Welcome to the SeeMeCNC Community

Ever since I built the Rostock, I have never been able to achieve a print which can fill the full size of the onyx bed because my hotend can never fully reach the bed on the left side, while the right side the nozzle is so close no plastic can flow out! (Doesn't tear the tape, but doesn't allow any filament to extrude)

As such I'm currently limited to a base diameter of less than 3 inches for optimal adhesion.

After reviewing the new build instructions, I see Gene mentions the use of a cut to size yard-stick to ensure proper height of each rail. I did discover one was 300-400 micron lower than the other two. However, I am still having an issue with one area of my bed higher than the other.

Any possible suggestions on what else to look for would be great!

how hot are you getting the onyx?

60c, 80c for abs. The problem area appears between the rails Z and X rails, and between the Z and Y rails.

See attached image for example of what happens. Here's a 150mm round calibration print.

If you look between the X and Y rails near the front of the print, you can see how the extruded plastic is perfectly round, not smudged into the bed. While the right side of the picture you can clearly see the print head is dragging so close to the bed that no plastic can come out. There's no doubt if I let this print continue for a full 10mm that I would experience warping of the first layer.

Looking for any possible causes. Initially, I suspected my aluminum heat disspator plate from Mhackney was the culprit, so I have removed that. The Borosilicite glass from SeeMeCNC appears to be perfectly level. Something else is going on here....

One thing that could help is to loosen all the screw holding the Onyx down. Bring it up to temp, then re tighten the screws in a star pattern. If you have a torque screwdriver, it would be ideal to tighten them equally.

Hmm, just tried that (raised temp to 70C to ensure the bed warps)

Trying a 6 hour vase print right now - has a 120mm base - doesn't look like it's going to stick for the full time, but we'll see.

Any other suggestions? I'm seriously considering posting this for sale and using the funds to buy an Orion. I'm tired to fighting with this thing, I want to get to printing my ideas!

i loosened all screws, took it to 100c, and with a plastic level (in line with 2 screws across from each other) pressed down and re-tightened in the star pattern. Also I cut a notch in my snowflake to remove the stress the wires could be forcing on the bed.

I've done everything possible to level the bed itself. I think the cause is something related to the arms or the top of the printer. Is there any other reason why some areas of my bed are up to 400micron higher than other parts??

cassetti wrote:I've done everything possible to level the bed itself. I think the cause is something related to the arms or the top of the printer. Is there any other reason why some areas of my bed are up to 400micron higher than other parts??

Is it safe to assume that you've followed the leveling instructions in the latest manual?

Are you talking about the G0 Z0 X-45 Y-77.94 F2000 instructions for adjusting the endstops? Yes, I have - at all 3 rails, the bed is perfectly level.

But between the X and Z rails, and the Y and Z rails, there appears to be a highspot on one side, and a low spot on the other side.

cassetti wrote:Are you talking about the G0 Z0 X-45 Y-77.94 F2000 instructions for adjusting the endstops? Yes, I have - at all 3 rails, the bed is perfectly level.

But between the X and Z rails, and the Y and Z rails, there appears to be a highspot on one side, and a low spot on the other side.

There was a forum post a little while back about a similar problem (I think ). Whoever posted about it (626Pilot perhaps?) had determined that his posts weren't exactly 120 degrees apart which is an "assumption" of the delta firmware. You should be able to verify this by as accurately as possible measuring the distance between the 3 posts. Or cut a yardstick or piece of wood to *exactly* fit between two of the posts and then see how it fits between the others. If it's off, that could cause the problem you're seeing.

I had this problem until I fixed two things, hopefully they will be able to help you.

1) The instructions say to place a piece of kapton tape on the onyx heat bed where the metal is exposed, I found that it adds a very small bump that actually causes the glass to act slightly like a 'teeter totter' and not perfectly flat. If you put multiple pieces of tape on each of the other exposed metal areas then the glass won't wobble, but it will be slightly tilted.

2) The major thing I found, was that after a few weeks, the U-joints got gunk'ed up and about half of them just would not turn anymore without using a significant amount of strength. An easy way to tell if this is the case for you, is to position the extruder close to the platform and use the manual controls to move it back and forth along the platform in the X and then Y directions, if the nozzle lifts up or lowers every time you change directions, it is stuck. To fix this, you will have to take it apart and use some kind of cleaner or lubrication until it moves smoothly again.

We're working on putting together some troubleshooting instructions for this exact problem. Edward came up with the software tower rotation solution and I worked on it some, and it did solve this problem for both of us, but it would be nice to get some more data points. Can you take a look at this post and let me know if it works for you? http://forum.seemecnc.com/viewtopic.php ... =20#p23853

Lochemage wrote:I had this problem until I fixed two things, hopefully they will be able to help you.

1) The instructions say to place a piece of kapton tape on the onyx heat bed where the metal is exposed, I found that it adds a very small bump that actually causes the glass to act slightly like a 'teeter totter' and not perfectly flat. If you put multiple pieces of tape on each of the other exposed metal areas then the glass won't wobble, but it will be slightly tilted.

2) The major thing I found, was that after a few weeks, the U-joints got gunk'ed up and about half of them just would not turn anymore without using a significant amount of strength. An easy way to tell if this is the case for you, is to position the extruder close to the platform and use the manual controls to move it back and forth along the platform in the X and then Y directions, if the nozzle lifts up or lowers every time you change directions, it is stuck. To fix this, you will have to take it apart and use some kind of cleaner or lubrication until it moves smoothly again.

Ah hah! Yes! I will try #1 right now.

I have emailed SeeMeCNC a few days ago regarding my delta arm blue, since I have indeed discovered a few gummed up arms. I have over-sanded my u-joints and had to compensate with shimming washers with mixed results. I understand SeeMeCNC has resolved this issue on newer units, and emailed to ask for a price, I have yet to get a response.

626Pilot wrote: We're working on putting together some troubleshooting instructions for this exact problem. Edward came up with the software tower rotation solution and I worked on it some, and it did solve this problem for both of us, but it would be nice to get some more data points. Can you take a look at this post and let me know if it works for you? viewtopic.php?f=54&t=2765&start=20#p23853

I'll try reading into this and get back to you with the results.

Just an update

Added some Kapton to the remaining areas of the Onyx to properly level the bed - I can see this made a slight difference in the bed. But when I tried to print, it was horrible - still way too far from the bed.

I think I need to run through the delta radius calibration steps in the updated manual. But even with that, I suspect I have an issue with tower rotation. @626Pilot, your documentation on that post doesn't really clear up tower rotation calibration, is there any in-depth discussion regarding this?

Still no word back from SeeMeCNC regarding my delta-arm blues. If I don't hear back from them by tomorrow I'm going to send a follow-up email.

I rue the day I can finally have a reliable printer like my Makerbot Replicator at work. Starting to regret my $1000 purchase......

just a sanity check, have you made sure that your glass is actually perfectly flat? I've purchased borosilicate glass before that wasn't actually flat.

Lochemage wrote:just a sanity check, have you made sure that your glass is actually perfectly flat? I've purchased borosilicate glass before that wasn't actually flat.

Yes, I have checked the bed, it's flat enough to not be causing the issue I'm seeing.

I have run through the complete calibration process included in the revised updated GeneB's instruction manual. I was pretty far off on the printer radius ( started at 198.25, ended up at 199.20!) but, I'm still not perfectly level.

I am finding the nozzle raises on the same area as before - directly BETWEEN the Z and the X rails. I think I have an opposite effect between the Z and Y axis in which the nozzle crashes into the bed.

It appears I need to adjust the tower rotation, but I don't see any clear instructions on how to calibrate this.

Can anyone point me in the proper direction? Thanks!

cassetti wrote:

Lochemage wrote:just a sanity check, have you made sure that your glass is actually perfectly flat? I've purchased borosilicate glass before that wasn't actually flat.

Yes, I have checked the bed, it's flat enough to not be causing the issue I'm seeing.

I have run through the complete calibration process included in the revised updated GeneB's instruction manual. I was pretty far off on the printer radius ( started at 198.25, ended up at 199.20!) but, I'm still not perfectly level.

I am finding the nozzle raises on the same area as before - directly BETWEEN the Z and the X rails. I think I have an opposite effect between the Z and Y axis in which the nozzle crashes into the bed.

It appears I need to adjust the tower rotation, but I don't see any clear instructions on how to calibrate this.

Can anyone point me in the proper direction? Thanks!

I'm having almost the exact same problem, my bed is dead level ands flat. in front of all three towers and in the center my nozzle zeros perfectly, however it raises off the bed between x and z. the tower rotation instructions say to change alpha a b & c in the eeprom, but my eeprom doen't have alpha settings

Just a followup.

I purchased replacement arms and carriage u-joints from SeeMeCNC and sanded those perfect (or damn near, if they're still a problem, i'll sand abit more). I think that resolves my Delta-Arm-Blues problems. But that didn't resolve my other issue with the bed.

It took me a bit of work adjusting the endstop calibration, and the Printer_radius, but I finally managed to print this (per 626Pilot's request) http://www.thingiverse.com/thing:50505


Attached is the pictures of my problem

On the left side between the Z and X rails, the platform rises considerably away from the bed, while close to the Y rail, there's a high spot. Most visible is the left side outer circle, you can see how the head was so far away, the filament doesn't even touch the bed!

Suggestions before I go Office-Space on this printer?

Help??

Hoping to get at least SOMETHING printed before christmas for someone, ha ha ha

I've been following your thread, not sure how much help I can offer. This machine is such a delicate balance between mechanical and digital it's hard to pin it down when something is off. My first instinct in your case is to (and I know you don't want to hear this) go back through and triple check all the mechanical side, then software. I'll throw some questions out just for the sake of maybe shaking something loose in the way of solving it.
1. Is everything square and within tolerance measurement wise with the columns?
2. Are they straight and flat? Not twisted or bowed?
3. Are your end stops functioning to spec? Are you getting some repeatability in z heights during calibration scripts?
4. Are your cheapskates dialed in? No lash? Do they run freely? (I had a tight one and it caused all kinds of issues )
5. Are your arms lash free? Are they moving freely? Are they the same length? This is the the first thing I would consider upgrading in your case, if you are still using stock arms? I went to Trick Laser CF when my stock arms started "galling up" really badly-again major issues CF arms were well worth the investment, highly reccomended
6. Are your belts tensioned well and in good shape? not too tight? IMO good tension keeps them from skipping, but not so tight as to make the stepper work hard.
7. Pulley's on the stepper's are snug? Grub screws tight on shaft flats and not allowing slippage?
8. Fastener's in general are all tight?
9. Effector is level? spacers are the same and torqued equally? Hotend is well mounted? (Is it stock?)
10. Borosilicate glass is clipped evenly? Have you found a "sweet spot" and put a witness mark on it so you can clean it, and then clip it back it the same orientation? Before I did this, I would take it off, clean it, hairspray it, then just drop it back on. And many times my first layer was out of calibration? So I did a careful calibration and then marked the edge with a sharpie so I could put it back the same way every time. I don't know how much difference it's made? But it's one less variable

Software-If you do a complete or partial rebuild, I would consider a "reset" to all default on the latest firmware and start from zero. Painful?... yes, but eliminates a lot of past variables in settings. I did this with my arm upgrade. i knew I would be changing some things so I went for it, so far so good

I haven't had to play with radius settings. I've done some large prints, and I do think I have some high/low spots, but call it luck or whatever the prints have "settled in" after several layers.

I wish you all the luck/skill or black magic it takes to get printing

Batteau62 wrote:I've been following your thread, not sure how much help I can offer. This machine is such a delicate balance between mechanical and digital it's hard to pin it down when something is off. My first instinct in your case is to (and I know you don't want to hear this) go back through and triple check all the mechanical side, then software. I'll throw some questions out just for the sake of maybe shaking something loose in the way of solving it.
1. Is everything square and within tolerance measurement wise with the columns?
2. Are they straight and flat? Not twisted or bowed?
3. Are your end stops functioning to spec? Are you getting some repeatability in z heights during calibration scripts?
4. Are your cheapskates dialed in? No lash? Do they run freely? (I had a tight one and it caused all kinds of issues )
5. Are your arms lash free? Are they moving freely? Are they the same length? This is the the first thing I would consider upgrading in your case, if you are still using stock arms? I went to Trick Laser CF when my stock arms started "galling up" really badly-again major issues CF arms were well worth the investment, highly reccomended
6. Are your belts tensioned well and in good shape? not too tight? IMO good tension keeps them from skipping, but not so tight as to make the stepper work hard.
7. Pulley's on the stepper's are snug? Grub screws tight on shaft flats and not allowing slippage?
8. Fastener's in general are all tight?
9. Effector is level? spacers are the same and torqued equally? Hotend is well mounted? (Is it stock?)
10. Borosilicate glass is clipped evenly? Have you found a "sweet spot" and put a witness mark on it so you can clean it, and then clip it back it the same orientation? Before I did this, I would take it off, clean it, hairspray it, then just drop it back on. And many times my first layer was out of calibration? So I did a careful calibration and then marked the edge with a sharpie so I could put it back the same way every time. I don't know how much difference it's made? But it's one less variable

Software-If you do a complete or partial rebuild, I would consider a "reset" to all default on the latest firmware and start from zero. Painful?... yes, but eliminates a lot of past variables in settings. I did this with my arm upgrade. i knew I would be changing some things so I went for it, so far so good

I haven't had to play with radius settings. I've done some large prints, and I do think I have some high/low spots, but call it luck or whatever the prints have "settled in" after several layers.

I wish you all the luck/skill or black magic it takes to get printing

Thanks for the long response (and following the thread this far)

This is all helpful information for anyone with problems to keep in mind. I'm fortunate that this wasn't my first 3D

printer. I've had the pleasure of starting with a Makerbot Replicator. Knowing what to expect has made fixing my printer so

much easier.

1 and 2) No, It is not. I can confirm there is a minor tower rotation (1 or 2 degrees perhaps) problem. I have calibrated using the new yardstick technique to ensure all 3 towers are secured at precisely the same height (found 1 loose t-nut and 1 tower off by 0.5mm approximately). 626Pilot supposedly has information on how to adjust the firmware math to compensate for this issue, much how you can compensate for a concave/convex effect by tweaking the printer_radius.

3) I am not a big fan of the mechanical endstops for Delta printers because the switch has pressure to move the cheapskates down several micron once the motors release after 15-20 seconds. To avoid this, I pre-heat my printer before pressing "Run Script" to ensure G28 is executed followed shortly after by commands to begin printing. However, they are functioning properly, and Z-axis height from bed is repeatable even 4 months after leaving the printer on the shelf (i was impressed how the printer needed no adjustment to the endstops when I first powered up printer before executing the upgrades).

4) Yep, no lash whatsoever, not tight enough to develop flat spots on the delrin wheels when not in use, and yet tight enough to ensure all wheels spin properly and don't drag (including the back wheel).

5) Arms move free, replaced with the v2 arms I recently purchased from SeeMeCNC (sanded the burs slightly to ensure proper fit). Also re-lubricated all joints with fresh white lithium grease

6) Belts are sufficiently tight, no howling or groaning. Enough 'give' for now. I will look into a tensioning upgrade at some point because I don't like the current "jam a screwdriver in and pry upwards" method of tensioning. Though I really haven't seen any official upgrade for belt-tensioning - suggestions?

7) Pulleys are Snug - have never had an issue with one slipping. They are the original all-metal milled pulleys. I might upgrade them at some point in the future, but so far, no slippage whatsoever

8) Yep, went around and gave an extra 1/16 turn where needed last time I rebuilt the extruder (less than 20 hours print time since then)

9) Last time I tightened, I loosened all screws on the onyx, pre-heated to 65C, then tightened screws in a star pattern of equal (as equal as I could) torque. Helped slightly but still hasn't resolved issue with the hotend raising away from the bed in one spot, consistently.

10) Haven't put a witness mark on the onyx, but considering the numerous times I've removed and reinstalled the glass, it's safe to say this isn't the culprit. The hotend assembly was removed from the platform and reinstalled two weeks ago, I can confirm it's snug as a bug. The problem has persisted before my upgrade and after, I don't suspect it has anything to do with the arms or the u-joints.


I can print, I have printed many awesome items if you check my flickr pictures (posted on my build thread). The problem is making full use of the bed. I know there's a way to calibrate it in the firmware, but I haven't attempted yet.

I'm fairly certain my problem is tower rotation, but I need to know how to compensate for it.

Phew, I can't imagine a newbie working through everything like this. I'd be pulling my hair out in frustration!

Yeah, I remember being a "noob", all I could think was-"oh man! what did I get myself into!". I still don't think I have a good handle on the electronics/software side of this. But trying to learn something new everyday

I really think the onyx needs to "float" with no fastener restriction. I have some ideas on how to do it, just no time to implement See this thread for my "mad ravings"- http://forum.seemecnc.com/viewtopic.php ... 173#p21173
This is what I put together for belt tensioning. Probably the first "mod" I did It does help to get even tension on all three belts. And I can make them run on center, sort of
Keep working on it Good luck

I am an absolute noob: Question about tower rotation... do you mean a rotation of one compared to another? like a twist or torsion? This is my problem now.
Or do you mean there isn't even spacing between the towers? or 1 or more towers are rotated so that one face does not aim directly to the center of the platform?

Sorry to hijack!

Thanks!

neurascenic wrote:I am an absolute noob: Question about tower rotation... do you mean a rotation of one compared to another? like a twist or torsion? This is my problem now.
Or do you mean there isn't even spacing between the towers? or 1 or more towers are rotated so that one face does not aim directly to the center of the platform?

Sorry to hijack!

Thanks!

I'm not quite sure what the problem is exactly, but I have reviewed this post which clears up a few things - http://forum.seemecnc.com/viewtopic.php ... =20#p23853

I've already calibrated my endstop screws, and the printer_radius. I can run scripts 1-4 in my Repetier-host to home and G0 Z0 at the 4 points around the bed perfectly. Given 626Pilot's explanation, it sounds like i need to calibrate my tower rotation calculations in the firmware, but I'm completely lost on how to do that. Still waiting to hear back from him on how I go about fixing.