What's new
What's new

What manufacturing process would you recommend for this design?

Rezathema

Plastic
Joined
Sep 22, 2019
Hi all,

I need to manufacture a piece which basically looks like the attached images.
How would you create such narrow gaps? Laser cutting? EDM?
The gaps are around 0.004'' and the depth is 0.04''.
The material would be Acrylic.

Thanks
 

Attachments

  • 2.JPG
    2.JPG
    47.6 KB · Views: 181
  • 1.jpg
    1.jpg
    81.4 KB · Views: 161
Can you say anything about quantity and surface finish? For a smaller run you could mill them or potentially use a slit saw. I see McMaster has slit saws down to .006" thickness so you'd have to see if you can get .004", but you can definitely get endmills at that diameter, though that is non-trivial CNC work. For larger quantities you could mould them but it would help to have a couple of degrees of draft angle and the mould might best be made by EDM. I would be curious if you could hot emboss something at this aspect ratio. I'd be dubious as the hot embossing I've seen is more like .5xdepth (for microfluidics) whereas this is 10xdepth, but I have no direct experience. I would also ask about surface finish as if you are using acrylic you might be concerned about optical clarity, if this is an optical part of some sort. One other thing; do you have a good engineering reason for using acrylic? Acrylic is a fine material which is affordable and has unique properties and is definitely the best choice for many applications. It's a terrible choice for others however so don't use it just because there's lots lying around and it's easy to laser cut!

NB I see you are in BC. If this is a university research project that is life-science related I might be able to help directly... If you need a metal mould you might want to talk to our good friend Marcus (Implemex) here.
 
...NB I see you are in BC. If this is a university research project that is life-science related I might be able to help directly... If you need a metal mould you might want to talk to our good friend Marcus (Implemex) here...

Oh boy, oh boy, oh boy......a real made in B.C. project. :D For a one-off job a slitting saw might do the trick
although the width of the cut and the spacing is really tight. You're going to need a solid, accurate machine
to hold those sizes. If you can't find a .004" slitting saw you could probably grind a thicker one down to size.

After many years of playing this game my first thought when I see a project like this is: "How do we re-engineer
this to make it simpler?" Not always possible but if there is a possibility it can save a lot of time and money. A
fresh pair of eyes will often see things a bit differently...
 
I would strongly suggest stepping back and explaining the problem you want to solve rather than asking how this can be made.

I have machined similar size grooves in aluminum for optical fiber packaging. Those parts were [larger and] most of a grand a piece in serial production. Not a cheap proposition, in other words.
 
Surprised you think EDM or laser cutting might be a choice for cutting grooves in acrylic plastic, unless you're making a mold for casting multiples? Given the aspect ratio of the grooves and the lack of draft etc. that doesn't seem right either.

I'd suggest you share the design intent of this block of grooved acrylic to get a decent answer.
 
Thanks. It is not for mass production. It is a simplified version of a microfluidic setup. I have considered slit saw and you are right I can see pretty thin slit saws on the web, and to be honest, I could compromise and choose the thinnest slit saw available, be it .006''. My problem is I cannot find anyone doing it for me.
Optical clarity is not an issue as this is not an optical device and a bit of rough surface does not hurt me at all.
And yes, we have chosen Acrylic for a lot of reason, it's mostly an academic exercise though.
I would be more than happy if anyone introduces a local business that handles samples like this.
 
And this device is basically a heat exchanger. So the fins are similar to what you have as a CPU heat sink. The geometry is chosen in this way for an academic exercise. The other side of the bock will have a micro-channel, but since they will be in the order of millimeters, I do not worry about them now. I figured the first step is finding a way to make such thin grooves first.
 
Sounds like the cross-sectional area of the groove might be more important than perfectly parallel sides, etc. If so, might engrave it?

The skiving suggestion could also work. No matter what you do, chip evacuation and keeping the plastic from melting may not be easy. A saw, running slow with water coolant, might be best if you can get one made or re-ground to .004". With a 10:1 depth to width aspect ratio and such thin sections in acrylic, this thing is also going to be fragile. Most any university prototype shop should be able to give engraving, skiving, and saw slitting a try to see if you're happy with the finished channels.

I do wonder about acrylic as a heat exchanger, but will assume there is a reason for this. To see flow???

Could also be you could laminate this out of .004" sheet, every other sheet displaced .04 and glued. The film would likely have better structural integrity than engraved/skived/sawn acrylic. You could probably find both a clear film and a double-stick tape near .004".

FWIW, microscope cover slips can be had near .004" thick and a stack of them is surprisingly strong. Be a total pain, but a stack of those, every other one displaced, could be embedded in epoxy at one end. Much like gage blocks, cover slips tend to stick together on their own. Even easier to laminate out of a metal, perhaps alternating with a film or clear double stick tape if you want some visible access???

You could also make a mold with .004" x .04" strips of aluminum tensioned across the bottom. Cast acrylic over it. Dissolve out the aluminum.
 
If it's a heat sink couldn't you relax the .004" to .006" and then get a stock slit saw? Acrylic is definitely a fine choice for microfluidics (fun fact, acrylic doesn't autofluoresce) but like all plastics it has such terrible thermal conductivity I'd be surprised if you got much cooling from a fin structure. Wouldn't it be better to run a cooling fluid through a passage close to your fluidic channel?
 








 
Back
Top