Dye-sub DIY (crazy idea?)

[damorgue]

Several people have mentioned that the dyesubs of HHKB are sharper than those of other manufacturers such as imsto or BSP. I noticed that Realforce and HHKB don't appear to be fully saturated legends on undyed background. They look more like dark grey on very light grey. I started thinking of reasons why, and I came up with a theory. I make the assumption that the dye follows the general rule of diffusion; that the rate of diffusion of the dye into the polymer depends on the difference in concentration. The theory is basically that they saturate the legend with dye, after which they dye the entire cap a light grey. They might even bleach the entire cap after this has been done.

Explanation:


Observe the following example where I applied this. Both of the "FONT" below were blurred the same amount, but the lower one was treated by introducing clipping. I exaggerated a bit which makes it really evident. The same effect although less of it would be achieved if the background had only been made a little bit grey.

I think treated lower one appears sharper. Does this theory sound reasonable at all?

[Krogenar]

I think treated lower one appears sharper. Does this theory sound reasonable at all?

I dyesub fabrics in my business (for the past 20 years), and we experience this sort of bleeding once in a while. What we do is we reverse print to paper with dyesub inks, which are then sandwiched on top of our target fabric. Then we run them both through a heatpress. The press gets to a high temperature and the pressure holds everything in place, so there's no slippage.

The way it works is that the ink (started as a liquid) is now in a solid state on the paper, but under high heat will become a gas -- skipping over the liquid phase entirely -- and work its way into the fibers. But still, even as it is a gas, there's some expansion of the gas, which is what probably accounts for the fuzziness. Once in a while I have a client who wants a super, super thin white rule line, on a black background -- and if it's thin enough, it will "fill in" a bit. If my pressure settings aren't right, it will be worse. Also, after coming out of the press, the paper and the fabric have to be separated and stay separated, or a double-hit shadow can occur. Also, putting down more ink than is actually needed will "blow-through" to the backside of the fabric. So I use a third substrate, a tissue, to catch that outgassing.

So, based on that, I think that's the phenomena you're seeing with the keycaps -- outgassing. The ink becomes a gas and doesn't go perfectly down into the substrate, it feathers out a bit. That said, I have no real idea how the keycap manufacturers are dyesubbing their caps. Maybe they're using the vacuum-sealed film (seems like overkill to me, but ideal if you were going to dyesub the entire surface of the keycap), or maybe they've just got a jig that allows for paper and a sandwich press sort of heatpress. I could see the paper route working if they're doing whole trays of just a single kind of letter, and they blanks are cheap enough.

I would love to see someone use the film to dyesub the entire surface of a keycap -- just so you could say it's been done.

Right, sorry -- damorgue, your theory is absolutely plausible. The contrast between the black-on-white is what makes the the black-on-gray look so much better.

[Matt3o]

here you can see the gold film in action

[Krogenar]

here you can see the gold film in action

Notice how the blank cellphone cases and the gold film are sort of 'prepped' in the top part of the machine? I guess you can't do this without preheating them both. Also, check out 1:02 minutes in -- 119 degrees C. Can ABS keycaps withstand that kind of temperature?

[Matt3o]

Can ABS keycaps withstand that kind of temperature?

No, that would be PBT, I don't think dye ink would stick ABS anyway (even if the material could resist 100+ C)

[Krogenar]

Here's another video, 8 cases at once.

Naturally, no prices listed.
http://sublimet.com/sublimation-products/sublimation-film/

[Krogenar]

Found this: http://www.dyesubforum.co.uk/vbforum/showthread.php?7532-3D-Sublimation-Using-Film

People are apparently making phones cases (hella harder than the surface of a keycap) using paper. But from the discussion it sounds like film has some issues as well.

[Matt3o]

Naturally, no prices listed.
http://sublimet.com/sublimation-products/sublimation-film/

I tried to contact them. THey are from Spain. Let's see.

[nkzt]

In my experience, I've never felt that bleeding is a problem. But I haven't tested such characteristics. Test chart for that purpose may reveal it. I'll try it.

I've encountered double exposure like shadow sometimes. I thought that the film slipped while heating. Cooling is important? hmm, I'll take care.

The gold film video is interesting. As far as I saw, the film is impenetrable and they used a hole punch for alignment (and for fixing too?). Similar and more sophisticated trick may be required for keycaps. Good luck, Matt3o!

I have some iPhone blank cases, free gift of transfer film. Some are hard (likely PBT) and some are soft (no idea). The soft cases deform under 200℃. Perhaps they are for low-temperature dye-sub ink.

[Euf0ria]

Regarding the video with Phone case and goldfilm:
Think of the positibility of this:
Instead of placing two pieces in The fixture, why not place 105 keycaps at a time and then cover with a goldfilm with the whole 105 keycap layout/design all perfectly aligned and then make a whole set in One go!
Mind = blown ;-)

[Krogenar]

Regarding the video with Phone case and goldfilm:
Think of the positibility of this:
Instead of placing two pieces in The fixture, why not place 105 keycaps at a time and then cover with a goldfilm with the whole 105 keycap layout/design all perfectly aligned and then make a whole set in One go!
Mind = blown ;-)

What I picture is taking a long section of the film and painting it directly with dyesub inks (in an artful way), letting it dry, and then putting it either into a rigged up oven, or using the blowtorch method to get the film to then conform to the sides of an array of keycaps. All these companies are very tight-lipped about the specs on the devices. Not a single one I found would declare prices or the temperatures necessary: just buy our stuff! *grumble*grumble*

I put in a request for more information to one of the companies, so hopefully I can get some information.

Nkzt, where did you buy the film? Which company did you use, if you don't mind me asking?

[nkzt]

Nkzt, where did you buy the film? Which company did you use, if you don't mind me asking?

As I wrote before, I bought my film from Shenzhen Basya http://www.aliexpress.com/store/735699. The item page is http://www.aliexpress.com/store/product/A4-size-transparent-film-for-3D-blank-phone-case-sublimation/735699_1489636166.html (They sell it again). I don't know the manufacturer of the film.

Regarding the video with Phone case and goldfilm:
Think of the positibility of this:
Instead of placing two pieces in The fixture, why not place 105 keycaps at a time and then cover with a goldfilm with the whole 105 keycap layout/design all perfectly aligned and then make a whole set in One go!
Mind = blown ;-)

In my case, plaster jig doesn't have enough strength for such large structure. Stainless steel or aluminium alloy can make such jig, but it costs much. In the future, if I come to make 100 sets, I'll invest to jig and machine (remote possibility).

[wcass]

Regarding the video with Phone case and goldfilm:
Think of the positibility of this:
Instead of placing two pieces in The fixture, why not place 105 keycaps at a time and then cover with a goldfilm with the whole 105 keycap layout/design all perfectly aligned and then make a whole set in One go!
Mind = blown ;-)

What I picture is taking a long section of the film and painting it directly with dyesub inks (in an artful way), letting it dry, and then putting it either into a rigged up oven, or using the blowtorch method to get the film to then conform to the sides of an array of keycaps. All these companies are very tight-lipped about the specs on the devices. Not a single one I found would declare prices or the temperatures necessary: just buy our stuff! *grumble*grumble*

I put in a request for more information to one of the companies, so hopefully I can get some information.

Nkzt, where did you buy the film? Which company did you use, if you don't mind me asking?

one of the "iphone case dye sub with gold film" videos on Youtube that i watched included preheat times (few minutes - don't remember) and temp (130 C). but i think the video was in Portuguese. it might have had info about printer and ink too, i just don't remember.

[Krogenar]

Regarding the video with Phone case and goldfilm:
Think of the positibility of this:
Instead of placing two pieces in The fixture, why not place 105 keycaps at a time and then cover with a goldfilm with the whole 105 keycap layout/design all perfectly aligned and then make a whole set in One go!
Mind = blown ;-)

What I picture is taking a long section of the film and painting it directly with dyesub inks (in an artful way), letting it dry, and then putting it either into a rigged up oven, or using the blowtorch method to get the film to then conform to the sides of an array of keycaps. All these companies are very tight-lipped about the specs on the devices. Not a single one I found would declare prices or the temperatures necessary: just buy our stuff! *grumble*grumble*

I put in a request for more information to one of the companies, so hopefully I can get some information.

Nkzt, where did you buy the film? Which company did you use, if you don't mind me asking?

one of the "iphone case dye sub with gold film" videos on Youtube that i watched included preheat times (few minutes - don't remember) and temp (130 C). but i think the video was in Portuguese. it might have had info about printer and ink too, i just don't remember.

Yeah, there was definitely a preheating phase to all this -- not just for the cases, but also for the film. The film looks like it was just inkjet printed. You could buy a liter of each of CMYK and you'd be good for a loooooong time. A single drop of the stuff is probably overkill. If they offered samples of the ink and the film, that would be perfect.

[Matt3o]

the preheating makes the process easier/better, but it's not strictly necessary. I've seen other videos where they do not preheat the materials.

I tried to contact 3 companies but none of them replied, I think the want to keep everything a bit "close", they don't want people start dye-subbing at home.

[Matt3o]

I've been contacted by one of the companies. The cost of the whole system is around €3000 for the oven and accessories and €1000 for the printer and dye sub film.

[bueller]

I've been contacted by one of the companies. The cost of the whole system is around €3000 for the oven and accessories and €1000 for the printer and dye sub film.

Hrmm, I wonder if we could DIY the oven....

[nkzt]

I've been contacted by one of the companies. The cost of the whole system is around €3000 for the oven and accessories and €1000 for the printer and dye sub film.

€4000 seems regular price in Japan, too. This price includes every possible help, I suppose. Such price may be reasonable if you are going to run iPhone case printing shop. Without their help, your shop may be suffered from the technical and supply instability.

But they have never tested printing on keycaps. They cannot supply any keycaps and any information about printing on keycaps. You want a oven? Buy it at Aliexpress or Buychina. Around $900. http://www.aliexpress.com/wholesale?SearchText=3d+sublimation&catId=0&initiative_id=AS_20140120062634

But I don't want oven yet. I can transfer dye-sub ink by common heat gun and vacuum cleaner. Oven helps much definitely, but expensive and space-occupying.

[Matt3o]

temperatures are not high, my kitchen oven reaches 220C so theoretically you can do it at home. the problem of course is fixing the film on the caps, but anyway your solution, nkzt, is simple and brilliant.

[nkzt]

I've printed a test chart on PBT keycaps. I've found ink bleed. About 2 px in 600 dpi. How can I compensate it? I'm not well informed such kind of problem.

[Matt3o]

what printer do you have? most printers have a heads alignment feature that you can run to optimize printing.

[nkzt]

Thanks Matt3o, I've tried "gap adjustment" just now. Vertical line becomes much better (printed on paper, not tried on keycaps yet).

But the bleed comes from dye-sub mechanism too. As damorgue wrote ever, deeper color bleeds more (measure the photo by a ruler).

[Matt3o]

I see what you mean. There's little you can do about it I guess, apart from making dark graphics a pinch smaller I guess.

[Krogenar]

Even with the color bleed, how well does a single character reproduce onto the keycap? Is it of acceptable quality? Can you post a picture of a keycap that displays the dot gain or bleeding?

[nkzt]

I see what you mean. There's little you can do about it I guess, apart from making dark graphics a pinch smaller I guess.

I've googled several hours and got a conclusion. The technology what I want is "non-blind image deconvolution." This is a black magic of mathematics and there is no handy way to use it for printing. I end up with as you wrote.

Even with the color bleed, how well does a single character reproduce onto the keycap? Is it of acceptable quality? Can you post a picture of a keycap that displays the dot gain or bleeding?

This problem will much suffer the printing quality when the image contains a lot of thin lines. I have no plan of printing such image for the moment, but I love general solution.

[Giorgio]

This was a wonderful idea.... too bad that it has stopped :-(