3D printed heatsink/hosts.

[Blarg King]

Found this neat stuff:

ColorFabb - copperFill

Essentially its 3D printer filament that has an 80% copper content.

Wonder if thats a high enough amount of copper content to be sufficiently conductive for heatsinking purposes...

Also says it can be polished like real metal... would be cool to be able to print and polish some cool hosts...

May have to get some of this stuff and try it out.

 

[rhd]

I'm just putting this out there...but if this product is designed to melt, it's probably not a great choice of heatsink material

 

[Blarg King]

Its designed to melt at pretty high temperatures. The print head is supposed to be 220-240 C.

I think it stays solid up to 55 C and if your laser gets to 55 C then theres bigger problems. XD

 

[iroquois]

I own 3d printers.
I've heard of the copperFill. Nothing special, they just mix a powder with known plastic pellets and extrude that. But I've heard one roll of such filament is gonna be expensive.

They pobably used a mix with ABS, which gets into plastic state at 100C and melts at 180.
Or PLA, which is lower but above 50 I think.

They won't melt, but there's another problem: home 3d printers are not very precise machines. The "100 or 20 micron layer height" is a marketing gimmick.
Truth is these machines are just melting plastic through a hole and smearing it like paste on a build platform or a previous layer of squeezed plastic.
So even if the machine is perfectly assembled and free of mechanical errors which most aren't, you're still gonna get inaccuracies here and there at few hundred microns range.

I guess you could post process it later to make it usable for snug fit heatsink, but won't work straight out of the printer.

The powder based commercial 3d printers on the other hand which use a completely different technology can make what you want with the resolution you need, but those cost >$30000. Some companies like Shapeways offer you to upload a 3d model, choose the material you want your model to be 3d printed than have them 3d print your model for you and send it to you for a fee. Not very expensive but not cheap either. They even allow you to sell your models on their online store.

 

[crazyspaz]

I own 3d printers.
I've heard of the copperFill. Nothing special, they just mix a powder with known plastic pellets and extrude that. But I've heard one roll of such filament is gonna be expensive.

They pobably used a mix with ABS, which gets into plastic state at 100C and melts at 180.
Or PLA, which is lower but above 50 I think.

They won't melt, but there's another problem: home 3d printers are not very precise machines. The "100 or 20 micron layer height" is a marketing gimmick.
Truth is these machines are just melting plastic through a hole and smearing it like paste on a build platform or a previous layer of squeezed plastic.
So even if the machine is perfectly assembled and free of mechanical errors which most aren't, you're still gonna get inaccuracies here and there at few hundred microns range.

I guess you could post process it later to make it usable for snug fit heatsink, but won't work straight out of the printer.

The powder based commercial 3d printers on the other hand which use a completely different technology can make what you want with the resolution you need, but those cost >$30000. Some companies like Shapeways offer you to upload a 3d model, choose the material you want your model to be 3d printed than have them 3d print your model for you and send it to you for a fee. Not very expensive but not cheap either. They even allow you to sell your models on their online store.

You say you own printers, but 100micron layer height it a gimmick? Not true at all...unless all you have are repraps...

 

[iroquois]

It is a gimmick because
1) Almost all consumer 3d printers today use the same stepper motors and stepper drivers. Just because you have a preset included in your software for your 3d printer telling it how to work at 100 or 20 microns doesn't mean your 3d printer can do it and others cannot.
2) Almost all of consumer 3d printers mention layer height, but not the X and Y resolution. Ever wondered why? Because they have an extruder nozzle with a 0.3mm-0.5mm hole, so you have 0.1mm Z resolution, *but* 0.3mm-0.5mm X and Y resolution.
Why don't they have a 0.1mm hole in the nozzle you ask? Because printing with it would take even longer and a 10x10x10cm print based on the complexity can already take several hours or days to print at 0.3mm. Plus any impurity in your less expensive filament could easily get stuck in the 0.1mm nozzle and cause a clog and failed print. Overextruding to print 0.3mm on a 0.1mm nozzle also causes clogs.
3) There's just so many other things affecting quality more than layer height (like stepper drivers, extruder type, shaft quality, etc).
If these aren't good enough reasons for you, then fine...

BTW, I don't own Repraps, but implying they are worse than the branded 3d printers shows you don't know much about modern consumer 3d printers and repraps. Most brand 3d printers are based on Repraps. Adding a pretty case with lights doesn't make it different when it uses the same GT2 or MXL belts, linear LMXUU bearings, ball bearings, shafts, same motors, same steppers and motherboards, same extruder designs and heated bed designs and same software...

EDIT: Read this: Feedback Friday: Solidoodle's Sam Cervantes Sets Record Straight on Desktop 3D Printing Resolution - Core77

 

[Blarg King]

The printer I own is plenty precise. I've printed planetary gear sets with .5 mm tolerances no problem off mine.

The copper fill has the same specs as bronze fill, which states that its 80% metal with PLA filler, and has a glass transition temperature of 50 C.

I think with 80% metal it would be possible to make a functional laser heatsink. It might not be as effective as a true metal one, but I think of it like thermal paste, where its fine conductive particles suspended in a poorly conductive fluid/gel. Its not the best thermal conductor, but it works well enough.

 

[iroquois]

The printer I own is plenty precise. I've printed planetary gear sets with .5 mm tolerances no problem off mine.

@crazyspaz: That is 500 microns. Just like I said, you'll get tolerance of few hundred microns, so only advertising the "100 micron" Z resolution isn't very honest.


Anyway, if you'll end up printing it, do let us know how it ended up.
Is .5mm tolerance acceptable for heatsink? I guess if you post-process the part where the module goes then yes. Otherwise there's gonna be gaps.

 

[crazyspaz]

I meant the DIY repraps (of which I have never seen a quality print from. And while most printers do use similar stepper motors, you have to admit some are better than others...if I remember to, ill print a single layer and measure it

Oh, and you may want to read the articles before posting...that supports the idea of a .1mm layer height....not sure what the point of posting that is.

 

[iroquois]

I meant the DIY repraps (of which I have never seen a quality print from.

Well maybe you haven't seen many Reprap prints then?
They are literally the same, same hardware and for most printers (like Ultimaker, Printrbot, Solidoodle) all these use use exact same firmware and software, only you have to source the parts yourself or buy a kit which is just all the parts sold by one person + optional assembly instructions, and then assemble it yourself. The price difference doesn't justify the wasted time to me personally to get a Reprap myself.

And while most printers do use similar stepper motors, you have to admit some are better than others...if I remember to, ill print a single layer and measure it

You'd be surprised how many 3d printers use the exact same Nema 14 and Nema 17 motors. And for those that don't, better means higher torque and life expectancy, not accuracy.

Oh, and you may want to read the articles before posting...that supports the idea of a .1mm layer height....not sure what the point of posting that is.

The point is simple: you said the claims of printing at 100 microns being a marketing gimmick is "not true at all", I linked to an article by a 3d printer company owner explaining how almost every machine can do that, proving that it is in fact a marketing gimmick.

What do you mean "it supports the idea of a .1mm layer height"? He just says the companies specify quality in terms of layer height because it is most visible to the end user, but it isn't. And he proves it himself in that very article:

To further compound the problem of specifying the X-Y resolution, it's important to remember that the printhead is extruding plastic through a 0.35mm nozzle. It's important to remember that ABS plastic oozes and expands a bit when it comes out of the nozzle. So imagine trying to define the exact position of a garden hose nozzle on an automatic sprinkler system to within millimeters. Even though the motor on the automatic sprinkler may be capable of positioning the water nozzle to within millimeters, in the end the water is going to going to spray out of the big nozzle over a wide area in a somewhat random manner. The same is true of the plastic extrusion nozzle of a 3D printer, although on a smaller scale.

You're referring to this:

So to simplify the specification of resolution, we typically define it in terms of the layer height, which will be most visible to the end user.

The text that is balded is a lie. Same Cervantes is the founder of the Solidoodle 3d printer, of course he'll say that, But it's a lie. You know it's a lie because of what he said right above that.
Try printing a 1x1x1cm cube. Compare the quality of the top face, where there are 3mm seams with the sides, where there are 1mm (the layer height) seams. Can you say with a straight face that the good quality of the side faces were more visible to you than the poor quality of the top face?
Unless you print very pointy models, the resolution of the X and Y axis will be no less visible to you than the Z axis.
So yes, it's a marketing gimmick. If you still don't believe it is, just ask it in the 3d printing communities. Others already have and have gotten the exact same response as mine: Ultimaker 2 and .02 layer resolution, just marketing or true? (Page 1) ? 3D Printer Discussion ? SoliForum - 3D Printing Community

I think I provided all the info there is, but if you have more to discuss about this, we can talk over pm, email, 3d printing forums or google group. This is getting pretty off-topic.

 

[Blarg King]

Well right now I can't print with the copperfill because I bought the XYZ Printing DaVinci 1.0 printer, and they use proprietary cartridges.

Its a real shame, because the printer hardware is fantastic, but XYZ wants to make it closed source.

Luckily I'm just modding the firmware and getting a programmer module (should arrive in a few days) that will turn my printer into an open source machine.

 

[rusirius]

Are you stating this based on the nozzle or the mechanics? The reason I ask is, I've just finished building acnc machine which uses nema17 motors and gt2 belts. I've used a dial indicator to measure it's resolution. It moves on all axes .006mm per step, and even after repeated movements in all directions it zeros back with only a .004mm deviance.

 

[iroquois]

Inaccuracies come from the plastic extrusion, not the motors, which are pretty accurate. It's a CNC after all.

I don't know how fine the copper particles are in copperFill which will determine how small the hole diameter of your hotend can be without causing clogging. I've seen 0.1mm nozzles for the E3Dv6 hotend. Even if the copperFill filament won't have problems with such diameters, I think you'll still need some thermal grease or similar to use the print as a heatsink properly.