Max v2 Calibrating Issue Between X and Y Tower

[Pyropainter]

Hello

I have recently calibrated my printer according to the Rostock Max v2 users manual. . All measurements at the towers are within .001 of an inch, using a travel indicator in place of the hot end. I also generated a few buttons in Matter Control that would allow me to check the position centered between each tower. The positions between the YZ and XZ were within about .002 inches. The front position XY was low by about .008 inches. I have tried researching this problem but have not found any answers. I am certain that this is not just a low spot in the heated bed, my new build surface is a piece of 1/4 inch mic 6 aluminium that has been lapped for flatness. Also the bed was rotated several times and retested. The end result was always the same.

Thank you for your help.

[ZakRabbit]

From what I've read, you might want to go back and verify that all of the universal joints for the arms are completely free of burrs, and move free and smooth. Also verify your cheapskates are adjusted properly, such that all four wheels are making equal contact with the metal extrusions.

[Mac The Knife]

I've manipulated the tower rotation settings in the eeprom settings to get a flat reading with a dial indicator. All was fine until the humidity jumped from 15% to 60%,,,

[Pyropainter]

The first thing I will do is go back and check for anything that may cause any mechanism from the Cheepskates thru the end effector for binding. I may be a day or so before I can report any results.

Thank you for your help.

[Pyropainter]

I've manipulated the tower rotation settings in the eeprom settings to get a flat reading with a dial indicator. All was fine until the humidity jumped from 15% to 60%,,,

Mac The Knife
You speak of manipulating the rower rotation settings in the eeprom to aid in calibrating your heated bed. This is my first printer, but I have spent many hours researching the net on everything related to 3d printers (still so much more to learn). Your method is new to me. Would you care to explain. Or recommend a web page that will allow me to make the adjustments.

Thanks

[Mac The Knife]

Here's a thread that covers the frustration, http://forum.seemecnc.com/viewtopic.php ... d+behavior

[Pyropainter]

Thank you for the information. I will check it out tonight but, may be a while before I can put its information to use.

[ccavanaugh]

Another good link http://www.thingiverse.com/thing:745523

It includes a calibration object that is insightful.

I learned a hard lesson that I think contributes to some peoples issues. I struggled with squareness of printed parts issues for some time. I went the route of laying out the primary plate on a CMM, etc. Towers were perfectly square and the plate was accurate, and yet, I still had issues.

After reading the above link and thinking it through, I came to the realization, that the issue was in the arm alignment on the carriages and not the tower alignment. I fixed the issue by carefully clearancing the alignment holes on the offending carriage (the holes the u-joint mounts in), and shifted the u-joint mount about 0.015". After that, I can now print the calibration part from the above link and a 60 degree drafting triangle fits all 6 sectors perfectly. This was all done without the need to fudge firmware rotations.

If you don't want to file or tune the parts, get yourself some TrickLaser arms and set the widths a touch wide so you can shift their position laterally withing the u-joint mounts and the carriage. The end result will be the same.

[Pyropainter]

Ccavanaugh

This machine came with Trick Laser arms.

In your reply you mention your method of tweaking your machine to allow it to accurately print the Asterisk file. I have not tried to print that file yet. All I have done is calibrate the bed and adjust the horizontal radius as close as I could. Then checked other positions on the heated bed and found that there was a low spot between the X and Y axis. Did you find a similar problem with your bed that was corrected by using your method to correct the angle between the legs of the printed Asterisk?

Thank you.

[ccavanaugh]

I did not test with an indicator, but struggled with bed leveling in some areas after repeated attempts at tower leveling. I could see the difference on the first layer of large parts. After making the lateral teak to the arm, bed level issues went away. I did not have to re-level the tower after the tweak. Also, I run modified scripts and adjust tower level with the arm in near vertical condition. This reduces the influence of how well the other arms are leveled.

[BenTheRighteous]

If it's just one spot on the bed you seem to be having trouble with, try shimming the bed. I did that last night (after reading the first 7 pages or so of the thread that Mac linked) and it worked like a charm. I was having trouble with the effector going high between the Y and Z towers, now it's flat all over the bed. Awesome.

[Pyropainter]

BenTheRighteous

I have read that one gentleman was able to calibrate his machine by shimming the bed. However I am now unable to do this because of some modifications that I have made. Here is a link to a post with some pictures that I posted in the "Show Us Your Machine Part of this forum" to give you an idea of what has been done. http://forum.seemecnc.com/viewtopic.php?f=6&t=8447 I am fairly certain that my bed and build surface is now flat, and something else is causing the low spot between the X and Y axis. When I get more time I will dive deep into this but, recently things have been kind of crazy around here. Right now I am trying to I am mainly trying to research possible directions to take when I get a break.

Thanks again for every ones suggestions and help.

[Jimustanguitar]

I have read that one gentleman was able to calibrate his machine by shimming the bed. However I am now unable to do this because of some modifications that I have made. ... I am fairly certain that my bed and build surface is now flat, and something else is causing the low spot between the X and Y axis.

Shimming the bed warps the glass to match the motion of the effector platform. The problem I encountered was inaccurate motion, not a bed that wasn't flat. I also tested with a known flat piece of Mic6 aluminum plate and had the same results, so it's definitely a motion thing.

[Pyropainter]

I also agree that this involves the motion of the machine and or build accuracy of the printer, rather than a warped build plate. I have replaced the glass build plate on mine with a piece of 1/4 thick mic 6 that has been lapped on both sides. I have run tests with a dial indicator and have rotated the plate and re tested, only to have the greatest difference error show up in the same location.

[ccavanaugh]

I have read that one gentleman was able to calibrate his machine by shimming the bed. However I am now unable to do this because of some modifications that I have made. ... I am fairly certain that my bed and build surface is now flat, and something else is causing the low spot between the X and Y axis.

Shimming the bed warps the glass to match the motion of the effector platform. The problem I encountered was inaccurate motion, not a bed that wasn't flat. I also tested with a known flat piece of Mic6 aluminum plate and had the same results, so it's definitely a motion thing.

Agreed. Best to fix the root cause of the problem and not put a band-aid on it.

[BenTheRighteous]

I don't suggest shimming because I think your build plate is warped. I suggest it because I think your build plate isn't level relative to your towers.

[BenTheRighteous]

Can someone please tell me what eeprom adjustments to make in order to get this perfectly flat, unwarped build plate to print without high and low spots without shimming?

Selection_264.png (4.97 KiB) Viewed 11807 times

I don't care how flat your plate is, if it's not level relative to the towers, you're not going to get good prints out of it.

[Dale Eason]

I don't care how flat your plate is, if it's not level relative to the towers, you're not going to get good prints out of it.

Isn't that what the leveling process is all about. You use the end stop screws to get the towers level with the bed.

Dale

[Mac The Knife]

But the bed need to be square to the towers,,,

[BenTheRighteous]

Isn't that what the leveling process is all about. You use the end stop screws to get the towers level with the bed.

That's a really good question actually, and I don't know the answer to be honest. But my guess is that the end stop screws only accomplish so much - perhaps they're more to compensate for the differences in the heights of the end stop switches.

Let's say - just for giggles - that we have a print bed angled at 45 degrees relative to the towers (this is impossibly extreme and no one in real life would ever have this). The center of the print (a.k.a. the centerline of the print bed) on the first layer would still be pretty close to the center of the 3 towers. However once you get to the 100th layer, the center of the print would be nowhere near the center of the towers. It would be skewed way off in the direction that the print bed was angled.

I don't think any endstop adjustment could make the effector somehow "know" about that gigantic offset and compensate as it traveled to higher and higher print layers.

Of course this is a purely theoretical example and not exactly the same issue, but my point is that I believe there is a limit to what endstop and eeprom tweaks can fix before mechanical adjustment becomes necessary.