perimeter bumps every start/stop spot

[airscapes]

Been trying to calibrate my M2 with this role of PLA that I bought from MG but use it up trying to figure out how to get it right
Thin walled calibration squares come out the proper height and thickness within .02. When I print a 100% solid square the perimeter were infill meets have blob that are too high. Layer nozzle is .35 layer width is .4 height is .2. This is worst on 2 of the 4 sides, I assume this is the start and stop of the travel. I have slowly one at a time adjusted the retraction distance up to the current 1.80, Extra Start distance -2.0, and retraction speed (set the other day for stringing) 4800. I also changed infill outline overlap from 15% down to 10% with little change.. What am i missing???
My square has a raise rim around the perimeter and is not flat.. Which setting should I be using to correct this?

[MasterFX]

It is not the start and stop of a travel. The Infill is a continuous line with a little bit overlap to the perimeter (you may try to reduce the overlap).
But it might be a little bit over extrusion too (check extrusion multiplier and filament diameter, and e-steps/mm in your printer).
Especially when you print a 100% solid model the perfect extrusion multiplier is extremely important since if there is just a little bit too much filament extruded it might cause a bump to the top (it squishes besides the nozzle and gets higher then the layer should and this accumulates over each layer).

[airscapes]

Did that. Nothing really fixes this.. it is most pronounce on 2 sides of the square and we are printing at 2 angles, so must be when the Z axis changes and drop to do the next layer.
One side is always highest which I assume is the last layer? I have tried with 0 overlap, under extrued, out side in and inside out.
Retraction up to 1.8 extra restart -.2 with no appreciable improvement. Someone on MG forum who I assume does not use S3d suggested setting the slicer to print infill first then perimeter but I have not found that option in S3D. Any other ideas.. I have a dozen calibration squares all the same... Infill when I measure the center of the square it is 2mm edges 2.1x-2.2x other 2 sides 2.05+
Yes I know squirting molten plastic at speed is not a perfect medium but was hoping for a flat top.

M2 RE 4 stock 3.5 nozzle, layer height has been tried at .1 and .2 no improvement. Bed 60C Ext first layer 215C 210C after that, 100% fan after first layer. MG PLA, bed level, staring height and extuder multiplier calibrated with thin-walled cube
What's my next move??

[zemlin]

What is your Solid Infill Underspeed setting? I drop solid infill to 50-75% depending on the material to improve surface finish.
i have also found that extruder accel and jerk settings can impact the u-turns on solid infill.

try this in your start script and see if it makes a difference.
M201 E10000 ; acceleration
M205 E60 ; jerk

[airscapes]

What is your Solid Infill Underspeed setting? I drop solid infill to 50-75% depending on the material to improve surface finish.
i have also found that extruder accel and jerk settings can impact the u-turns on solid infill.

try this in your start script and see if it makes a difference.
M201 E10000 ; acceleration
M205 E60 ; jerk

Solid Infill is default at 80%. So that sounds like the first one to start altering ..
Although I may have other speed related issues as I have not tried to alter any of them from the M2 default profile. (see attached)
For the acceleration and Jerk, are these not yet implemented in S3Ds interface requiring the use of the start script?

Speed setting screen shot

[MasterFX]

I would reduce jerk even more. Try more conservative with Acc 500, Jerk 20 (this is default on CR-10).
Also you might check your belt tension, especially if you also have problems with ghosting/ringing on sharp edges/details.

For the acceleration and Jerk, are these not yet implemented in S3Ds interface requiring the use of the start script?

Yes

PS: you should change the speed units from mm/min to mm/sec, it is more common and more understandable in my opinion

[zemlin]

I would reduce jerk even more. Try more conservative with Acc 500, Jerk 20 (this is default on CR-10).

note that i am talking about accel and jerk for the EXTRUDER. Your numbers are in line with conservative numbers for the X and Y axes.

[airscapes]

WOOHOO!
Dropped solid infill speed to 50% and it really made a huge difference.
Now on the Acc and jerk setting.. not sure about them.
try this in your start script and see if it makes a difference.
M201 E10000 ; acceleration
M205 E60 ; jerk

Here is what is set now... I think sent M503 to display settings

M201 X900 Y1000 Z30 E2000 -----------------------so this looks like a change to 1000 would be a large change (cut in half) and I will give it a go
Acceleration: S=acceleration, T=retract acceleration
M204 S2000.00 T3000.00
Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum XY jerk (mm/s), Z=maximum Z jerk (mm/s), E=maximum E jerk (mm/s)
M205 S0.00 T0.00 B20000 X4.00 Z0.40 E1.00 ------------- However, E1.00 changing it to 60 seems like that is a bad idea.. is this the correct parameter?

Will go try the M201 E10000 ; acceleration and report back in this same thread if no one post before I get back

[zemlin]

However, E1.00 changing it to 60 seems like that is a bad idea.. is this the correct parameter?

Will go try the M201 E10000 ; acceleration and report back in this same thread if no one post before I get back

That's what I run. It's going to vary with different printers, but I got pretty similar results with extruder numbers when I was running very different machines. It's a low-risk move. Factory extruder jerk is 1.0 on my machine too, but I ran test prints and got the best print quality at 60. The right answer is for you to run your own systematic study and find what works best on your printer.

[airscapes]

factory extruder jerk is 1.0 on my machine too, but I ran test prints and got the best print quality at 60. The right answer is for you to run your own systematic study and find what works best on your printer.

Ok great! Can you tell me what exactly the jerk setting is doing? I am a Unix Admin and programmer so will be getting into the Gcode in due time but need to know what things do. Yes, one change, print, measure, examine, repeat!

Update: The above settings have corrected the infill to perimeter bumps! Sill would like to know what Jerk does..
Going to do a search for the next 2 defects and if need be start another thread for them..
Thanks again!
Doug