would have it thrice as well. Besides, I think he would have bigger problems if that was upside down....

[geneb]

After some of the builds I've seen, I don't take _anything_ for granted any longer. I'm surprised I don't have a drinking problem already.

g.

[m4r1n5]

I measured from the bottom(under the heat bed ) to the top and got this. all with 1mm diff.
[img]http://i1326.photobucket.com/albums/u65 ... 8b5404.jpg[/img]

Also have a lean, apparantly.

[youtube]https://www.youtube.com/watch?v=e2UUEL3 ... e=youtu.be[/youtube]

Im running this with eeprom off right now.

[geneb]

Did you build your machine against the first edition of the manual, or the second?

Measure the height of the base at the position of each tower - from the top of the top plate to the bottom of the bottom plate - are all three dimensions the same?

A 3/4" offset means you've got something mechanically way off.

g.

[geneb]

...and I just noticed the markings on that tape. If that tape is marked in inches and you're measuring from the top surface of the base top plate, you're at roughly 690mm, which is nearly double that of a stock Rostock MAX.

Somehow I don't think we're getting the whole story on this machine.

g.

[Nylocke]

Are you measuring from the base of the tower or from under the ONYX?

[dpmacri]

...and I just noticed the markings on that tape. If that tape is marked in inches and you're measuring from the top surface of the base top plate, you're at roughly 690mm, which is nearly double that of a stock Rostock MAX.

Somehow I don't think we're getting the whole story on this machine.

g.

His measurements agree with mine. I think, Gene, that you're thinking of the print volume height (360mm) which is just over half of the actual height of the cylinder (700mm)

That said, m4r1m5, please take some pictures of the entire machine so folks can see that start and end of the tape measure -- that would take out the confusion.

[lordbinky]

I'm curious.... Has M4r1n5 printed anything with it like this? Are the prints correct? I'd question the motor if the construction is even fairly correct.

The other thing is, what are are your settings for these values

X/Y/Z Max Length
DELTA_DIAGONAL_ROD
PRINTER_RADIUS
CARRIAGE_HORIZONTAL_OFFSET
END_EFFECTOR_HORIZONTAL_OFFSET

How DELTA_RADIUS is handled has always irked me during calibration. It just doesn't seem right for how I picture the ways it shifts the edge positions. I may be picturing it completely wrong and I'm not diving into the code to find out (for now), but if the nozzle is equal at the base of all three towers, then making the curve between edge and center more or less concave/convex should NOT change change that equality. (the way I picture it) Those points should stay equal and the radius change be completely compensable by changing the Z-height accordingly. It doesn't always work that way though (if it should work that way, it just makes me think there is skew in my towers I can't tell).

[m4r1n5]

@lordbinky- I have printed with this and came up with torqued looking calibr cubes and a few decent 20x20x10mm boxes, but they all had a lean to it.

SRY about not replying soon enough. Graves+kids= NO TIME FOR ME!

After being hammered by geneb and others on how straight my towers were (and they are) I started putting pics together per dpmacri request. I normally used the large right angle ruler(cuz the manual sez), but I'm taking pics so I used my small leveler right angle and got Z[img]http://i1326.photobucket.com/albums/u65 ... b6423c.jpg[/img]

THEN Y..........[img]http://i1326.photobucket.com/albums/u65 ... 66f13d.jpg[/img]

THEN I GET TO X. My problem child....[img]http://i1326.photobucket.com/albums/u65 ... 5602eb.jpg[/img]
UUUMMMMM...>WHAT???? How could this side be this off on level....I even checked my table.
[img]http://i1326.photobucket.com/albums/u65 ... 63b56e.jpg[/img]

AFTER geneb has woken up after passing out from the shear pain of pulling out his hair, and others have picked themselves off the floor from laughing....COLOR me stupid and guide me on where to fix this.....

[geneb]

Not stupid, just inexperienced. That comes with time (and even DUMBER mistakes. Trust me on this. )

Is it just me, or do the first two photos show a complete adjustable square, but the third photo has the ruler missing?

Another thing to check is to make sure that the top is squared as well - you can help this by using a trimmed down wood yardstick as a height gauge that will allow excellent repeatability.

Check out how I show it being used in the 2nd ed. assembly manual.

g.

[m4r1n5]

Not stupid, just inexperienced. That comes with time (and even DUMBER mistakes. Trust me on this. )

Is it just me, or do the first two photos show a complete adjustable square, but the third photo has the ruler missing?

Another thing to check is to make sure that the top is squared as well - you can help this by using a trimmed down wood yardstick as a height gauge that will allow excellent repeatability.

Check out how I show it being used in the 2nd ed. assembly manual.

g.

I took the ruler off because it was also square, but I was making the emphasis on my heat bed being woefully unleveled. The hieghts have all been fairly the same (give or take 1 or 2mm). I will recheck the top being squared also. I don't see how it changes why the heat bed is lower on the X tower side compared to the other 2. If the top ends up being squared, Will i need to add "shims" under the bed to get it leveled?

[Nylocke]

Don't measure from the heated bed!!! I was trying to figure out if you were doing that or not. Just get rid of your bed until you have your towers properly leveled and such, your heated bed introduces way to many variables into an already frustrating equation. make sure you have the machine square without it, you can account for an unlevel bed by calibrating it with the endstop screws.

I didn't even square my towers by the way gene, I just got all of them at the same exact height and it prints beautifully. No lean, no problems.

[m4r1n5]

......So per request, I was going to make a quick video of my 'stock and show that all was level, staight and still messing me up. As I was resquaring the towers and ran "g28" I ended up with this [img]http://i1326.photobucket.com/albums/u65 ... 10497a.jpg[/img].....WHICH happens to be from the X-tower. Same one Im having massive issues with.

This has to be a defect, cuz i havent printed really anything pn this thing.....just more money to sink in I guess.

[dpmacri]

When did you get your kit? Based on your post of 9/11, I assumed you had one of the newer kits (like mine which I got in August) that have all aluminum pulleys. But yours appears to be one of the older ones with plastic pulleys.

[m4r1n5]

I bought mine in June. I thought they went away from the all aluminum ones since they're selling mine on the website right now.

[geneb]

They'll replace that gear for you - it's defective. I'm willing to bet that's the root of your entire issue too.

i'd check and replace the other ones as well.

g.

[m4r1n5]

Im sure they would. Theyve replaced the Arduino (bad z-axis issues) board and the adapter board( my bad on the soldering). I'm going to buy the pulleys because I want extras just in case. Theyve done a great job working with me and my limited knowledge of 3d printers. Hopefully I'll get back to working the printer by wed or thur.