Ghosting Defect

[pyronaught]

Here's an interesting problem I found which I think is related to vibration during printing. In the image below, there are three cutouts in the case wall for the power jack, knob and switch. To the right of each of these objects you can see a trail of fading surface defects that match the profile of the cutouts. The surface is parallel to the X axis (extruder travel) and the extruder is moving left to right when it happens. I think what happens is the extruder speeds up to get over the non-print area, then when it suddenly slows down to begin printing again there is an inertial shock that cause vibration. The evidence for this is how the frequency and severity of the ghosted images shortens over time to match the dampening of the vibration.

The gantry seems pretty rock solid so it is hard to imagine any flex going on there, so I'm wondering if maybe the entire machine is shaking since it sits on rubber feet. Maybe it's time to nail this thing down. Anyone else ever run into this?

ghosting.jpg (438.25 KiB) Viewed 13916 times

[Tim]

It's a well-known issue and I've suggested that it might be caused by the sharp transistions in velocity in the Marlin code. The sharp corners on the trapezoidal velocity profile implies sudden instantaneous changes in acceleration. That will inevitably cause some ringing which can never be completely eliminated. There are other firmware variations (which are probably not directly compatible with the RAMBo) that do smooth velocity transitions, which might reduce the effect significantly. Otherwise, your best bet is to slow down the print speeds. You should be able to get by with printing the outermost perimeter at a much slower speed, and leave the inner perimeters and infill at a much faster rate.

[pyronaught]

I'll have to watch what it does when it crosses over those holes next time. I'm guessing it zips across at a faster speed than the print speed, then slows back down to start printing again. In other CNC programs there is a setting for "rapids" which controls the max speed used when moving between work points, but Simplify 3D doesn't seem to have an equivalent setting.

Slowing down the outer perimeter would probably work as a last resort, but this part takes six hours to print as it is and I'd hate to slow it down any further. It would seem a shame to slow down the entire perimeter for the entire layer stack just to fix something that is only happening in a few small areas. I could make a stack of processes with different settings for just the layers with holes, but that would get convoluted too.

[Tim]

(1) This will happen so fast you are not likely to see anything without the use of a high-speed camera. Note that the velocity profiles computed by the firmware will have jumps in acceleration both when printing and when moving across spaces, and the same applies to acceleration as well as deceleration, so you can get ripples in either direction, not necessarily related to the direction the head is traveling.

(2) Note that the outer perimeter is just a small part of the whole object (surface area vs. volume), so you can slow down the outer perimeter quite a bit without increasing the total time by much, although everything depends on the object, so I can't really assign numbers to those statements.

[jsc]

Slow down the allowed maximum accelerations by using M201 in your startup gcode. There is a long thread in the old forum where someone did an exhaustive test to see the effects, and determined that around 800 is a point of diminishing returns. It doesn't slow down your print all that much.

M201 X1000 Y1000 is what I'm using now.

[Dale Reed]

pyro-sub-0,

The speed for rapids (non-extruding moves) is set in the S3D FFF Settings dialog, Advanced veiw, Other tab, in the "Speeds" group on the left. "X/Y Axis Movement Speed" (def = 12,000 mm/min) is the setting. Just to save you some hunting around through all those settings! Wish they would add the acceleration settings here....

Dale

[pyronaught]

I've been using M201 X900 Y900 Z200 E10000 for the acceleration, which is what the print in the picture was done with. I'll try lowering the X/Y axis movement speed next (never would have guessed that is the "rapids" setting), currently it is set to 18000 mm/min. This is where I think the fix is, but if that doesn't work I'll just lower the outline underspeed as the last resort, which is currently 60%.

[jimc]

I changed mine from 18000 to 9000 and outline i vary depending on what im doing. That will got from 60 all the way down to 20 on occasion

[ednisley]

it zips across at a faster speed than the print speed

No zipping involved: the nozzle traces a continuous line from the inside of the object, around the edge, then along the outside of the object (or the other way around). Feed the G-Code into gcode.ws (or whatever toolpath previewer you prefer) and look at the layers to see how the nozzle moves.

If the object has only one opening on a layer, you'll see that there's a single thread going around the entire object, with no rapid (non-extruding) motion at all. You'll still see ripples on the down-stream sides of that opening, though.

Those particular ripples are due to the platform vibrating very slightly as it stops moving after the nozzle deposits the edge of the hole (parallel to the Y axis) and begins moving along the X axis. The M2's platform rests on three springs compressed by the alignment screws, so it has a tendency to shake very slightly after high-acceleration stops.

I added silicone plugs inside the springs to damp that motion:

http://softsolder.com/2013/05/31/makerg ... ilization/

That worked reasonably well, but there's still plenty of mass lashing around.

As other folks suggest, you can reduce the Y axis acceleration by a factor of two or four, which reduces both the force and the ripples. You can also reduce the speed on perimeter threads to get much the same effect: less speed requires less force to speed up and slow down.

[glx51mm]

I have printed dozens of cubes while changing the speed, acceleration and jerk settings on the fly. Results were mixed. I have even managed to print at 18000 mm/min with low acceleration and it wasn't too bad (higher temp than normal was used). No violent shaking.
For most prints using lower than the default acceleration will do no harm and you can safely double the printing speed without annoying shaking of the printer. However, this has a huge effect on the prints that need supports. That is because of the zig zag nature of the supports there is an increased amount of material deposited at the support corners as well as more material flow at small area supports. This in turns means that your supports need to be spaced far enough from the part that they remain removable. If you do this, the supports end up being further than actually needed in order to provide adequate support.

I think the most convenient and flexible solution would be to have an option in the slicing software so that different accelerations can be selected for infill, supports and outlines. Infill can easily use high speeds and low accelerations while supports and outlines can be printed at high accelerations and low speeds. Also, infill wouldn't need to overlap the outlines anymore as the increased material deposition would do that anyway.
The second solution would be to modify the firmware so that extrusion goes along with velocity, as already stated.
Still, both solutions would pose an additional problem at higher speeds, accumulated pressure in the extruder.
It is really amazing though printing at 9600 at low acceleration. It prints like a different machine. It feels more smooth, robust and precise because there are no sudden moves.
I think Simplify3D, being an amazing paid software, should provide custom codes/macros running right before outlines, supports and infill. This is the kind of versatility I am missing from it.
Maybe there is some way to build a post processor doing this.