Does Your Rostock MAX v2 Hold Its Calibration?

[dkaustin]

The experience of various users seems so disparate that I am curious what the general experience has been.

[mvansomeren]

It probably is closely related to the build process and making sure all of the mechanical aspects are correctly installed and functioning (i.e., no binding, looseness, etc.) My printer has held it's calibration since the day I built it and I have about 408hrs of use on it so far.

[BenTheRighteous]

I'm recalibrating all the time and I think it's due to the weather. I live in a humid area and it really has an impact. Is it hotter this week than last? Do we have the windows open, or are we running the A/C? I've noticed that if the environment changes, it tends to affect first layer quality and I find myself tweaking endstops more often than I'd like. If I could locate it in a constant-climate room, that would make things a lot easier, but oh well...

[CRPerryJr]

I only need to recalibrate when doing preventive maintenance to the printer, such as checking and tightening the cheapskates or belts. Other than that, I pretty much just throw on the glue stick and go.

One thing that helped me with my calibrations is to extend how far out the x, y, and z tower macros go. It's helped a lot with larger (Wider) prints.

Z Tower

G28 ; Home All Axes
G0 Z0 X0 Y110 F3500

Y Tower

G28 ; Home All Axes
G0 Z0 X87.94 Y-55 F3500

X Tower

G28 ; Home All Axes
G0 Z0 X-87.94 Y-55 F3500

[jlmccuan]

I have 2 Max's. They just don't require recalibration, either one. There are 500 odd hours on the first and 200 or so on the second. I load them up in my truck with the passenger seat laid back and carry them by grabbing the Y tower and lugging it like a suitcase. I did make some shims to lock the towers in place and carry some of the stress that cause the machine screws to open up the holes in the melamine. I'm not sure why some folks seem to have to repeatedly make adjustments. I'm just glad that I don't.

[3D-Print]

I have only had to recalibrate when I have moved or removed the print surface. Otherwise with about 600 hours is money on!!

[KAS]

I would say yes my printer holds calibration although I do see fluctuations in overall Z-height sometimes. Even-though it holds well, I still like to tinker with it trying to better my previous calibration.

Okay, I voted. Who said they only calibrated it once during assembly?

[EL Cuajinais]

I used to have to set my Z-height frequently for the first 2 months I had the printer. It seems I've been going for close to a month now without the need to recalibrate. I've checked about two or three times and it is always were it needs to be. My horizontal radius has been the same since I assembled the printer. While trying to print to the edge of the bed I tried to recalibrate the horizontal radius twice, and found both times I had set it well in assembly, no additional tweaks to horizontal radius have been needed. That said, I can't print to the edge of the bed, but then again, I looks like that's a fairly common issue here.

So:
Horizontal radius is untouched from assembly.
Z-height required many re-calibrations initially but has been stable for the last month.

BTW I copied the code from the above poster and will use those locations the next time I try to do the horizontal radius. Thanks!

[MrJoe]

I calibrated the Rostock Max V2 when building it, and many times thereafter while I was learning about the printer.

I've upgraded to the PEI print surface and the bed hasn't been removed since then.
I check the calibration often and it usually does not need adjustment.
I was using glue sticks prior to the PEI upgrade and had to remove the bed to clean it occasionally and this may have contributed to the need for calibration.

I use a Cartesian printer at work and it requires calibration more often than the Rostock.

[3D-Print]

I just saw Mhackney posted this and it may be a part of what is being discussed here.

My printer which as noted above is very stable and really doesn't require calibration is in a stable room without much fluctuation in temperature. As well with my enclosure and halogen lights, I try to keep the chamber temp at about 40 degrees C. This may be part of my stability!

http://forum.seemecnc.com/viewtopic.php ... 725#p71237" onclick="window.open(this.href);return false;

[BuckeyeVolunteer]

I've had some success in keeping it's calibration but mine is in a garage it is kept in a heated and air conditioned environment but, I do notice z height differences between PLA and ABS due to the temps. It maybe due to a warped heater plate, also I found the screws that hit the home switches are rather a bad idea they can and will move the more you home it because the area holding them is very weak. When my kit arrived I had black rollers for the cheapskates and blue rollers so the black are on one side and the blue on the otherside of the cheapskates whether this makes any difference it wasn't covered in the instructuction manual. I have mine sitting on a block of marble which is leveled and the base is leveled to the same as the marble setting on a metal cabinet. I am sure I am a dummy and have done something wrong but it's been a struggle to get good prints. The new matterhackers update really screwed me up. then I added the new updated arms and haven't had any consistency since. Maybe I should tear it apart and start over again.

[dajay23D]

I'm constantly recalibrating and adjusting Z. However, this printer is awesome and prints fine if you don't care and just calibrate it once and leave it alone. I'm always picky and trying to get some real accurate #s and I just can't. Here's what my v2 does;

1. My Z height moves if I check it everyday. Literally. It moves .10-.30.
2. I can only choose 1 calibration procedure. If i try both, v3 of manual, end stop screws heights won't be the same.
3. I like the the alternate cal procedure in the v3 manual (std procedure in v2 manual). It saves a bit of time since it doesn't have to run thru all the axis before you can adjust. But, I'm confused if I need to adjust Z every time I adjust the end stop screws like the other procedure. I can get all 3 axis to be fairly close, but the bed center will never equal the axis, regardless of horizontal radius. The more I adjust, the lower the build height gets. If I keep trying, I'll run out of threads on end stop screw.
4. Issues from 1-3 is in constant loop.

With all that said, and as mentioned earlier, I can print really good and even fairly level. I just have live w/ the weird things and just print. it seems like my printer likes 130.25 for hr.

[KAS]

I would say calibrating your overall Z-height would be necessary anytime the endstop screws are touched and/or endstop values in the EEPROM are changed. Also if you adjust your horizontal radius, just to make sure.

Temperature and humidity can definitely fluctuate the values. I find myself printing a test print before normal use Just to verify that my first layer is laying correctly. The minor Z height correction can be adjusted in the slicer, at least with Simplified 3D.

Also, when you change the horizontal radius, the center height doesn't change. The height of the effector changes towards the edge of the platform, the center stays the same. You'll want to find center of the bed height and change the HR so the outside X/Y/Z match the center.

[ccavanaugh]

Something else to look at is belt tension. If the idlers/tensioners are slipping, it can cause minor changes in apparent z-height for a tower. I've noticed a 0.2mm change in z-height after retensioning.

Belts may stretch a little on a new build as well and then stabilise.

[dajay23D]

Also, when you change the horizontal radius, the center height doesn't change. The height of the effector changes towards the edge of the platform, the center stays the same. You'll want to find center of the bed height and change the HR so the outside X/Y/Z match the center.

thanks for the clarification. This is what confuses me the most... Many say hr doesn't change height, including the manual. But, it also says to change horizontal radius to raise bed height. Either way, I've tried numerous things, and it's just not working. I've played w/ belt tension numerous times.
So the chart for alternate method on 3rd edition manual should say; Adjust X Tower End Stop --> Adjust Z height --> Adjust Y Tower End Stop --> Adjust Z height.

[geneb]

No, it doesn't say change the horizontal radius to raise the bed height - at least how you're implying it.

Changing the HR affects the concavity or convexity of the mathematical model used to define how "flat" the effector travel is.

Follow the flowchart.

g.

[captainpotatofudge]

In my efforts get calibration sorted out I came across this thread. NOPE my calibration doesn't hold worth a darn. I've had my Rostock Max v2 since Christmas week 2015. It has held calibration at most 4 days in a row. The rest of the time I'm constantly chasing something, when I nail one number down two more shift. Then I nail them down and another changes. I frequently see changes from print to print like something is loose and moving around, but I sure can't find it. The room it's in stays 68-72F and 35-37% humidity FWIW.

[dchauran]

I find I have to do a minor calibration tweak every couple of days. I've only had mine for less than a month. I've been meaning to re-tension the belts, and recheck the towers for squareness, but so far have not had the time when it just takes a minute or so to recalibrate (I have a mount for a dial indicator always installed).

[Eric LB]

After building my Rostock last month and printing the fan shrouds, I designed some large prints which will come close to maxing out the build area. So I decided to recheck calibration ( which I have been doing with feeler gauges). The first thing I did was go into the terminal to plug in the coordinates for the tower positions 125mm (further than in the user manual, see the post above) from the center of the bed. So far so good : excellent results and horizontal radius is set just right. Next I decided to check calibration 125mm from the bed center IN BETWEEN THE TOWERS (let's call those three locations A, B, and C). To my surprise the z height increases .2 mm to .3mm as you go out from bed center to points A, B, and C. This is hugh and I can predict before wasting any filament, that it will render large cyllindrical/ circular objects unprintable (due to first layer adhesion issues even if one overlooks the geometrical innacuracy).I saw a forum post about this and now can't find it again. The consensus was that it is a firmware math problem with Delta arms. I would best describe it as a horizontal radius problem BETWEEN THE TOWERS ( thus not detectable during any of the normal calibration procedures outlined in the manual). The G code coordinates I came up with for A, ,B annd C are: G1 X-107.94 Y62.32 Z(can be any number you want to use with feeler gauges, I use 0.2); G1 X107.94 Y62.32 Z0.2; G1 X0 Y124.64 Z0.2. As I recall, as of the end of the forum thread, nobody had any answers except more bed screws. But the v2 has screws already in the six positions (X,Y,Z,A,B,C). If the answer is to use shims, can the glass really be warped that much by shims and those black clips. The glass looks pretty flat to me. Of course my bed and hot end are at operating temp for calibration/ all my measurements. Based on the thread I refer to, I know others have or have had this same issue.

Can anyone help with this or refer me to anyone/where? Maybe I should do something with horizontal radius, but I don't know what and I think it would mess up the z at the towers. Maybe I should try taking the firmware out, and downloading again since I did it the first time with a Windows XP computer ( I have since acquired a Windows 10 computer to run the printer). I just don't know.

I second the comment about the end stop screws loosening the more you home and the more you adjust them. They are tight in the plastic carriages the first time you wind them in, then get looser with every little adjustment. Any suggestions there?

[nebbian]

After building my Rostock last month and printing the fan shrouds, I designed some large prints which will come close to maxing out the build area. So I decided to recheck calibration ( which I have been doing with feeler gauges). The first thing I did was go into the terminal to plug in the coordinates for the tower positions 125mm (further than in the user manual, see the post above) from the center of the bed. So far so good : excellent results and horizontal radius is set just right. Next I decided to check calibration 125mm from the bed center IN BETWEEN THE TOWERS (let's call those three locations A, B, and C). To my surprise the z height increases .2 mm to .3mm as you go out from bed center to points A, B, and C. This is hugh and I can predict before wasting any filament, that it will render large cyllindrical/ circular objects unprintable (due to first layer adhesion issues even if one overlooks the geometrical innacuracy).I saw a forum post about this and now can't find it again. The consensus was that it is a firmware math problem with Delta arms. I would best describe it as a horizontal radius problem BETWEEN THE TOWERS ( thus not detectable during any of the normal calibration procedures outlined in the manual). The G code coordinates I came up with for A, ,B annd C are: G1 X-107.94 Y62.32 Z(can be any number you want to use with feeler gauges, I use 0.2); G1 X107.94 Y62.32 Z0.2; G1 X0 Y124.64 Z0.2. As I recall, as of the end of the forum thread, nobody had any answers except more bed screws. But the v2 has screws already in the six positions (X,Y,Z,A,B,C). If the answer is to use shims, can the glass really be warped that much by shims and those black clips. The glass looks pretty flat to me. Of course my bed and hot end are at operating temp for calibration/ all my measurements. Based on the thread I refer to, I know others have or have had this same issue.

Can anyone help with this or refer me to anyone/where? Maybe I should do something with horizontal radius, but I don't know what and I think it would mess up the z at the towers. Maybe I should try taking the firmware out, and downloading again since I did it the first time with a Windows XP computer ( I have since acquired a Windows 10 computer to run the printer). I just don't know.

I second the comment about the end stop screws loosening the more you home and the more you adjust them. They are tight in the plastic carriages the first time you wind them in, then get looser with every little adjustment. Any suggestions there?

There are two problems that can cause the "high between towers" issue.

One is that you have your steps per mm wrong. This is unlikely.

More likely is that you have a twist in the top of your frame in relation to the bottom of your frame. I don't know how you would go about sorting this out with a Rostock Max, but I've seen this symptom on both my builds, and in both my builds it was resolved by carefully ensuring that the twist was removed.

[Eric LB]

Thanks for the suggestion. I think my steps/mm are okay because my fan shrouds came out well, but I readily agree that tower twist is quite possible given the fact that the melamine is so subject to deformation and the design of how it holds the towers. The process of getting all tower angles 90 degrees to the bed was difficult and I wouldn't be at all surprised if some subtle twist was introduced and not detected. The huge question I have is: How do I get the twist out ( yes loosen tower attachment screws and apply opposite twist force to the top but HOW do I measure to know the twist is out??? measure diagonals between bases and tops of adjacent towers??? Would that be accurate enough? I fear not. How did you measure for twist? How did you you know you got it out? Once it's out how does one prevent it from recurring after the slightest bump to the machine (short of going through the whole calibration that I used to detectmitnin the first place? I wish these frames weremalot more rigid. I already know that one of my towers is about 1mm taller than the others - but tech support said that will be okay and can be tuned out with end stop screws. But that would in fluence any diagonal measurements - which could be accounted for with a the pathagorean theorm.

Will getting the twist out then require me to reset my horizontal radius ( which is perfect now)?

[Xenocrates]

Personally, I'd use a trio of squares on a tower, one facing each other tower, and one facing inwards, and work until the whole assembly comes out balanced on all three. Use the slot to hold them in somewhat, and use a bungee cord or similar to hold them against the tower. It won't be perfect, but should reduce the effects. They are fairly good at holding calibration once the bolts are in (See Mike taking his for a ride in his convertible and printing immediately afterwards) but when the bolts are loose, such as when you are adjusting the tower angle to the base, it's much less rigid (Which is the point)

[Eric LB]

I checked with the three square positions on each tower, plus I ran perpendicular measurements between towers, and sited down the Rostock top and base edges: no twist; everything amazingly square. Then rechecked calibration: it has not changed. Things are very close at x,Y,x, 0 but .25 mm, .33mm, and .48mm high between towers. What next?

[Eric LB]

I worked on the calibration at 0,X,Y,Z some more and got more uniformity in those "high between towers" z values. They are now .254mm, .254mm, and .33 mm. I think if calibration was perfect they would all be equal at around .254mm, and so represent a mathematical firmware phenomena. Maybe the firmware thinks my towers are twisted when they are not detectable as such. If that is the problem, it's not something I can straighten out by eye and carpenter's square(s).

[Eric LB]

My bad my bad!! What I should have done rather than writing all those posts was: call Matter Hackers support.

The following was learned: The variances in tower spacing are within accepted cnc cutting norms (I of all people should have known this being an cnc guy; my bad). The way to eliminate the Opp tower gaps and really dial the calibration in after having calibrated with endstop screws is: in Matter Control go to options then calibration, turn it on and do the 13 point calibration.A guide can be downloaded from Matter Hacker support. The prompts lead you through probing the bed in 13 places. The software will tell the printer what gcode offsets are necessary to make the nozzle travel in a flat plane. voila! In suggest that geneb mention this in the manual as well as some introductory comments emphasizing that the go to resource for any problems that have anything to do with programming, or matter control, is MatterHackers support, not Seemecnc support. There are a lot of us newbys out there that need this pounded into our heads ( at least I did).

For moving end stop screws: after using them for the very first rough calibration - put some crazy glue on them and that should move you into the ",only calibrated once" category above. Hope this helps. Though I never asked, I believe the 13 point calibration procedure in Matter Control can also be used for the Kossel pro.