Each layer external shell starts out rough.

wirlybird · 2016-09-22T18:26:09.000Z

Yea, I thought of that but I am giving the barrels away as pencil holders. I am in the oil and gas business so barrels are a fun desk trinket! At least it puts the material to use but you are right I should remember to do that for testing.

MindFuLL · 2016-09-22T18:51:18.000Z

I don’t use S3d but @Perry_1 knows what he’s talking about. Ringing is probably the most frustrating and difficult to diagnose problem I have dealt with in 3d printing(and that is saying something). I just want to add to what he said about trying different speeds. You are fighting vibration, so you want to not just lower the speed but do so in an imperfect interval. For example, if you double or half the speed(an octave) you can end up just stretching or compressing the ripples, while trying something like 89% or even 107% can SOMETIMES eliminate them all together.
I think @perry_1 is dead on about the momentum from the travel speed since it’s happening in only one spot and you had the seam set at that location. Did you have that seam set like that on both of the settings?
-Jesse

Perry_1 · 2016-09-22T19:17:31.000Z

@Mindfull has good info here. People always wonder why I tell them to print at 73mm/s!

There is a strong indication that your resonance can match stepper movements, etc. I had this exact situation on some extruded text on a print. I kept getting slower and slower on the print… First 60mm/s, then 30mm/s, then 15mm/s. All that happened was the ringing changed. It actually almost went away at 73mm/s. I have been printing at 73mm/s a lot after that (xy print speed).

wirlybird · 2016-09-22T20:02:04.000Z

Yes, and it makes some sense because I can see that it does the inner loop then hits the outer and reverses direction so there is momentum from direction change and lateral position change.

Before changing speeds I am going to try a print (just top part) with random and also with most efficient start-stop and see what I can see. Just for fun.

wirlybird · 2016-09-22T20:06:57.000Z

Interesting. I was told recently that a full step on the Z is 40 microns to make layer heights a multiple such as 1.2 instead of 1. Another thing I pick up is this for better resolution prints and save some time is drop to 1.6 layer height and print the infill at 3.2 every other layer. Seemed to work out ok.

wirlybird · 2016-09-22T22:23:08.000Z

Ok, test 1, slowed non print X-Y movement down from 4200 to 1000 and no difference.

Now running at reduced print of base 1000mm/min and 50% outer shell. Almost done.

Watching the preview while printing it is the direction change doing it I am convinced. It goes from the inside ring to the outer with a direction change at the seam and then the outer to the inner with a direction change again at the seam and I can see a similar effect on the inside in the opposite direction from the seam.

wirlybird · 2016-09-22T22:56:08.000Z

Ok, speed at 1000mm/min results are pretty good. Going to ramp up a bit and see if I can find the break point.

Perry_1 · 2016-09-23T00:05:03.000Z

This thread is dealing with vertical ringing along the x/y, so I dont want you to come to the wrong conclusions when you apply that to layer height, which has a slightly different effect.

As far as infill at every other layer, this can be good if you are printing under certain circumstances, and at a small enough layer height. When your height to width ratio gets too high, printing every other layer of infill, you can get infill that does not adhere well, making your part weak, or it can become so wispy it does not hold your top layers up well.

But yeah, printing infill at every other layer, and support at every other layer, when it works, is GREAT! (really increases printer time), and I do it for my personal printing sometimes, but not for my professional printing.

wirlybird · 2016-09-23T00:17:46.000Z

Yea, the layer was a deviation on topic!

So at 1000mm/min it is good, at 1500mm/min it is definitely passable but the ringing is starting.

Slowing the non print XY speed didn’t do anything and I think because it is simply reversing direction and going from inner shell to outer shell (2 shells) so the movement is small but abrupt.

I can try really setting the non print speed to extremely slow and see!

Also I learned about the ratio and I think 1.6 and infill at 3.2 every other is about the limit maybe 1.8 and 3.6.

MindFuLL · 2016-09-23T00:18:27.000Z

What is your x/y jerk setting? Might try lowering to 15 or 10.

MindFuLL · 2016-09-23T00:22:53.000Z

Yeah, the common wisdom seems to be that you should keep your layer height at 80% or less of nozzle diameter. I personally never go above 50%, even on infill, for reasons above.

wirlybird · 2016-09-23T00:50:25.000Z

Ok so whrere do I find that?! I’m on S3D.

MindFuLL · 2016-09-23T01:00:18.000Z

It’s a setting that is stored on the board itself. If your board uses EEPROM(and has it enabled) it’s as simple as sending a few gcode commands. If not you would have to re-flash the firmware. Which firmware does your machine use?

wirlybird · 2016-09-23T01:15:30.000Z

This one is a FFCP 2016 with sailfish 7.8

MindFuLL · 2016-09-23T01:40:47.000Z

I mostly use Marlin and have zero sailfish experience but it looks like the gcode is the same. “M501” will return all the current settings stored on the EEPROM.

FIRST THING you should do is take a screen shot of these setting so you have them as a back up/reference.

You should see the lines:

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 X20.00 Z0.40 E5.00

In the example above, the jerk is 20. To change it to 15 you would send “M205 X15” and then “M500” to save the changes. If you do not send “M500” after entering any setting, it will not be stored. You should then send another “M501” to verify the changes were made and are correct(a decimal in the wrong place can be a disaster). The jerk settings are kind of subtle unless you make extreme changes(don’t do this). If 10 or 15 doesn’t fix anything don’t try going lower. Regardless, you just got access to some stuff you didn’t know was there. Enjoy.

wirlybird · 2016-09-23T11:59:58.000Z

Thanks, I’ll check this out and I need to research this setting since I don’t know what it is for or what it does.

How about acceleration settings? I don’t need this thing lightning fast but smooth and steady with out using the “granny” gear!!

More to learn!!

wirlybird · 2016-09-23T13:29:47.000Z

I was just looking at stuff and reading up and I think I can change this stuff via Replicator G. I did use that to fix a lighting issue and change a couple other minor things on the advice of FFCP support.

Does this seem right? I can post a screen shot this evening of what I am seeing.

Thanks for the help guys!

Perry_1 · 2016-09-23T13:34:12.000Z

You have to be careful comparing certain M commands that work in marlin that do not work in Sailfish, and vice versa.

So with marlin, you can easily recompile the firmware, and there are a few GUI ways to do this.

Sailfish, which I personally like, some of the slicer software allows you to do change parameters.

Sailfish has accelleration built into its firmware, and messing with the firmware settings messes with some complicated formulas that provide excellent prints from Sailfish.

The simple 1D kinematic equation for distance including jerk is:
s = vi*t + (1/2)*ai*(t^2) + (1/6)*j*(t^3)

where s = distance traveled, vi = initial velocity, ai = initial acceleration, j = jerk (i.e. rate of change of acceleration), and t = time

You can change “jerk” settings with sailfish. The key is that this is often a choice of last resort, because it can be difficult to change back to defaults without reflashing the software.

S3d does not have jerk setttings, and I recommend getting dialed in with S3d, but if you really want to know all there is to do, and expand your control, you can mess with firmware settings. Jerk is explained in the sailfish manual at "4.3.2 Maximum Speed Changes " under firmware settings. In effect, it will do the same thing though-slow down your printing.

wirlybird · 2016-09-24T13:12:50.000Z

Thanks. I guess I better leave it alone. It just seems odd I can’t print this small barrel faster than 1000mm/min. The one large enough to hold pens and pencils will take over four hours to print!

I am going to draw one myself to see if it is any better.

It has been a good learning experience.

MindFuLL · 2016-09-24T14:52:17.000Z

Jerk (not the actual physics term, which is the second derivative of velocity) has to do with how fast an axis makes a direction change. It should only affect your overall speed unless you make significant changes but small changes could have a big impact on the type of issues you are dealing with. Don’t mess with it if you are not comfortable.
-Jesse