26 Reply by jaques (edited by jaques 2014-11-04 07:06:11)

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

I see distortion at same 5 and 11 oclock as me (your images are rotated, could tell by the bed screws).
Sorry man, but I think you're deluding yourself tongue Doesn't look any different when comparing in an image editor.

So doesn't seem to be an issue with the motherboard hardware. Which is good, I guess.
Now we have to think of something else.
I havent yet tried reflashing because I'm away from home, but I guess you've already done that since your Rumba didnt come with Marlin preinstalled, right?
I'll still try it myself.

What are we left with now?

Mechanical problems

1) Problem with the motors (unlikely as 1. if underpowered for what it is intended for, unless everyone else owning a SD4 hasnt noticed this, 2. if 4 of us have faulty motor(s), odds of that are also extremely low)

2) problem with the rods (still bending?) or problem with the bushing setup

3) problem with pulleys and belts and all of us fail to calibrate them properly (unlikely)

Software problems

4) Problem with this version of Repetier Host and Slic3r? (unlikely, jagowilson tried with Simplify3D as well)

5) Firmware (Ireinstalled the firmware from the official Marlin webpage, no difference)

6) Driver? (at this point, who knows?)

Electronics problems

7) PSU (at this point unlikely, jagowilson uses a very poweful PSU, Ive tried running with a car battery)

8) wires?

9) motherboard (unlikely, jaqowilson tried with Rumba board)

Are you also sure you're lights are not really dimming and fans are not slowing down? I remember you said you had vision problems. And fan sound change might be hard to notice when you're concentrating on everything else.

27 Reply by jagowilson (edited by jagowilson 2014-11-03 06:22:50)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

This is most unpleasant lol. However I have not noticed bulging in taller cylinders which is ... Odd.  My next step is to replace the Y motor, followed by replacing the bushings with bearings and also replacing the stock pillow block. Following that, I will replace the Y rod as it is possible it is bent. If I am still not making circles, I will replace the entire carriage system. If at that point it is still not working, I will have to replace the entire body. Maybe it's not square. I have a plan, but I don't have the time and money to do it all right away.

I am absolutely certain my lights are not dimming. Trust me, I've got more power than I know what to do with. To get my e3d to work I actually had to tweak the firmware just to stop it from drastically overshooting.

Sorry I led you down a rabbit hole. Please post a picture of your circles superimposed on mine. I find it hard to believe my denting is as extreme as yours.

28 Reply by jaques

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

jagowilson wrote:

My next step is to replace the Y motor, followed by replacing the bushings with bearings and also replacing the stock pillow block. Following that, I will replace the Y rod as it is possible it is bent. If I am still not making circles, I will replace the entire carriage system.

just buy a makerbot, lol.
But seriously, don't be so hasty, unless those aren't cheap to get from where you are.
I've checked my rod. It isn't bent. I mean it might be bending (not permanently) while printing.

am absolutely certain my lights are not dimming. Trust me, I've got more power than I know what to do with.

Well I powered mine with a car battery while in the car and still had that issue.

To get my e3d to work I actually had to tweak the firmware just to stop it from drastically overshooting.

What did you change?

Sorry I led you down a rabbit hole.

its okay

29 Reply by jagowilson (edited by jagowilson 2014-11-03 06:21:42)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

A NEMA 17 is $25, I'm not sweating that. I had to adjust PID_MAX which is the PWM frequency sent to heaters. And by replace, I meant upgrade the carriages.

Also please post a picture of your circles superimposed on mine. Thanks.

30 Reply by jaques

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

jagowilson wrote:

Also please post a picture of your circles superimposed on mine.

That won't work perfectly. Better do a side by side comparison. Camera lenses and shooting at slightly different angle will look different. That doesn't prevent me from doing a comparison as I've worked with 3d for few years now.

31 Reply by jagowilson (edited by jagowilson 2014-11-03 06:34:24)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

True. Well rest assured I am frustrated enough to continue to pursue this problem. I hate it. So much. Here's hoping it's the motor. I'll find a 3rd party vendor and buy one tomorrow.

The money I'll end up spending to pursue this would've been better spent on a different printer, but I'd rather learn than take the easy way out. I will solve it. This is freaking ridiculous.

I want to know what the culprit is so I can at least call SD out on it. It's stuff like this that will make them tank. Thanks SAM for making my life so difficult.

32 Reply by jaques (edited by jaques 2014-11-03 06:35:45)

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

Why not try a different software and slicer as I suggested first?
Might be a trivial bug in this version of either of the programs or their presets.

33 Reply by jagowilson

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

Already did that with the old board so I know that's not it.

34 Reply by jaques

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

Did you also change the slicer program? Sorry, just double checking.

35 Reply by jagowilson (edited by jagowilson 2014-11-03 06:39:00)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

I also still have Y backlash and it's driving me nuts. I can't print circles. I can't print squares. UGH. Nothing is making any of this better.

And yes I've tried slic3r+RH and Simplify3D.

36 Reply by jaques (edited by jaques 2014-11-03 07:00:23)

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

Any chance Simplify 3d use code from Slic3r?
You could try skeinforge

37 Reply by jagowilson

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

No simplify3d is its own slicer.

38 Reply by nbourg8

  • nbourg8
  • Member
  • Offline
  • Registered: 2014-11-02
  • Posts: 85

Re: Odd circle test results

I've put in a request to Solidoodle to aid in the progress and pointed them to this thread.  Maybe we'll get a better answer.

It does appear to be a result of friction and belt tension overpowering the y-axis stepper motor during the micro adjustments of creating a circle.  When the motor slows down, it can't make the large steps that the rest of the circle makes.  While the x-axis can make these adjustments easily, it is only overcoming the tension of 1 belt and no other belt needs to be the same tension as it.

I'm like you two, very interested in solving the problem rather than going around it.  By now, I have the knowledge to make a printer myself.

39 Reply by jaques (edited by jaques 2014-11-03 19:33:46)

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

Hey, if you haven't since this, here's a video I've made whichs shows me printing this test pattern + a close up of the y motor.
You can see that the "dents" at 5 oclock and 11 oclock happen when the y motor is changing direction.

40 Reply by nbourg8

  • nbourg8
  • Member
  • Offline
  • Registered: 2014-11-02
  • Posts: 85

Re: Odd circle test results

Have we tested the angles between the support rods?  If they are not actually parallel, we could see a reason for motor slippage of the y-axis.

41 Reply by nbourg8 (edited by nbourg8 2014-11-03 22:05:23)

  • nbourg8
  • Member
  • Offline
  • Registered: 2014-11-02
  • Posts: 85

Re: Odd circle test results

For everyone's review, this is the response I got from solidoodle technical support:

Hi Nbourg8:

Imperfect circles are usually a symptom of poorly calibrated belts. Another factor can be stepper voltage, which from your message I can see you have already considered. However, when a stepper chip needs voltage adjustment, the issue is usually accompanied by a loud noise, but not always. This is why I would suggest checking belt tension/calibration before testing with multimeter, as you did not mention any loud noises. Another suggestion, would be to make sure that all of your Repetier Host and Slicer settings are set to default. Below are links on how to do so. My last suggestion before stepper voltage adjustments that can cause this issue, would be axis shifting. This is more common with Y axis, but can also occur on X axis. Once you have eliminated these causes, I would then begin to test the voltage on your stepper chips. Because voltage output can vary in different environments, it is not uncommon to make small adjustments to your stepper voltages, however, I would try these other troubleshooting tips first.

Belt Tensioning/Calibration

Please visit Belt Tensioning for more information.

Y/X Axis Shifting

Axis shifting tends to occur most during a print. The most common shifting is in the Y axis. The X axis can also exhibit this symptom, but occurs less often. Please refer to the following list to look into the most common reasons for axis shifting:

1) Belt Tension. This is the most common reason for shifting. You will want to take a look at this tutorial and follow the instructions. Belt Tension Tutorial Remember, the main idea behind this calibration is to get both Y belts with equal tension. The belts should have some bend to them so be sure not to over tighten.

2) Stepper motor voltage. You can check the voltage for the stepper chip using a multi-meter and this tutorial. Multi-Meter Testing

3) Check to make sure your filament is spinning freely. If it tangles it can cause skipping since the extruder pulls harder on the spool.

Some additional things to check that could cause skipping:

1) Make sure you do not have any of the zip ties that were attached when we shipped the printer. They can eventually move to a point that would interfere with printing.

2) Rods may need more grease. This will be more relevant to older printers but can rarely apply to newer printers.

3) Turn the motors off (or simply remove the power to the printer) and move the axis with your hands the full range of motion. The belt or carriage can get stuck and this helps loosen it.

4) Inspect the belts to see if there is any sign of wear and tear to the teeth.

5) Motor or motherboard overheating. You can try pointing a small fan at the motherboard and print again.

6) In general, make sure the printer is not in a location can be moved suddenly by accidentally bumping into it or the surface it is on.

7) The small belt for the Y motor (located in the back of the printer) can be too tight. This is rare since it does not have a big range of tightness. This can only be adjusted by loosening the 4 hex screws that hold the motor in place and then using the motor itself to tighten or loosen the belt by repositioning it on the frame.

After reading this, when I get home from work, I'm going to:

  • adjust the y-axis motor belt in the back to be slightly less tension

  • using a hex screw, allow the rod collars to spin freely while I tension up the y-axis rods.  This should remove any differential in upper and lower belt tensions

  • Verify motor voltages are nominal

  • Verify y-axis rods are parallel and perpendicular to the rear rod

  • Re-print all calibration stl files and prey for better results.

42 Reply by nbourg8

  • nbourg8
  • Member
  • Offline
  • Registered: 2014-11-02
  • Posts: 85

Re: Odd circle test results

Here's the results of my previous post after calibration of x/y-axis steps/mm and lubrication of each rail with gun oil instead of grease.

If it's better, it's only minuscule.  I still can't produce a "perfect circle".

Post's attachments

IMG_0049.JPG
IMG_0049.JPG 2.46 mb, file has never been downloaded. 

IMG_0050.JPG 2.72 mb, file has never been downloaded. 

IMG_0051.JPG
IMG_0051.JPG 1.61 mb, file has never been downloaded. 

IMG_0052.JPG 1.77 mb, 1 downloads since 2014-11-04 

You don't have the permssions to download the attachments of this post.

43 Reply by jaques (edited by jaques 2014-11-04 04:32:14)

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

Sorry dude, but it isn't better, its the same. Ive done this for weeks, I know.

The email response you got is a copy-paste. We've all gotten the exact same response.
Every suggestions except 1 and 2 is useless and we had all done that before contacting for help.
What's the true problem we don't know, but we're going to get to the bottom of this. Could be a cheap chinese Y motor that was used instead to compensate for the price drop.

What we know for a fact is these SD4 we got were *not* tested prior shipping as claimed. They were not even assembled properly with loose screws and not properly tightened belts which shipping couldn't possibly affect.
They didn't even bother to update the stickers so they wouldn't lead to dead links.

Solidoodle is aware of this thread and they don't even bother to respond, we have SD employees in this forum.
Maybe they are waiting for us to do their work and find out what the problem is with their Solidoodle 4 product.
If they are reading this I hope they are aware of consumer protection and what a hassle it will be for them if one of us decides to send a formal complaint letter.

44 Reply by jagowilson (edited by jagowilson 2014-11-04 04:43:36)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

So wardjr and I have been investigating this and I'm going to try an experiment. Both of you, look at the right side of your Y carriage. See how the Teflon only makes contact at 2 points? How many degrees of freedom does that allow ???

Tomorrow I'm going to design and print a block to replace the bottom piece with one that will make contact at 3 points. I will let you know how it goes. If anyone wants to beat me to the punch feel free. I won't get around to it until the evening.

Wardjr saw my frustration and has stopped me from quitting once again big_smile

45 Reply by wardjr

  • wardjr
  • wardjr
  • Moderator
  • Offline
  • From: Minnesota
  • Registered: 2013-06-10
  • Posts: 5,623

Re: Odd circle test results

big_smile

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

wardjr's Website

46 Reply by jagowilson

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

Relevant images.

Post's attachments

image.jpg 1.2 mb, file has never been downloaded. 

image.jpg
image.jpg 941.17 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

47 Reply by nbourg8

  • nbourg8
  • Member
  • Offline
  • Registered: 2014-11-02
  • Posts: 85

Re: Odd circle test results

Not sure what is meant by Teflon making two points of contact.  I've uploaded photos to this post for your review.

LOL @ Jagowilson
Did you take the rear fan off because it was too loud?

Post's attachments

IMG_0055.JPG
IMG_0055.JPG 1.24 mb, file has never been downloaded. 

IMG_0056.JPG
IMG_0056.JPG 1.31 mb, file has never been downloaded. 

IMG_0057.JPG
IMG_0057.JPG 1.59 mb, file has never been downloaded. 

IMG_0058.JPG
IMG_0058.JPG 1.49 mb, file has never been downloaded. 

IMG_0059.JPG
IMG_0059.JPG 1.21 mb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

48 Reply by jagowilson

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

I broke it. Wired it backwards into the Mobo and it started smoking. I had a quiet one in there and it sucked that I ruined it. $12 gone -_-

49 Reply by jaques

  • jaques
  • Banned
  • Offline
  • Registered: 2014-10-16
  • Posts: 177

Re: Odd circle test results

what are we talking about?

50 Reply by jagowilson (edited by jagowilson 2014-11-04 06:38:21)

  • jagowilson
  • PRINTiT
  • Offline
  • Registered: 2014-09-14
  • Posts: 1,701

Re: Odd circle test results

I bet the right side of your carriage moves up and down. Try it. Do you see why that is obviously a problem? When the motors reverse direction, there are all kinds of things happening that will induce backlash. Most crucially, the left side is the only side being moved correctly because it's using a bushing. The right side, however, does not contact as well because with only 2 points of contact, it has a lot of wiggle room. So the right side is just being dragged along rather than synchronized with the left side.

Print a circle. I bet you will FEEL the right side move up and down/lag behind the left. I can. It makes an audible, but quiet, thump.

Wtf were they thinking? Why didn't any of us who actually own an SD4 think of it? This is wardjr's discovery bassed on pictures of my printer.