Cheaper LED's and a look into LCD for Exposure Split from Saati Screen Toaster Thread

[Maxie]

This looks a lot like the Saati 300w unit but who know what intensity it has, nm, etc?
The Saati unit also has two fans I the back and I can expose 2 screens with it from 36".
Like Most new things LEDS still need to be researched, you can get more technical info with incandescent lamps.     Watts, Kelvin etc, I have not been able to find a LED catalogue with similar info.
I am sure that as they become more common prices will drop.

[Full-SpectrumSeparator]

---Formula for the Invention: LED + LCD = CTS 2.0   see below...


This seems like a great place to share my contributions to this evolving technology.

How much do you think those panels can be made for?  I want to get one right away for testing.

Also trying to get the point-source bulbs as well, for my purposes the power of 1680 watts from those 336 5-watt bulbs might work or not, if each area is only 5 watts whereas I need something like 400watts to pass the filters which I can't really work around.   

I am also making the project open-source DIY, etc... and I hope that all the manufacturers would adopt the technology and push for actual changes in the panels made so that we have true UV passable filters in the right ranges, and it would even triple the resolution instead of the RGB.   

However the proof of concept was tested last year and I'm just working on trying to see the low to high-end ranges of what I can get... but the timing needs to come down to minutes.... even with these high-power point-source bulbs it would change the 10-seconds to minutes I am sure.   But I need to test and show it.     

The great part is the cost is entry-level.... just paying for the bulb and driver, or something like that panel you've shown is perfect if it actually works in my "invention".    It is not really my invention or able to be patented, but I honestly think it is just too cool to not share and show, maybe if people want a solid prototype with some of the ideas I have to make it robust and worthy of high-volume production and such... it would become a dream-machine rip-station stencil-maker... but these combinations of technology are being used in a lot of ways,  as there are people using these techniques for 3D printing, etc.   

I first thought of this back in 2005 when I first entered screenprinting from the digital print world and a couple years out of college, and seeing the rather archaic and analog process of vellum and toner, inkjets and film, etc... to make masks, to put on light-tables, to expose screens... compared to what I grew up with and had known in the digital copy and poster, color laser and inkjet world, etc... (imagine if I had seen Rubylith!) - well I was thinking one day, why don't they just use an LCD screen?

10 years later I finally tried it and it works.

So what I am going to make some DIY videos and start showing the testing and developments of, is a "Digital CTS LCD Stencil-Mask" technology or "hack".     With current technologies, this is certainly the next step forward because of the merging of UV LED costs and high-definition LCD costs.

This is basically what you'd call the "poor man's DMD CST" machine.... you know the CST machines that use a plotting head of direct UV light passing through a DMD projector to expose only the negative areas of the design, no mask required...   Well I did research those over the years and they are still outrageous in price, especially compared to this.   

Now these are very crude examples I have attached, but this is what I have to work with... and only a proof-of-concept test, and with a regular 1080p LCD panel,  just the block of UV flashlights you see, and well... nothing else but the emulsion, and like 30 minutes to get the light strong enough to pass through and get a cure on both sides of the screen mesh, while the stencil still blocked the uv.

All I need to test is these higher-power UV LED bulbs single-point and something like the panel in the above post.     Showing the DIY way of putting the LCD on it is a piece of cake, and fun, but gotta be careful... and thats why the reason for open-source and let's build the best digital table for bringing screenprint into the true CTS arena...   for example it should be inside a housing where the screens slide in and press up to meet the LCD panel, the light from above, and a computer-system attached with sep and rip capabilities to run the entire process, memory-card readers and wifi, scanner and camera, etc... a whole stencil-making system is what I envision...  but I have neither the resources nor the profit-motives to keep this under wraps until I can build the perfect complete system, and even then each part would add costs to the whole unit, more for professional shops...  but the everyday printers by the thousands who are just running small one-man shops etc, everyone can benefit from this new technique,  at least we need to test the variables of what we can get and can't get....   

However, even with a 1080p screen, and sizes you can get a square-ish basic 14" x 12" for like $40, older maybe not the same res,  but a modern HD monitor is only about $150,  and the 4K monitors are like coming down to the $400 range...   I am getting about 100 DPI resolution exposure on the 1080 p monitor if it is a smaller size like   24" perhaps... I have to check the maths again on the diagonals to the sides and the actual resolutions, plus the RGB pixels split by the LCDs and the grids around the LCDs play a factor (hence the need for very high-power UV to blast through both the polarizing filters which cut out UV and the RGB which most UV passes the blue LCD) - however the (closed / off / no light) mode of the pixels cuts out the light so significantly that it works, or at least you can see is proven in this test to have worked with those low-power lights at a longer exposure of about 15 minutes perhaps.... with the flash-lights directly behind the LCD pixels (hence only exposed the areas where the flashlights were, just a crude test proving it works)...    When going to 4K monitor panels, you get more resolution - but this depends on your actual SIZE of exposure - the monitors diagonal or height/width dimensions factored with the stated resolution 1080 or 4K or even 8K is already in manufacture, then we are talking 500DPI plus resolutions at least for 20" etc sizes... 



Now if we could just get enough of a market demand for a new type of LCD panel to be manufactured with UV LCDs not RGB, and with UV backlighting and UV-passable polarizing filters, in the right ranges -- then you TRIPLE the resolution, so getting to film-capabilities and thermal and beyond if we can get there...  where it is directly exposing the entire area of the screen at resolutions of perhaps 1500+ DPI, pixel-perfect and masked perfectly too.... the comparison to "film density" is just, I don't know we would have to make the comparison terms and figure the math for it.   

-No RIP - except for visual graphical digital image conversion and display to an LCD.
-No Film, No Ink, No Analog Conversions besides the digital conversion to the LCDs.
-Uses existing screen emulsion, mesh, etc technologies, existing light tables can be converted if high-enough-power UV lights are added or the existing lights work within reasonable exposure times. (Metal halide maybe?)

-No waiting for printing the mask on the screen in any way.   
-Directly exposes the entire area of the image negative, masks the positive digitally, and does not "plot" over the image.

-Super Low entry-level costs for spot-color down to ten or 8 point text and halftones of perhaps 30 LPI, with low-resolution older cheap monitors...
-Future evolution potentials off the charts of current screenprint technology

I will post more images of the various tests, and start making vids of how to take apart and set up the panel,  if anyone wants to work together we just gotta try a high-power UV bulb.

You won't believe how easy it is to make this work.    I am just not in any positions of having resources for taking the R&D on this project to the next level... but its open now and I don't think it can even be patented because there are already patents on using LCD masking for various scientific processes and even in 3d printing, but we can certainly DIY and use the fact that the LCD monitor industry is booming with cheap LCD panels... now the LED market is about to boom with cheap UV LED price-points for the bulbs and drivers.      Everything else is a matter of how you want the protection of the panel and the cooling of the bulbs, the vacuum to the mask, etc...   I know most of that can be figured out through some of the things mentioned.

Things are really getting fun now.... and also by the way the photoshop toolkit plugin I made is also officially open-source and free to the industry for any and all uses, no limitations, enjoy... well I need to make some videos showing how to install it properly and get it running for certain versions of photoshop... but you can download and even just run the .jsx scripts without the panel if you know how, feel free to contact me if you want me to do a 1-on-1 training or installation. 

Now, any questions about the digital LCD-LED-CTS method?   Let's answer them together, it's open-source DIY, it is not even that special, people have been doing this for 3d printing already, if people want to build screenprint exposure tables and put this tech in them I'm sure it will work fine, hopefully LCD and for screenprint etc they aren't violating existing patents, but if so just DIY it, because it is really a piece of cake, not really any modification to the panel just disassembly and re-mount differently.   

[sben763]

That's far from a single source/point.  After over 2 years of development and research it's not as simple as just using a led in the 395nm as you say your going to use. I've bought 3w/5w 100w all the way to 500w 385-405nm. Yes they will work but not as good as some of the others. If you spend the time and do some scientific research comparing the cross linking, under cutting and over all detail retention vs the other units you will find as I did it won't compare. Its why I developed my own bulb. It has taken over a year of testing and tweaking to get a product that wasn't just haphazardly thrown together like some of the current units that used their customers as test subjects and then upgrade their units based upon customer results. 

Saati's exposure unit is 450nm.  Its oil cooled which does cut down on the UV to the screen. Anything over 300 on a single die almost requires liquid cooling.  The have a lower wattage bulb.  I personally think there a few major mistakes on these units. I was able to use one for some testing at a friends shop.  It's better then the Vastex unit as it was when released which is likely the results one would get with a panel like posted. 

[Full-SpectrumSeparator]

That's far from a single source/point.  After over 2 years of development and research it's not as simple as just using a led in the 395nm as you say your going to use. I've bought 3w/5w 100w all the way to 500w 385-405nm. Yes they will work but not as good as some of the others. If you spend the time and do some scientific research comparing the cross linking, under cutting and over all detail retention vs the other units you will find as I did it won't compare. Its why I developed my own bulb. It has taken over a year of testing and tweaking to get a product that wasn't just haphazardly thrown together like some of the current units that used their customers as test subjects and then upgrade their units based upon customer results. 

Saati's exposure unit is 450nm.  Its oil cooled which does cut down on the UV to the screen. Anything over 300 on a single die almost requires liquid cooling.  The have a lower wattage bulb.  I personally think there a few major mistakes on these units. I was able to use one for some testing at a friends shop.  It's better then the Vastex unit as it was when released which is likely the results one would get with a panel like posted.

That's why I am pretty sure that method of multiple-led panel won't really work for the LCD masking Computer-to-screen technique either...  unless it is really close and then all sorts of variables like undercutting etc need to be re-evaluated because of the LCD vector.   Single-point from the right distance should blast enough UV through each LCD pixel while the LCD(off) state should still mask the UV appropriately, and maybe we can get it down to under a minute or so,  it really all needs to be tested and I don't have a bulb to test it with, lol.   

But how many of you have bulbs out there at these power-ranges of UV like 300watt, 500watt, 1000watt etc metal halide whatever...   go pick up some old cheap LCD monitors or take one apart you have laying around, and put it on your light table, drive it with a computer attached and display a black and white image,  start testing.    The cost of testing the LCD methods are a lot less expensive than the LED testing and are literally giving you CTS capabilities, not just an LED exposure unit,  you can even try the LCD masking with existing exposure tables if the wattage will pass through the significant UV-cutting of the panel itself for a long enough time to expose and cure the emulsion, and the mask should block it to get the stencil.    Exciting stuff, but me no have high-power UV! 

[Full-SpectrumSeparator]

LCD = Liquid Crystal Display

LED = Light Emitting Diode

CTS/DTS = Computer-To-Screen / Direct-To-Screen


Just making sure you understand what is being discussed in my posts above is something I have never seen done anywhere in the screenprint industry, and I've shown it works, and only needs a higher power UV light to get the full area to expose.    If you already have a high-power UV light table like a UV Led or mercury vapor or metal halide,  its only going to cost about $40 to test and see, and you might have yourself a Computer-To-Screen digital film-less/ink-less exposure method.

How much would the entire industry save if we eliminated film, and film ink, and all analog processes of transferring the stencil to the screen and then exposing.

My or "this" method with LCD and UV exposes only the area of the screen image you want exposed, and the rest is masked out with LCD technology, so it is entirely digital except for the UV light and the screen doing the digital-to-analog conversions through the lcd panel...  There would be no waiting or time/labor for printing the stencil first like most CTS/DTS systems, and there would still be a perfect registration etc.    In the future you can even program in dynamic dot-gain compensation, so the image RIPPED to the display can change over time during the exposure,  think about that for a minute.



Please someone tell me if you have tried this or seen anything similar, or if you're willing to help test it.   

[jvanick]

In the future you can even program in dynamic dot-gain compensation, so the image RIPPED to the display can change over time during the exposure

this is a super interesting idea... and would certainly help in under cutting... with some tweaking, you could end up with perfectly sharp 'channels'...  hmm...

[ebscreen]

I've tried it and have been convinced that lcd screens will not pass UV in the spectrum or amount we need.

[Full-SpectrumSeparator]

I've tried it and have been convinced that lcd screens will not pass UV in the spectrum or amount we need.

Did you see the pictures I posted of my printed samples proving it does indeed work?

https://en.wikipedia.org/wiki/Proof_of_concept

(Engineering - In engineering and technology, a rough prototype of a new idea is often constructed as a "proof of concept". )


Did you try a powerful enough UV bulb and in the right spectrum?

I've gotten those little flashlights which are like 3 watts not even, to work just up against the panel and within about 12 minutes gets a positive cure through both sides of the emulsion, and the masking works fine....  I'm just trying to take it to the next level of a full-area coverage by getting more LEDs or a powerful single-point source to work from the right distance.

So I'm confused about your being "convinced"... of those two things:

1) "lcd screens will not pass UV" --- They do pass UV, it is proven and I have shown the proof.

2) " in the spectrum or amount we need."  --- It is in the right spectrum, and the amount works, I just need to increase area/distance with power.

The "proof of concept" is real and works, not something we need to be "convinced of" on the actual possibility side.  It's proven possible before my very eyes, and repeatable, empirically, by anyone else who wants to try it.   

The testing and increasing of the variables we need is just to get higher power or more even grid of LEDs to get the whole panel to cure, not just little store-bought flashlights - that was only proof-of-concept because normal UV bulbs like fluorescents or cheaper UV strips would not get a cure ALL The way through the emulsion without taking upwards of 30+ minutes and still break down washing out.    But I pushed that envelope with just testing the UV flashlights directly behind the panel right up to the emulsion.  Perhaps you didn't try to push the lights closer to actually get positive UV power and spectral passage like I have shown.

But we are beyond the "proof of concept" stage already that is why I'm sharing it openly to get the testing and other more production and quality-oriented factors worked out. 

There are major concerns when it comes to everyday use such as protecting the panel so you never even can touch it or put a screen near it, to not crack or damage pixels etc,  and getting larger dimensions like a 27" monitor would cover the inside area of a 22" x 24" screen which is about 21" x 19" I think, inside the frame, and going with 4K monitors we will have about 300 DPI resolution -- probably not good for really high LPI work, but perfect for index and also people seem to forget that you lose grey-levels of halftone percentage when you increase LPI, you don't get more shades, you lose shades while increasing spatial resolution, but if you don't go to 600, 1200 dpi in film, and even then going to 50, 60, 80 lpi just kills the amount of actual % levels you can reproduce from the digital halftone conversion.    For most of the screenprinters out there even just showing a simple DIY method of taking a high-power UV and a cheap monitor that will give you even 100DPI will hold like 30 lpi halftones and 8 point text or lower,  and that will be perfect for most startups and small spot-color, simple-design type of printers.     

It only takes a bit more investment in physical aspects and designs and R&D to build this to a higher-end professional level, with vacuum and even computer sep/rip-station built in to drive your "RIP" - just displaying your separations to the monitor and blast the UV when ready to expose the screen -- a housing that slides the screen into place and presses gently with protection for the panel below thick glass for backing support, and the vacuum seals the mesh to the panel against the thick glass, maybe the light from above, all housed inside so you can't mess with it and accidentally damage the pixels/panel with a screen-frame edge etc...   been thinking on a lot of all that stuff over the last year, but really its time to just show we can at least get to the next step which is a full-area exposure, in lets say under 5 minutes would be a good start.   Deal with seeing what kind of detail and spot/text/halftones we can get, the resolution etc, after just putting together a higher-power bulb or the grid of high-power LEDs  - either placed close or far enough to still work, and the timing is good, and then try going further from there to dial in the quality desired.

I am still wondering if you merely gave up after one or two trial-and-errors on the concept, or what could have "convinced" you that something won't work which I've proven works.   :/

[ebscreen]

Holy hell I didn't mean to open that can of worms. I'll try and find the page but I've seen it stated that
lcd screens simply can not be made to pass uv in the quantity and spectrum that we need. Yes, I obviously
see your photos. It's a looooonnngggg way from what you have to what we need.

DLP probably more viable.

[ZooCity]

Love this idea fullspectrum, I came up with a concept similar a couple years back when I saw that transparent LCD screens were being manufactured and a light flickered on in my head.

My idea was just like yours but utilizing an optically clear, thinnest possible LCD.  Drop in replacement for all expo units and all light sources.   Could also be built as standalone units optimized with other features like that very clever dynamic dot gain idea you had. 

The key would be the thickness of the panel to avoid excessive light scatter and undercutting.  Otherwise, while it would still be viable, it would be regaled to the simplest spot color images shot onto fast burning PP emulsion.

So when are we going to have nano bots form a stencil for us across an electrically charged, high tpi screen matrix? 

Zanegun, the custom Saati setup looks super efficient next to the CTS, nice work.  If that was true single point and higher wattage it would easily be the best way to go.

[zanegun08]

This is the light that is in the "Screen Toaster"



I don't think they have it online, but they have it in their print brochure.

[Full-SpectrumSeparator]

No worries,  its not a can of worms, hah.

I'm open to all exploration of the pro's and con's and physical quality limitations or pathways to all of this tech going forward.

I just know what I've seen and actually printed, and yes those are crude examples but it was only proof-of-concept.

Thanks for sharing the Saati brochure I need to buy that light ASAP and test, show you guys it is easy to get a full-screen exposure in probably under 10 or 5 minutes, with details down to 30 or 40LPI halftone dots, or essentially a 100 DPI, until I test a 4K screen - at about 300 DPI.

The LCD panel is right against the emulsion/mesh, so there for  undercutting or gap, only perhaps the thickness of one layer of the polarizing film, sure that needs to be compared....  but its not relegated to spot-colors only, I have already gotten halftones at 30 lpi probably down to 10% dots, and 8 point text.

No contention or argument going on here, this is just open discussion sharing what I "think" I've done, results of tests etc, all in the spirit of progress,  I understand the images of the first proof-of-concept tests and prints are very crude, and a long way away from what we need to be useful for even a 1-color spot-print at normal print dimensions and in normal exposure times.

But we'll have plenty of time to approach each factor and test it, I need to save up and buy some bulbs and test some methods, but there are some extremely high-power UV out there, some are just outrageously expensive panels,  but I know through small flashlight tests it is not as impossible as it may have seemed in the past... perhaps UV LED technology has come a long way and maybe I won't get it to work from a distance, but I don't think in the past without UV LED there was a way to test lots of high-power UV bulbs up close to the LCD panel? 

But I mean I could just build my own series of closer-packed UV LEDS -- I tried buying the strips but those are way too low of an output power, took like 15 minutes just to get a small dot of exposure...  which is to be expected from such low-power output.      Obviously the amount of UV cutting through the polarizing filters and the RGB LCD crystals is SIGNIFICANT--- I understand this, it is cutting a TON of UV when I can put UV LEDs right up to the emulsion through the panel and it takes 12 minutes to get a cure. 

But I get a cure.    The emulsion is cured, the stencil part that was blocked by the arrangement of the LCDs being charged cuts out so much MORE of the light that it works.    The only thing to solve and I think that is why this is the right timing for this transitional tech (- the 3d-printed screens and the nano-micro-electrical-machine-controlled digital screens are of course way way off into the future, additive manufacture will eventually become material replication like star-trek, the field of printing is actually the field of reproducing anything from an original pattern, etc... we are just in primitive days but emulsion-mesh-screenprinting is like making small prototype 3d prints held by mesh, and 3d printers are using LCD PANELS WITH UV THROUGH THEM for curing layer-by-layer of emulsion into full 3d objects.... I think it works for screenprinting just fine if we just work at it) ---  but the transitional nature of this tech may work because of a few factors which aren't really about the tech...

-- for example I also thought about and was going to buy the transparent-LCD types of panels.. but this is specialty and very expensive, and when opening up a normal LCD panel, it is already TRANSPARENT.    the LCDs do not "light up".     There is either a simple CFL or a modern LED backlight, at the sides and it passes through various diffusers to create a very bright white smooth even backlight behind all the LCDs. -- we are just taking apart everything and only using the front LCD panel itself...   RGB liquid crystals sandwhiched between light polarizing films in grids with small electrodes that fire charges which cause the crystals to become jumbled up and when they are not arranged correctly the light does not turn to the correct polarizing angle and it doesnt pass the filter.,... normally the light passes through because the construction is that the polarizing filters are at right angles, and the light would NOT pass if you didn't have the RGB crystal layers where within each, like the Red crystals, are arranged to turn the light 90 degrees, and therefore allow it to get through the other filter...  There is definitely a reduction in light-passage in visible and UV etc wavelengths, but it is not absolute or 100% reduced...  holding up the disassembled panel you will see its like sunglasses dark, hence the normal white backlight is very bright, but thats why we will just use our own UV HIGH POWER backlights.     Haven't you ever realized you gotta hit the thing with a bigger hammer if you want it to work?    The right tools for the job and we'll see if taking it to production-levels is possible,  I mean the first test for me in my small arena is to get a system going that I will replace the current older fluorescent-bulb vastex unit we use.     Obviously that will require more of the other parts like vacuum top etc, and protecting the panel in ways so I dont damage it.... and just hooking up at computer to be the rip-station to display the "seps" on. 

But anyway, the transparent LCD panels are too much of a niche and expensive,  this transitional tech might work now because of the cross-over of things like higher and higher resolution LCD panels at affordable costs to both test and build our own (rather than waiting for some big company to make another high-ticket-price item... this brings digital CTS to entry-level and still has high-end potential) --- and also the LEDs at higher and higher powers and closer tightly packed configurations... of course if we just had LEDs packed as tight as 1200 dpi then thats it, right up to the emulsion for like 1 second lol, done.     But that is all requiring too much advanced prototying and manufacture,  I know enough about how profit-motives and business-models and manufacturing/industry works to know that screenprinting will not get these huge LCD panel manufacturers and science/engineers to create UV -liquid-crystal-displays.   You can't even get monochrome LCD larger than like 10 inches.   But you don't need to,  the RGB LCD panels are transparent when you just use the panel, and transparent enough in the UV range with the right power to make a CURED STENCIL.    Yes, the only step next to at least prove a print-job-usable stencil  for a spot-color job or with halftones in a compressed range of possible levels, would be how do we expose the full-area with higher-power, either a point-source from the right distance or a lot more higher-power LEDS packed closer in a grid and from a closer distance, but too close and you get just the spot-light-areas of the exposure and wherever there is no LED light it wont cure...  but these are simple steps to test next with various current LED tech.

Thank you so much for the discussions and what everyone is sharing,  also the Saati brochure I gotta get that to test right away, or Sean as well I think may have some bulb/driver units ready soon.      But who knows maybe I will need 2 or 3 of them arranged or closer/trying to force a higher power spot-light, the cool part is that the LCD pixels actually force the light to arrange straight forward, so you're kind of like actually exposing pixels, not having light pass through at angles that undercut...     I am working on the level now of writing formulas and direct image processing code,  I can write RIPs very easily, halftones etc, the fun stuff is developing cluster-detection and other post-RIP halftone dynamic % range controlling where you could then have certain ranges expand or shrink over the exposure time to have like a dot-compensation curve dialing in realtime.. .but that is the stuff thats down the line after building a true RIP-station for the industry,  as other industries have like offset CMYK that I worked in, computer-to-plate has been around a lot longer than screen technology computer-to-screens, and they've had rip-stations running halftone previews etc for decades,  we could just combine with separation and other features.     I see the screenroom being the center of the art-to-print transitions in the future, and literally stencil-making systems that are as easy to make screens for printing as using a copy-machine for making color copies.   I've worked in these various industries and it is just a matter of time, not a matter of technology capability.    I mean really with just a few of these things all cobbled together you'd have a system like that...   the ease of doing separations and rip/screens for any type of job or # of screens just went off the charts for me, after a few things which I'm still updating in the plugin and going to show, which is all easily accessed through photoshop's index mode, and you can trick index into halftones so easily,  so combined on a computer with the setup to 2 monitors - one to the normal viewing and the other to the digital RIP LCD panel display,   then you literally press a few buttons check out a few options to get the type of seps and # of screens you want to sep to/colors/halftone type etc/,  then display your RIP image (its already there when you view each sep as black and white) - to the other monitor panel,, load your screen, and hit the UV expose, done, wash it out etc.      Let me correct myself... I see this being done in any area you want it,  but I see there being a possibility or potential that the art from any image to separation/rip/screenmaking can take place even at a single machine in the screen room.     It could become machines that are pre-loaded with dried-coated screens and people go to a kinko's and put their art on the glass or plug in their cards or upload wifi images form their phones and get screens made to go print with... just as they go to get their films printed to make screens with.   

The other place where this is really cool to test out is in the Lum InkoDye or other UV-curing inks and elements,  it is already being used to make DIY 3d printing systems using liquid emulsions - resins,  as screenprinters we have been working with UV emulsion-resin tech for a long time and making what are essentially complex 3d prototypes with each screen.

Looking at 3d printed screens it is way off because of many factors... but really it is interesting study on materials science to consider...  just how strong a high-tension mesh of nylon threads stretched and that are woven together, truly is... you can stand on it, etc,  and compare that to a 3d printer tying to make a mesh, lol.    But its possible to still make a stencil that might let ink through,  there are newer flexible and nylon materials, etc,  you 3d print the layer of the design, then the mesh itself too... but it also would have to change a lot of factors...  using a larger x-y bed of liquid the size of your design or screen... then it cures a thick layer once for the design, and then another thick layer for the mesh...  you'd have a stencil you can print ink through...  but think about that difference between nylon threads that are woven and stretched, and an emulsion just by itself or even nylon extrusion printing a simulation of woven... would it be able to get stretched after?    I doubt it is anywhere near the same materials variables...   So 3d stencils are a long ways off, expensive to test etc etc.

All I'm saying is... this takes $30 to test if you already have high-power UV.   Maybe a few hundred to test if getting these high-power bulbs and panels.    But I think some of the companies may already know all about this tech, did you guys think maybe thats why Ryonet has a table pc attached with bluetooth?     One would hope they are ahead of anything us little guys think of, but who knows where their research and developments are going...   I'm just saying I am not about all that business-stuff, I want to push forward in science and discovery, trial and error, and sharing openly as I go.    But of course anyone in this thread could decide they are gonna go test it and then not say anything and develop their own table to sell etc, that is the purpose,  anyone is free to do what they want with the information.    I'm just excited to keep testing and working out the issues, see what's possible or usable for my own shop or the place I work etc, and hopefully there will be some others who are excited too and have more resources than I do to take the testing and developments further, but take it or leave it, please don't take me the wrong way,  I'm not trying to open any more cans of worms anywhere,  its a can of liquid crystals and light emitting diodes with ultraviolet photons cross-linking polymer chains. 

[Sbrem]

It's getting pretty busy, and I can't spend as much time on this, but this is great, my head is totally swimming. The folks that have posted using the Saati light have them in enclosed systems from what I remember I've seen; has anyone tried one on a stand, to be used just like a MH facing an exposure frame, without it being enclosed?

Steve

[jvanick]

It's getting pretty busy, and I can't spend as much time on this, but this is great, my head is totally swimming. The folks that have posted using the Saati light have them in enclosed systems from what I remember I've seen; has anyone tried one on a stand, to be used just like a MH facing an exposure frame, without it being enclosed?

Steve

Greg Kitson has had one in his shop for a while now.. I think it's the older 3 LED style one tho, and not the new 450w with reflector setup.

[Prōdigium]

An exceptionally interesting concept, and accepted proof of concept. The issue will remain that most, if not all LED/LCD screens will simply NOT produce enough power (per pixel) unless specifically manufactured for this process. An expensive endevour to say the least. This would especially be true for trying to expose a thick cap film, but could have application in the printed circuit boards field?

The Saati unit is at 450nm, seems a bit high on the scale for most emulsions. As I really do not know much about the unit I would say that the spectrum is more intended for specific applications, CTS/DTS which often use very specific emulsions. From the friends LCD factory here in China they produce over a dozen chips with spectrum breaks every 5nm, so I can get anything I want from 365 and up.

I will leave on one interesting fact, and that is the UV spectrum only goes to 400.  The UV region covers the wavelength range 100-400 nm and is divided into three bands: UVA (315-400 nm)    UVB (280-315 nm)    UVC (100-280 nm). Once you break into anything above 430 your in the realm of human visible white light.

The LED Chip below would produce a better spectrum for most "standard" emulsions on the market.

This is a single chip UV LED testing at only 20 watts, but is rated for 350 watts @ 395nm
It gets hot, but does NOT require liquid cooling only a heatsink and a fan similar to a CPU fan, but from Saati's selling point liquid cooling sounds (marketing) better and is probably more efficient? I can group 4 of these on a single plate and offer 1400 watts of single UV power.