Slanted prints with the RoStock MAX

[ayp5082]

Hardware: RoStock MAX, EZStruder, RAMBo 1.1b.

Software: Repetier Host 0.90C (built in Slic3r, have also tried standalone Slic3r).

Firmware: Repetier firmware (latest release from SeeMeCnc)

I finished the build for the RoStock MAX some time ago and have been battling this problem for weeks. I am witnessing fairly uniform slanting in my larger prints, as shown below. It is especially prevalent with wider objects, and I do not see nearly as much slanting with objects such as the calibration cube.

I thought it might have been an issue with one of the stepper motors (the one "in line" with the slant) and switched it out, to no avail. I thought this slant might be exaggerated by printing on the extremes of the board. When I printed the above part closer to the edge of the bed, the same slant was witnessed. The funny thing is, when I tried printing 3 smaller parts (the same .stl file copied 3 times) on the edges of the bed, the prints did not exhibit much of a slant. I am not sure if it is having some issue with the larger .stl files, but the g code looks fine for any file when I preview it.

I have completed numerous calibrations to ensure the tip is equidistant from the bed as it prints, measured the rail-to-rail distance to ensure an equilateral triangle is formed by them, switched out a stepper motor that I believed to be faulty, and am still getting this issue. At this point, I believe it to be a firmware issue or an issue with the RAMBo card.

My RoStock shows a lot of promise in print capability, but it has really got me on my last nerve. I am thinking of trying to switch from Repetier to Marlin firmware, but am worried about how well it will work/if it will play nicely with Repetier Host. Does any one have any advice or ideas on the issue and how I can proceed from here? Thank you.

[lordbinky]

Measure the arms and let us know what you find.

[ayp5082]

Binky,

I'm not entirely sure which measurement you want me to take? The distance between the aluminum bars is 14 3/8" for all of them, and ~10.5" on the arms connecting the aluminum bars to the extruder. I apologize for not initially knowing which measurement you were after, but please let me know if it is something different.

Thanks.

[lordbinky]

as seen in the picture in this threadhttp://forum.seemecnc.com/viewtopic.php?t=1959&p=17311 you're trying to measure delta rod length. Not all the arms are always perfectly the same, and if you happen to have the two worst arms on the same tower, it could affect the print the further you get from the center. The other thing to do is mark and measure the points you use to calibrate height as shown in geneb's manual. These are just things to check to eliminate possible problems right now.

[ayp5082]

Binky,

I took all of the rods off and measured them. The biggest disparity between the holes of any two rods was ~1.5 mm, which is fairly large. I also found that 3 of the rods did not have the holes drilled perfectly straight, which potentially skewed some of the measurements. I have put these rods on opposite towers and paired the longest rod with the shortest rod, etc. I went to run a print but am having issues with the firmware (the code will not load). I've already reprogrammed the Atmel with an AVR a few times on the RAMBo and found that I get one upload, then the firmware seems to mess with the bootloader. I am still investigating that issue, but I will print out a part and follow up if the part remains slanted. Thank you.

-Alex

[lordbinky]

To elaborate on the mark and measure your bed leveling calibration points, you want to measure the distance from the center point to each of the other 3 (scripts 1-3). If the distances to center aren't the same it's straightforward that the slant is coming from the printer thinking it's moving the same distance but is actually moving further in reality.

If it was just that issue alone, your prints such as a circle would come out a little oblong, but... there are so many variables in play you just gotta start somewhere and start eliminating what you can.

[kbob]

Binky,

I took all of the rods off and measured them. The biggest disparity between the holes of any two rods was ~1.5 mm, which is fairly large. I also found that 3 of the rods did not have the holes drilled perfectly straight, which potentially skewed some of the measurements. I have put these rods on opposite towers and paired the longest rod with the shortest rod, etc. I went to run a print but am having issues with the firmware (the code will not load). I've already reprogrammed the Atmel with an AVR a few times on the RAMBo and found that I get one upload, then the firmware seems to mess with the bootloader. I am still investigating that issue, but I will print out a part and follow up if the part remains slanted. Thank you.

-Alex

Wow, those are way off.

I wonder if that's part of the reason people are so happy upgrading to TrickLaser arms — maybe TrickLaser has better quality control on the length.

[lordbinky]

AVP contact SeeMeCNC and let them know about the arms, they may want to help resolve it. They have been fantastic from everything I have seen on the forum and my personal dealings with them.

I am really excited to hear some feedback on the Orion, especially the updated and improved arms SeemeCNC developed. I'm hesitant to criticize or go beyond identifying specific issues with the Rostock MAX kits due to lack of knowledge on key manufacturing variables. SeeMeCNC put together a great kit for the price. If we

I have no reason to suspect their decisions on the design isn't based on best bang for your buck. The way they support their customers is so counterproductive and contradictory to a pure profit oriented business model, they'd have a certifiably crazy person in charge if they weren't in it for more than money.

If I didn't know that the people at SeeMeCNC weren't great guys making the best choices they have, I wouldn't hold back on any critique or decision . Instead though I'll twiddle my thumbs and keep in mind there was a reason for their design decisions with lots of manufacturing specifics we aren't privy too ( which just makes me more curious...) that impact the things we are interested in such as price, part tolerances, material selection to other things we're less interested in like scalability, part lead times,waste, etc.

[nathanstenzel]

I am having slanted prints too. Good luck, kbob. Hopefully, we will both get our prints looking straight instead of slanted.

[ayp5082]

Binky,

I took the measurements you suggested and did not find much discrepancy. I also took your advice and contacted SeeMeCNC (I had bothering John already with other stuff, and decided to ask him about this as well). He suggested further isolating the end stop lines from the stepper motor lines and making sure they were nowhere near each other. I rerouted the wiring, and as it turns out the noisy stepper motors were causing false triggers on the end stop lines. This resulted in the motion of that axis "freezing up" for a split second every layer, causing the slant. I've run into noise issues with steppers before but just didn't recognize the symptom. It's easily eliminated by filtering the end stop signal with a low pass filter with a cutoff of about 100 Hz, and I'm kind of surprised the RAMBo doesn't do that. Maybe the next iteration.

Regardless, John helped me to solve my issue and has been invaluable so far. I've also been extremely impressed with their customer service and pleased. Thank you for the suggestions on here as well; it was much appreciated.

[Eaglezsoar]

Binky,

I took the measurements you suggested and did not find much discrepancy. I also took your advice and contacted SeeMeCNC (I had bothering John already with other stuff, and decided to ask him about this as well). He suggested further isolating the end stop lines from the stepper motor lines and making sure they were nowhere near each other. I rerouted the wiring, and as it turns out the noisy stepper motors were causing false triggers on the end stop lines. This resulted in the motion of that axis "freezing up" for a split second every layer, causing the slant. I've run into noise issues with steppers before but just didn't recognize the symptom. It's easily eliminated by filtering the end stop signal with a low pass filter with a cutoff of about 100 Hz, and I'm kind of surprised the RAMBo doesn't do that. Maybe the next iteration.

Regardless, John helped me to solve my issue and has been invaluable so far. I've also been extremely impressed with their customer service and pleased. Thank you for the suggestions on here as well; it was much appreciated.

I wonder if a pullup resistor on the end stop would make any difference? Looking at the schematic for the Rambo, it looks like the end stop inputs directly to the logic chip with no
pull ups or buffering of any sort. Just a thought.

[lordbinky]

LOL I purposely didn't say check the end stop wires thinking I suggest that one too often. Glad you got it sorted out

I was thinking something about filtering the noise as well, I was thinking in the 1 uF ceramic capacitor across the switch might be enough to clear up the noise. I might try that that this weekend.