Re: New Member: Electrically Conductive Filament Exists

[Tim]

My guess would be that people are not too excited (yet) for several reasons:  (1) carbomorph is not wiring.  "Electrically conductive" does not mean low-impedance.  Carbon is what is used for resistors, and the material will be highly resistive.  The several examples shown in the article are applications that can tolerate (or actually prefer) high resistance in the wiring.  (2) The resolution of existing printers is a bit coarse for anything more than a small wire bundle.  And if it could reach the resolution needed for good printed circuit board work, then the prints would take so long as to be cost-prohibitive.  The semiconductor industry only took off when bulk processes were invented that could print all the transistors or all the wiring in a layer onto a chip, package, or circuit board at the same time.  (3)  Stuff that comes out of an extruder melts at low temperature by necessity.  Try soldering to that. . .  (4) Wired magazine and Popular Science both have a long history of being wildly optimistic about new technology and making wild and sweeping claims about every new advance.

The availability of conductive material to add to printing will be a nice addition when multi-purpose, multiple extruder heads are available, and will be good for simple wiring and battery terminal connections and such, but I would be skeptical of claims about printing computers into a piece of plastic or anything of that nature.

[Matt Minuti]

Based on some of my own experiments with making conductive
glues/pastes, even though lampblack can be very conductive, the binder
material is usually a pretty good insulator, and you wind up with
something that's only usable as a megaohm+ resistor. So at least in
this amateur's experience, it's going to be a bit harder than just
mixing some pigment with elmer's glue ;)

At the same time, things like the Lyman Filament Extruder might make
it more doable, though I don' t know the insulative properties of ABS
or PLA offhand...

> --
> You received this message because you are subscribed to the Google Groups
> "MakerGear - Make Today, Change Tomorrow" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to makergear+...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Allen Jeter]

Along the lines of dual extruder though not related to being
electrically conductive. Did anyone see that the Cube X printer from
3D Systems is using PLA to print soluble support for ABS. This printer
looks like the result of 3D systems acquisition of Bits from Bytes as
it looks like and Ed Sells design to me. It's a really clever idea I
had not seen anyone else do yet.

--
Allen Jeter
Lumensa LLC
www.lumensa.com
720 975 6717

[Justine Haupt]

Those are very good points, thank you. On the issue of resistivity, I just checked out their paper and found the relevant paragraph:

The printed composite was tested for its resistivity both in-plane of the printed layers and perpendicular to the layers. Tests for resistance were carried out on 5 mm cubes of carbomorph using a two-probe measurement with the two opposite cube faces painted with silver conductive paint to minimise contact resistance. The measured resistivity of the composite in-plane with the layers was 0.09±0.01 ohm m⁻¹. Perpendicular to the layers, the resistivity was 0.12±0.01 ohm m⁻¹. A reduction on the resistivity of 25% was encountered when moving from the perpendicular resistance to parallel resistance mode. The difference in resistivity is explained by the way in which the blocks were printed. In the plane of the layers, the printed filaments provide a complete conductive path between electrodes. Perpendicular to the layers the establishment of a conductive pathway is reliant upon melting between printed layers. Current-voltage (IV) analysis was carried out on the printed composite cubes in both orientations between −5 and +5 V using a potentiostat and showed the IV response in both orientations to be linear.

Unless I'm misreading this, it looks like the conductance is actually extraordinarily good: .1 Ohms per meter? Could that be right?

To me, the power of something like this wouldn't be in making conventional-esque (2D layered) circuits on a 3d printer, but rather in expanding the electrical design possibilities by removing the constraint of having to lay circuits out in layers.

~Justine

[Tim]

Unless I'm misreading this, it looks like the conductance is actually extraordinarily good: .1 Ohms per meter? Could that be right?

Part of the "extraordinarily good" figure comes from the cross section;  the test geometry was a 5mm cube, so that's a cross section of 25mm^2.

If you scale that down to the cross section of a 100um-diameter extruded filament, that's a cross-sectional area difference of about 3000, which

means that a thin extruded wire is about 300 ohms/m,  making a 1cm long printed line somewhere around 3 ohms.  Which is still not bad;  I'm

pretty impressed by that.

 

To me, the power of something like this wouldn't be in making conventional-esque (2D layered) circuits on a 3d printer, but rather in expanding the electrical design possibilities by removing the constraint of having to lay circuits out in layers.

I agree completely.  But, I have heard other people gushing forth about the ability to print entire logic circuits on plastic, and wanted to dispel any such notions.

I can come up with a number of interesting applications myself;  with a short printed wire less than 1 ohm and at the approximate size of a bond pad on some less aggressive fabrication processes, one could possibly use the 3D printer as a quick & dirty chip-on-board wire bonder.  MEMS applications making use of the variable resistance are the most likely.  Applying directly on top of a chip is an advantage for extruded plastic, because I would not want to  try that with laser sintering.  The filament process should be pretty ESD-safe, too, with decent grounding.

[Matt Minuti]

This is why conductive paints/films are usually specified in
ohms/square instead of a linear distance. Still, those numbers are FAR
better than I would have expected. I might just have to read through
and see if I can apply any of their techniques to some quick home
tests...

[Justine Haupt]

>Part of the "extraordinarily good" figure comes from the cross section;  the test geometry was a 5mm cube, so that's a cross section of 25mm^2.

If you scale that down to the cross section of a 100um-diameter extruded filament, that's a cross-sectional area difference of about 3000, which

means that a thin extruded wire is about 300 ohms/m,  making a 1cm long printed line somewhere around 3 ohms.  Which is still not bad;  I'm

pretty impressed by that.

Okay, right, not quite as spectacular, but still usable in place of wire in many instances. And 100um is very small. For something like this perhaps .5mm would make a fine wire, and that would bring it to ~.13 ohms/cm. If instead of wire-like structures this material were used for its structural strength as well (I wonder how strong it is), large volumes of the innerds of parts could be made as conducting paths, and resistance would really become negligible.

>I might just have to read through
and see if I can apply any of their techniques to some quick home
tests...

Maybe you would be the first. I can't find any info or pictures of material people have actually made. The process of making it seems a bit daunting though. After rolling the material into filament between the glass plates, how to get it on a spool? Ugh.

I would really like to see this be made available commercially.

[Jerrill Johnson]

I've been looking for anyone who duplicated the University of Warwick's carbomorph results for about a year now. I even hoped that someone would have tried to commercialize it by now, but all I see are links to the same article from a year ago. Nothing else. Has anyone seen real samples of this material?

Jerrill