Consistent Horizonal lines on the prints of a Delta printer

Hello all ,

I have a trouble with my Delta printer.
As you can see in the image below There are some consistent lines on the walls of my prints.
[imgur.com]
I'm moving carriages with wheels and belts on my printer.
I checked my belts , eccentric nuts behind the wheels and tighten them all but the problem still exists.

Thanks in advance for any help or advice.

Most likely reason is wrong PID parameters for heat bed or hot end temperature maintenance. When temperature is fluctuating then also the distance between bed and hot end tip is changing. This change in distance has impact on the actual layer height of the layer currently printed. At the same flow speed that means wider or narrower wall leading to the pattern you see. Check whether the temperature fluctuates and whether its period is consistent with the period on the printed object.

Quote
hercek
Most likely reason is wrong PID parameters for heat bed or hot end temperature maintenance. When temperature is fluctuating then also the distance between bed and hot end tip is changing. This change in distance has impact on the actual layer height of the layer currently printed. At the same flow speed that means wider or narrower wall leading to the pattern you see. Check whether the temperature fluctuates and whether its period is consistent with the period on the printed object.

Thanks hercek for the reply.
Umm, I tuned my PID parameters for the hotend.The hotend temperature fluctuation is just 2 ℃ and as I'm printing With PLA I turned the heated bed off.(It's PLA so I just use a simple glue stick for adhesion).

Can this amount of fluctuation result in that pattern?

Probably not. 2 °C looks too low for it.

Is the length of the period of the pattern the same as the timing belt pitch? Maybe something with belt not matching pulleys well?

If your idler is not a toothed pulley then make sure the belt is turning around idler with its smooth side. Otherwise it would cause small errors with belt pitch period.

Might be also eccentric pulley on extruder. Possibly.

No more ideas.

Quote
hercek
Probably not. 2 °C looks too low for it.

Is the length of the period of the pattern the same as the timing belt pitch? Maybe something with belt not matching pulleys well?

If your idler is not a toothed pulley then make sure the belt is turning around idler with its smooth side. Otherwise it would cause small errors with belt pitch period.

Might be also eccentric pulley on extruder. Possibly.

No more ideas.

Sorry for the delay.
I will check that out.
About the idler pulley I'm Using DC42 way of assembly which a M3 screw passes through flange bearings.When I home the printer and the carriages get close to top vertices I hear a scratch like sound from one of my idler pulleys and I think that's the problem.

I've read in some threads that this way of idler pulley assembly puts a lot of force on the bearing inside the pulley.Is that right?And if it is,What would be the alternatives?

Thanks for your help

Edited 1 time(s). Last edit at 09/03/2020 03:01PM by Mj996.

I do not know DC42 idler pulley assembly.

If the scratch happens only during homing and cause some error only when a carriage is near the home position then it cannot cause the pattern on your print. You must look for a periodic error along the part height. Not a one time error near homing. The homing error may be there but it is a separate issue.

Quote
hercek
I do not know DC42 idler pulley assembly.

If the scratch happens only during homing and cause some error only when a carriage is near the home position then it cannot cause the pattern on your print. You must look for a periodic error along the part height. Not a one time error near homing. The homing error may be there but it is a separate issue.

This is the idler pulley assembly.
[ibb.co]

Ok the first thing I would check when I get to the printer is looking at belts and pulleys to see if they match.
How can the eccentric pulley on the extruder cause this issue?(I'm using MK8 bowden setup extruder)

What is the pitch between the ripples? Is it consistent?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

More extruded material means thicker walls. Less extruded material means thinner walls. An eccentric pulley on extruder will cause variation in the extruded amount of plastic with the period corresponding to one pulley rotation. If the cube walls are thin enough and filament thick enough then one extruder pulley rotation will correspond to more layers and it may correspond to the pattern you see. You can run the numbers to check whether this is possible.

Quote
dc42
What is the pitch between the ripples? Is it consistent?

Thanks for your reply DC42.
a kind of.We can say almost 1mm for pitches.
Ok I checked belts and pulleys and they match perfectly.I noticed that one of my carriages has a bit rotation.I can see my mag balls on that carriage rotating along the tower.I took a look at wheels assembly I turned the eccentric nuts so that the screw position is close to the tower and also I tighten nylock nuts but it was still a small rotation in carriage.Anyway I printed the cube again and now it's so much better than the last print. My carriages are 3d printed and they are thin so maybe this is the reason why I still have a bit rotation.

Quote
hercek
More extruded material means thicker walls. Less extruded material means thinner walls. An eccentric pulley on extruder will cause variation in the extruded amount of plastic with the period corresponding to one pulley rotation. If the cube walls are thin enough and filament thick enough then one extruder pulley rotation will correspond to more layers and it may correspond to the pattern you see. You can run the numbers to check whether this is possible.

Nice explanation.Thanks

One of my carriages had a bit rotation.I don't know if it makes any sense but when I re-tighten my nylock nuts and turned eccentric nuts behind the wheels ,It does improve the print very well.
and the strange point is that when I replaced my cooling fan with a high power turbo fan those lines completely disappeared !!
I also printed a benchy to see the overall quality of the print and here is the result :
[ibb.co]

I would be very happy to hear your opinion.

The chimney looks terrible and I think that's because I have problem with printing small objects.When it comes to print small objects my printer prints a few layers and after that the nozzle clogs and sometimes it finishes the print with terrible quality as you can see in chimney of benchy.

I've read in some threads that it can be a retraction issue or something related to layer time.
I want to try retraction test print but nozzle clogs in the middle of the print.I even lowered the retraction distance from 4mm down to 2 mm but it didn't make any difference and the nozzle still clogs.I'm using E3d V6 hotend with mk8 extruder (bowden setup) with 650mm bowden tube.
Here are all my slicer settings :

[ibb.co]
[ibb.co]
[ibb.co]
[ibb.co]

The trick to printing small objects in PLA is to use lots of fan cooling and either print more than one at a time or reduce the print speed. Most slicers have a Minimum Layer Time parameter to slow down printing of small objects. Try increasing it.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
The trick to printing small objects in PLA is to use lots of fan cooling and either print more than one at a time or reduce the print speed. Most slicers have a Minimum Layer Time parameter to slow down printing of small objects. Try increasing it.

Thanks DC42.That helped a lot.
So i tuned my retraction setting and now I almost have no stringing :
[img.techpowerup.org]

As you can see That's not a good print yet.

I do not have an explanation how fixing a play leading to the carriage rotation can fix the problem with periodically changing wall thickness you presented in the first picture. Carriage rotation play leads to significant errors in the XY-direction. But the problem is that this play should lead to an error which has period of one layer. While the error in your picture has obviously period spread over more layers. And the error is very consistent.

If you would have a problem with retraction distance and timing relative to the layer time then again I would expect the period of the error to be one layer and not spread over multiple layers. You can verify whether it is connected to retraction/layer time ratio just by printing a bigger cube. That will change the layer time if you are not already at minimum layer time limit.

No experience with E3d hotends and their clogging. Anyway extruder periodic partial clogging/unclogging can lead to the error you have. My only doubt with this theory is again the nice periodicity of your error. My dirt cheap old style reprap hotend clogs only when there is some garbage in it (e.g. a piece of metal from a broken quick coupling).

May be you can run the numbers on the expansion of your hotend when its temperature changes by 2°C. I guessed it is too little to be significant but it was only a guess. Or ignore the computation run PID calibration to get the fluctuation below 0.5°C and try a test print again. Maybe your hotend temperature sensor has poor heat transfer connection with hotend? Is the fan rotating at the same speed while the test cube was printing? If fan speed changes with the same period as the error pattern together with the badly connected temperature sensor then we can get the problem in the test print.

Quote
hercek
I do not have an explanation how fixing a play leading to the carriage rotation can fix the problem with periodically changing wall thickness you presented in the first picture. Carriage rotation play leads to significant errors in the XY-direction. But the problem is that this play should lead to an error which has period of one layer. While the error in your picture has obviously period spread over more layers. And the error is very consistent.

If you would have a problem with retraction distance and timing relative to the layer time then again I would expect the period of the error to be one layer and not spread over multiple layers. You can verify whether it is connected to retraction/layer time ratio just by printing a bigger cube. That will change the layer time if you are not already at minimum layer time limit.

No experience with E3d hotends and their clogging. Anyway extruder periodic partial clogging/unclogging can lead to the error you have. My only doubt with this theory is again the nice periodicity of your error. My dirt cheap old style reprap hotend clogs only when there is some garbage in it (e.g. a piece of metal from a broken quick coupling).

May be you can run the numbers on the expansion of your hotend when its temperature changes by 2°C. I guessed it is too little to be significant but it was only a guess. Or ignore the computation run PID calibration to get the fluctuation below 0.5°C and try a test print again. Maybe your hotend temperature sensor has poor heat transfer connection with hotend? Is the fan rotating at the same speed while the test cube was printing? If fan speed changes with the same period as the error pattern together with the badly connected temperature sensor then we can get the problem in the test print.

I know what you are talking about that doesn't make any sense to me either.
About layer time and retraction,I wasn't talking about that pattern.Actually my printer prints small object specially column , pillars ,etc with terrible quality(like the chimney of my printed benchy in the images above) and Some people recommended increasing layer time and tuning retraction to solve this.Those helped a lot but It isn't still satisfying.


Nozzle clogging problem was solved by replacing the old nozzle with a new one.

Edited 1 time(s). Last edit at 09/06/2020 03:53PM by Mj996.

Quote
Mj996

Quote
hercek
I do not have an explanation how fixing a play leading to the carriage rotation can fix the problem with periodically changing wall thickness you presented in the first picture. Carriage rotation play leads to significant errors in the XY-direction. But the problem is that this play should lead to an error which has period of one layer. While the error in your picture has obviously period spread over more layers. And the error is very consistent.

If you would have a problem with retraction distance and timing relative to the layer time then again I would expect the period of the error to be one layer and not spread over multiple layers. You can verify whether it is connected to retraction/layer time ratio just by printing a bigger cube. That will change the layer time if you are not already at minimum layer time limit.

No experience with E3d hotends and their clogging. Anyway extruder periodic partial clogging/unclogging can lead to the error you have. My only doubt with this theory is again the nice periodicity of your error. My dirt cheap old style reprap hotend clogs only when there is some garbage in it (e.g. a piece of metal from a broken quick coupling).

May be you can run the numbers on the expansion of your hotend when its temperature changes by 2°C. I guessed it is too little to be significant but it was only a guess. Or ignore the computation run PID calibration to get the fluctuation below 0.5°C and try a test print again. Maybe your hotend temperature sensor has poor heat transfer connection with hotend? Is the fan rotating at the same speed while the test cube was printing? If fan speed changes with the same period as the error pattern together with the badly connected temperature sensor then we can get the problem in the test print.

I know what you are talking about that doesn't make any sense to me either.
About layer time and retraction,I wasn't talking about that pattern.Actually my printer prints small object specially column , pillars ,etc with terrible quality(like the chimney of my printed benchy in the images above) and Some people recommended increasing layer time and tuning retraction to solve this.Those helped a lot but It isn't still satisfying.


Nozzle clogging problem was solved by replacing the old nozzle with a new one.

Look at it this way, you may not be over/under extruding in a pattern, you may be mis-registering your x and y position. If those are not consistent, you will see banding. This is a delta printer, so all three towers must agree on where to go, if one of them is a bit off, you will get a bad scale, if one is intermittently off, you might get banding.If one of your tower carriages is "wobbly", that could be your issue. In my experience, banding like this is a mechanical issue.

Your smoke stack ickiness comes from overheating the PLA and it blobbing. Increase your fan/blower speed or align it better so it hits the model in a better place. If you cannot increase your part fan speed, or improve its aim, then decrease you filament temperature by 5 degrees each time until you get what you want or you reach your minimum temperature. If none of THOSE things work, use a breeze wall, that will force a minimum delay time on the small detail areas.

DLC

---

Kits: Folgertech Kossel 2020 upgraded E3Dv6, Anet A8 upgraded E3Dv6, Tevo Tarantula enhanced parts and dual-head, TronXY X5SA Pro(E3DHemera).
Scratch: Large bed Cartesian, exchangeable heads, Linear slide Delta, Maker-Beam XL Micro Delta, 220x220CoreXY.