Infill types and percentages?

Through my tests I’ve been using gyroidal infill and varying degrees of infill percentage. I’m wondering if anyone out there has tested different types and percentage against others? Full disclosure, I haven’t gotten a true success yet, but it might also be the model shape in printing.
Also… if everyone just uses 100% infill that would be good to know too.

Using 100% infill - I would imagine anything else would slump into the cavity due to lack of support. Might be able to get away with lower infill that changes direction on each layer - e.g 60% infill 45 degrees on one layer and the -45 on the next. Haven’t tried that yet though. This might reduce the slump and save some filament.

I’m printing a couple different variations now, we’ll see what happens :man_shrugging:
If nothing is more promising than the other types, 100% just may have to be what I use

If you plan to sinter, I think you will find that 100% infill will typically provide the best results. In many cases when sintering if there are unsupported areas you will get sagging due to the heat needed for sintering as others have suggested. In some cases you may be able to intentionally design and orient your part with the idea of limiting sagging in unsupported areas during the sintering process.

All that being said this is very new to most and nobody has all the answers. It is likely that with experimentation like you are doing we will all learn new things. Thanks for the questions and posts. Look forward to hearing your results and keep going with the experimentation. This is how we will all learn more about what the limits are and what is possible.

1 Like

Agreed, already learned a few things from others and certainly plan to share my results :+1:

To save on filament costs I’ve scaled my upa’s down by half and am trying 3 types of infills at the same percentages.
I tried a 40% tri-hexagonal and noticed a completely different end result so I feel like there’s something here…

Also, I know shrinkage of infill can cause warping as it comes apart at different rates than the walls, but I also imagine gas escaping can cause an issue… Will try to drill a small hole in one and leave the other as a control to see if it makes a difference

All for science!
download (1)

1 Like

Haven’t cleaned them up but… my scaled down versions seemed to work (40% infill I believe)

I should really also try to maintain one variable that I change… but I also used my casted sinter “boat” because it fit a lot more all at the same level (so this might have contributed to success since its a different material than my graphite crucible)

I have only been printing with the bronze filamet for about two weeks. After adjusting some print settings and filament settings in my slicer, I was able to get 2 successful prints after about 10 failed ones. The first two objects I sintered produced tears, deformations, and a rough, granular surface on the results. The tears and slumping made me think about the infill problem. I have tinkered in metal fabrication and jewelry making, one of the big lessons concerns hollow-form pieces; that is, how to apply heat to a hollow item without negative effects.

When the temp increases during sintering, the air trapped within the infill will also heat and start to expand. This puts pressure on the walls of the cell containing the air. At the same time, the temperature of the metal is approaching its melting point, which is exactly when the metal is at its weakest. So increasing pressure inside weakening metal leads to rends and tears in the metal as the hot air attempts to escape.

The remedy is to provide a way for the air to escape (i.e., holes in the printed object that link all of the infill chambers and can allow the hot air to leave the printed object without causing tears and other problems).

The other solution is to not trap air at all and print at 100% infill. For my objects of interest, which are small and flat, this approach does not increase cost or time that much. One important aspect is also minimum top and bottom horizontal shell thickness. If the shell is too thin, it will be more likely to slump around the infill , the infill patter will show on the surface of the sintered item, and it will most likely tear at some point.

2 Likes

Lines up with the testing I was doing (gas blowing out the top). Actually was going to drill small holes in these little upa’s but at the time I didn’t have a small rotary bit. One alternative may also be to introduce liquid phased sinter + lower infill parts.

I’ve been in pyrex mode for a while but have some more tests coming with this.
The idea is that the infill will be filled with the other lower melting point metal (maybe reducing gas expansion or offering some level of support?)

I’ve noticed that this “puffing up” happens sometimes even in fully solid parts and can happen during the debinding process itself, and which then leads to some of the ripples/porosity in the final part. I would think that after being debound there are many physical openings for the gas to escape through during the sintering process. After all, it is permissible to print with (closed cell) infill when using markforged filament, with the main difference between us and them being their chemical (non gas-forming) debinding process…

Trapped gas can also leave by being dissolved in the metal and transported to the surface (a limitation for inert gasses like argon which can get trapped Effect of entrapped inert gas on pore filling during liquid phase sintering | SpringerLink).

I guess a big question is, if it can be a good design process to place holes through the part to encourage thermal debinding, or some other useful trick.

I’m not sure how the gasses escape during debinding and I don’t know much about physics or metallurgy, but it seems like the ratio of trapped gas in relationship to its encapsulating structure would be a determining factor. A large air pocket with thin walls could be problematic whereas microscopic air bubbles diffusely trapped throughout the structure and surrounded by relatively thick walls might not be a problem. Also, my prints are relatively flat and any type of infill causes 30% or more shrinkage along the z-axis.