13mm bridge, PLA @ 225C = FAIL

[Geoff]

Trying to print a simple tray for an Arduino Uno. The wall opening for USB measures 13mm across. The bridge at this location is not pretty due to multiple sagging extrusions. The support option in slic3r 0.9.9 was not effective as it only supported a fraction of the 2.1mm wall thickness, leaving the remaining portion to sag.

I am currently operating with the default slic3r config provided by Maker Farm (Prusa 10" i3v) which has the PLA extrusion temp set for 225C along with a bed temp of 70C.

I've done just a little research and this has led me to consider lowering the extrusion temp to ~200C. Good idea, or will that cause problems in the 0.4mm Hexagon hot end? Using 3mm PLA.

Jim

Don't be afraid to play with things, the worst you will get is some clicking...

The bridging in Slicer is in a section on its own in print settings under Speed, have you tried changing the actual bridging speed?

[JimG]

...The bridging in Slicer is in a section on its own in print settings under Speed, have you tried changing the actual bridging speed?

Ummm, no ;-] Was unaware that this was configurable.

I see that it is set for 60 mm/s. Does that seem reasonable?

Jim

[CalifDan]

Jim,

I think that 60mm/sec is way to slow for bridging. I think that is particularly slow at 225C. I don't really use bridging much, but I suspect 120 - 150 might be more appropriate. I noticed that Slicr 1.6 also has a flow ratio for bridging, not sure about earlier versions.

Rather than print all the way up to the bridge on this part (wastes a lot of filament and time if it does not work), you could design something much smaller where an equivalent bridge happens quicker.

[JimG]

I actually need several of the trays, so I can use the failures.

At 190C and 60mm/sec bridge speed, the bridge result was acceptable. Not perfect, but plenty good enough. But I've noticed that the first layer is not well bonded to the 2nd layer.

Trying 200C and 100mm/sec next.

Jim

[CalifDan]

Jim,

In Slicr, you can set a different temp for the initial layer. Better yet, in Cura, you can select which layer the temp changes. For example, I often use 225 hotend and 75 heatbed for layers 1 and 2 and then change to 220/70. Another factor in the initial layers is getting a good first layer. If you are using the standard slicr settings, you may have a first layer of .35 and then .2 after. I have been using .3 and then .2 with success. I also try to get that first layer to squeeze just a little but not so that the bead looks flat. As many here have said, the first layer is critical. Sometimes, I watch the first layer or two go down and just stop it right there and adjust.

I am looking forward (with some trepidation) to setting up the auto bed leveling on the next build.

Dan

[Drone]

For my i3V the standard MakerFarm firmware worked great... as long as I was very thorough with bed leveling. If you're not using the ABL code, bed leveling needs to be as precise as you can get it. Even with the ABL code, good mechanical bed leveling will always be a help I think. I'm using Slic3r 1.20 experimental and get better prints than I did with the 0.99. I also have used Cura and found it didn't give me as good a result on the first layer. It overall gave very good prints, but there was always some strange small gaps in the first layer that wasn't as appealing to look at, but was a good printed part overall. I'm sure the best slicer to use will vary depending on the part your printing and how the slicer sets the tool path for the print.

Another area that took quite a bit of trial and error for me at first was how close to the bed my Z axis needed to be for the first layer. Once I learned how close it should be for a good first layer (this was much closer that I first thought) I started getting very good prints.

I now use Roxy's ABL code, the DACB fork, and while I was getting very good results with the stock MakerFarm code until I got the ABL working I can say that the ABL code is a very good firmware to use that produces very good and very repeatable results. But I would recommend starting with the MakerFarm code and do lots of experimenting with prints until you get a good understanding of how everything works. Pick an object that only takes 10 or 15 minutes to print and try all kinds of different settings. Before you know it you will start to understand how the code and hardware work and you will be able to start fine tuning your prints with a clear understanding of how things work.

[JimG]

The higher bridge speed (100 vs 60) with 200C did not make an improvement. Apparently extrusion temperature is the more important variable for bridges, so for this part I think 190C is going to be the choice.

The loss of bond between the 1st and 2nd layers is not something I have noticed before, and is present at 200C also. The first layer is bonded very well to the glass, to the point that takes a pretty good whack to break it loose (using glue stick treatment on cold bed, then warming to 70C).

But the first layer is not well bonded to itself or to the layer above it across the wide, flat base of the tray object. No bonding issues elsewhere in the object.

Following the latest print and overnight cooling I found the nozzle makes contact with the glass when homed. The nozzle gap looked good when the print started, though.

Z-axis adjustment has been wandering a little, I would guess it varies +/- 0.4mm "on its own." Haven't studied this carefully yet, but I am guessing it is temperature related. It may be necessary to set the Z-stop while the bed is heated? (I have a micro-adjust Z-axis system, similar to the one shipping from MakerFarm these days).

Thanks to all for the input.

Jim

[CalifDan]

Jim,

I have to fine adjust my z axis starting with each day I print. Sometimes, later in the day, I have to do it again. Once I dial it in, it seems pretty good for a number of prints. I do it cold, but heating might make a difference. I use hairspray on the glass. That works well and I only clean it off about every five or ten prints. I shoot a light squirt on before most prints. Other variables must come into play as well. I wonder about how accurate the z axis micro-switch is from one time to the next. I wonder how much temperature of the room might make a difference. We are working with very small increments and it does not take much to make things change.

I am concerned about your adhesion issue. I would give Cura a try (so far 1.7 seems best for my stuff) and try the z axis add on to reset the temperature at lower temperature farther into the print. I noticed that Drone said he had problems with Cura filling on the first layer, but I had the opposite issue. Slicr was missing stuff on the first layer and Cura was catching it all. That is not an opening for a debate. Frankly, and as Drone said, I think it is incredibly part dependent on which slicer program will provide the best results.

I could not even get Slicr 1.2 to load on my computer. It fails at the install. Too bad, I would like to try it out.

Dan

[gmay3]

Hi JimG,

Looks like you're getting lots of suggestions!

Yes you are printing the try in the correct orientation. Typically, I have seen PCB enclosures in two pieces cutting all of the connector ports in half to avoid printing the bridging.

Here are a few examples:
http://www.thingiverse.com/thing:400624
http://www.thingiverse.com/thing:284923

Don't worry about updating slic3r from the stock version Colin recommends. I am using the version listed in my signature right now with no issues.

Here are couple resources that might be useful (as they were to me) to improve slic3r settings and PLA printing

http://richrap.blogspot.com/2012/01/...tings-and.html (3 part series on slic3r tuning, generally a very helpful resource to take a look at)
http://reprap.org/wiki/PLA (shows a range of 180-220 for printing PLA)

[JimG]

....Typically, I have seen PCB enclosures in two pieces cutting all of the connector ports in half to avoid printing the bridging.

The wisdom of that approach is now very clear to me

Jim