For a project I'm working on, I need to manually laminate an extremely thin, stretched Mylar membrane (around 1μm) with a gold leaf.
I've had success so far with using mist spray glues, like 3M 77, but I would like to make sure my glue layer is as thin as humanly possible without a special application process.
I was talking to some guy on the internet who's doing something similar as I am, but he refused to explain what glue he was using, he was just willing to tell me this:
'I apply the glue with my hands using a special roller. I make my own glue. based on components that only I have. The layer dries after application and becomes 8-9 times thinner. Its peculiarity is that when it dries it does not lose its uniform thickness. The second feature of the glue is its stickiness for three days after application, then it becomes the same, consistently elastic, and completely loses its stickiness.'
He told me that by using this technique, and his specially formulated glue, he was able to leave a glue layer that is only a single micron thick.
Does anyone know their shit and about glue and could point me in the right direction? I would like to reproduce this myself. I offered to pay the guy but he didn't care.
You can literally soak off the adhesive used on tape with a light solvent, re-apply it, snd the solvent will evaporate leaving a thin film.
>You can literally soak off the adhesive used on tape with a light solvent, re-apply it, snd the solvent will evaporate leaving a thin film.
They sell spray adhesive. Lightly spray the surface from a distance and mist it with acetone if you need to thin it on the applied surface. That should give you an ultra-thin layer of adhesive. And you might want to put a dust filter on a box fan if you're trying to be accurate. Just a thought. Good luck!
> ingredients only I have
cum
Poverty threads annoy me so much, why not just use established methods to achieve your goals such as sputter coating, electroless gold plating, or even something as basic and simple as bar coating?
Could be for a school project. I had to make some of my own apparatus from random junk that was in the lab or stuff I could scrounge up myself.
Sputter coating is beyond my means, electroless gold plating doesn't seem to apply to my particular use case as the receive substrate is plastic, not metal, and I'm pretty much trying to do a poor man's bar coating.
You're close, I'm building planar magnetic headphones drivers.
At the moment, it's a personal project. So far, I've only used 5μm aluminized mylar, photoresistive lacquer and chemical etching with a lot of success and very clean results. But the thinner I can get, the better my driver will be. I have a few rolls of 1μm mylar and a roll of 0.5μm mylar as well, neither of them aluminized.
I have had access to the most expensive planar headphone out there. It's 6K$. I opened it up to examine the membrane and was shocked when I realize that they're probably handmade, and by the looks of it, not using any ultra fancy jig or rig. It is literally your run of the mill 24 carat gold leaf that's been gilded onto a mylar that is probably 1μm.
See attached pic. You can very clearly see that it's gilding leaf, applied by hand, the creases are a telltale sign. If it was a fancy deposition technique, it would be a lot shinier. It's literal gold leaf, which has the advantage of being just 0.15μm on average.
I figured that with my own means, I could come up with a process and a set of tools to allow me to come very close to what I was seeing.
I'm going to look into the mist machine idea, it could work if it doesn't immediately clog the thing.
I gave screen printing a try already, but the layer thickness I'm getting for using the thinnest mesh density I have (which may just be the thinnest you can get I think) isn't too reliable.
Melting the plastic won't do it, it's already much too thin and I don't have any control over it. It is so extremely thin that it can be torn by a gust of air at times.
>planar magnetic headphones drivers
>not going full electrostat
>i shiggy diggy
Anyway, given this new information, I would do the following:
Find a suitable adhesive that can be thinned enough to blow through an airbrush. Acetone apparently does not affect mylar, so CA glue thinned to a watery consistency with it (probably like 20-30% acetone) and sprayed on should work.
Create a stencil to mask the prepared diaphragm. Spray evenly with the adhesive, and let dry just long enough that it's barely tacky. Press your gold leaf between the film and a block of PTFE. HDPE may also work okay, but, either way, the surface should be very smooth (not textured like a cutting board). Use a flat object with something like neoprene foam on it to ensure even pressure.
Allow to cure. After it has, give it some attention with a paintbrush to remove the un-adhered portions of gold. You'll have to do some experimentation with thinning out the adhesive and practice applying it in order to get a reasonably even, thin film thickness, but this should work pretty well. You could skip the stencil and instead just etch the gold after the fact if edges of the stencil collect too much adhesive.
Good fricking luck verifying the actual thickness of the adhesive layer, though. I'm very well equipped, and even I don't have anything that can measure reliably in the single-digit micrometer range (yet).
I wish I could go full estat. It's just that, as a DIY venture, it's much easier to just make a membrane, CNC mill an enclosure, slap contacts and plug your headphones vs. having to acquire or build a complex amp and deal with voltages that would kill you in a heartbeat. But yeah, I've never recovered from my encounter with a flagship Stax.
I did find some specialty CA glue which seem to yield very good elasticity once dry. I ordered some and will try it out.
I think a combo of a 2um Mayer Rod and diluted CA would get me very close to a layer that would end up close to 1um or less once dry.
I'm sadly not equipped to measure that kind of thickness. I heard even Audeze has trouble measuring their >0.5um membranes even.
I'll just trust the process, should be good enough. Hifiman does a terrible job at it and their shit still sounds fricking amazing.
I'm also going to just coat the entire thing, gild it and etch it with a photosensitive lacquer instead of trying to glue a stencil to the membrane. Etching works wonder and since I have a high-res printer, I can achieve some insanely thin layer lines.
Do you DIY electrostatics?
>Hifiman does a terrible job at it and their shit still sounds fricking amazing.
Not exactly surprising. You already know both PMDs and electrostatics are inherently superior in terms of performance, as long as you can deal with their peculiar drawbacks.
>trying to glue a stencil to the membrane
Wasn't suggesting you glue it on, just set it on there to pattern the adhesive layer. Like how you would do with spray paint stencils. Actually just now occurring to me that lifting it off the membrane with a few tabs of paper would prevent the "adhesive collecting in the corners" issue. Not that it matters if the bar coating works out, but keep it in mind.
>Do you DIY electrostatics?
It's on the backburner. Along with like 15 other projects. The idea of having a wireless, open-back electrostatic headset/microphone pleases me greatly, but it's fairly low on the priority list right now.
>They sell spray adhesive
I don't think you realize just how small 1µm is. The line left by a *sharpie marker* is in the range of 2.5-3.0µm: https://www.youtube.com/watch?v=46DBNUfhATo
Typical film thicknesses for canned spray adhesives is in the range of 2-10 mil. That's roughly 50-250µm. Any adhesive I can think of needs to be heavily thinned in order to get anywhere near single-digit micrometer thickness when sprayed.
What about electroless copper as a base metal? The process is well documented as it's widely used for PCB via plating, works on most surfaces if you treat them with activators like Sn/PdCl2 or more modern Sn/Ag formulas. For corrosion protection strike nickel layer should be sufficient.
If you insist on using gold leaf I would try spin coating epoxy diluted with appropriate solvents, sub um layers are achievable. Curing would be best done with a miniature (heated) laminating press, that's how strain gauges are made only they use slightly thicker materials. Both contraptions can be DIY'd to work on a small scale
Damn, thanks for pointing me to bar coating, that's exactly what I was looking for. I'm ordering a bar coater than can deposit a 4um layer height. That's perfect.
Now I just need to figure out a glue that gradually thins as it dries without shrinking.
I get the feeling you're trying to make a microphone diaphragm. Was actually thinking about doing this a while ago, even have a turbopump and controller stashed away for the eventual vacuum chamber build.
Is this a personal project or commercial? If it's commercial, just bite the bullet and put together or buy a PVD rig. If it's just for the lulz, it's probably not worth it over just using aluminized mylar.
Getting coatings as thin as you want them on the kind of substrate you want it on is extremely difficult. The best I could possibly think of is to heavily thin out something like CA glue (might have to use something else if the glue/solvent dissolves the mylar) and use spin coating to distribute it onto the unfinished capsule.
Screen printing with ultra fine mesh screen can apply uniformly thin layers of material.
Ultrasonic atomization is also used to do thin layer spray coating, the commercial stuff is pricey but you might be able to bootleg something using a cheap diffuser to fog a water thin adhesive on your piece.
Another option for this kind of thing is to apply a solvent that will melt the plastic surface and make it tacky, which can eliminate the need for adhesive altogether...mylar is pretty impervious to most solvents but allegedly Tetrahydrofuran (THF) (aka oxolane) will do it. Might be tricky on something thin and stretched, but at least it won't melt the gold.
FWIW, this might be of interest to you, if not directly helpful-
https://www.nature.com/articles/s44172-023-00093-0
>Does anyone know their shit and about glue and could point me in the right direction?
In composites, they want low resin content because less fiber crimp = more stiffness. In the resin infusion process, they use low-viscosity epoxy and it's introduced after the parts are compressed together with a vacuum bag.
If you wait for the glue to get smaller, it doesn't seem like the parts are going to get closer due to gravity. On the small scale, gravity is not very strong compared to vicious forces. Whichever method you use, you should control the bond line thickness with mechanical pressure, like under felt on a flat surface plate. You can't introduce the adhesive later because there's no voids in the material.
That's a very good remark, I hadn't thought of that.
Fortunately, since I'll be using gold leaf, it will have to be burnished on the surface using a very soft brush pushing against it to flatten it out and remove the folds.
But I'll make sure to press it under felt afterwards for a while to enhance the bond.
I guess I'll have to make sure my layer is as thin as possible. I'll DIY a coating rod that can do 2um, I couldn't find anything under 4 that I could access easily without getting special quotes.
>I was talking to some guy on the internet who's doing something similar as I am, but he refused to explain what glue he was using, he was just willing to tell me this:
>
>'I apply the glue with my hands using a special roller. I make my own glue. based on components that only I have. The layer dries after application and becomes 8-9 times thinner. Its peculiarity is that when it dries it does not lose its uniform thickness. The second feature of the glue is its stickiness for three days after application, then it becomes the same, consistently elastic, and completely loses its stickiness.'
sounds like some bullshit he completely made up to try and act superior or smarter than you on the internet