My LED Exposure Unit Build

[Jhtitan]

So I have been reading just about everything I can on here. The Wife and I just bought out everything we needed from a semi local shop going out of business. Only thing missing was the exposure unit as his was a loaner and had to go back to its owner, who was not interested in selling it.

Attached is a JPG of my current design. Glass bed is currently @ 43"x 30". Right now that is pretty much just to cut 30" off the end of a sheet of MDF for the bottom. I wanted it to be large enough to get 2 screens on it at once if the need arises. Not sure if that is necessary or not.

I have planned 20 UV LED strips at @39" long sitting 3" below the glass. I have read about calculating this distance but have so far found no equations or formulas. LED's have a 120° cone. I have it figured at 20M of LED Stips and I bought an extra 5M strip to make sure I have some to replace any that needed it.

I am planning on running the vacuum thru the metal frame like the one they had and a few others I have seen in pictures. Lots of drilling and some welding but nothing I can't handle. In this configuration the vac blanket gets attached to the top of the frame with a adhesive gasket seal around the the other side for the glass contact seal. I had played with the idea of building it into the box. You can see the hole in the rear corner of the box I had planned on attaching a shop vac to.

I have not started on a base design yet. I am torn between making it hinge on the wall so I can drop it out of the way when not in use or putting prepped screen storage under it. Close off and light seal 3 sides with the front being some sort of door or curtain to access ready to burn screens.

I am open to suggestions if anyone has any. I have parts on order so I am still about a week away from starting the build and I plan to order the glass last just in case there are any last minute design changes.

As far as control, I plan to use a PC power supply for the 12VDC and I have this controller ordered:
https://smile.amazon.com/gp/product/B07LBJ9PYJ/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

Guy we bought everything from did have some neoprene left from replacing the blanket in his loaner unit and he had bought enough to replace it again if he had needed to.

Once I have it built and tested I could offer up some plans for this. Would help offset some of the build cost if anyone were interested.

Thanks
Jon

[blue moon]

are you sure this is the way to go? You should be able to pick up an older Metal Halide unit for about $1K. It is a much smarter purchase than building your own.
It will have a properly calculated distance from the bulb, a good working vacuum setup and will expose any emulsion you might want to use.

There are many, many sub par and poorly performing LED units on the market and these were designed by the companies with an R&D team that supposedly know what they are doing. Don't take me wrong, it is certainly doable (Lou designed one that is now sold under Vastex name and still has the best field distribution I've seen), but is it worth it? You will have several hundreds of dollars into a DIY and for just a little more can go into a pretty nice unit.

pierre

[blue moon]

oh, and that LED layout will cause massive hot spots/variations in the light field. The distance between the LEDs has to be the same in all directions. You will need to get strips closer if the spacing on them is represented correctly.

pierre

[mk162]

I'd listen to Pierre on this one...

[Maxie]

I have a home built unit, I have a Saati led light about 40” below the glass.      It’ s not that expensive and is made for the job.      I think Saati have 2 units, I have the cheaper one.
I would also recommend that you have the rubber seal sitting on the glass, I built it with the seal on the metal frame and this gave me trouble.
I had an electronics guy set up two timers so the first one switches on the vacuum for a set time and then the lamp switches on.
Good luck, don’t try and save too much, to do good printing you need good screens.

[Frog]

I have a home built unit, I have a Saati led light about 40” below the glass.      It’ s not that expensive and is made for the job.      I think Saati have 2 units, I have the cheaper one.
I would also recommend that you have the rubber seal sitting on the glass, I built it with the seal on the metal frame and this gave me trouble.
I had an electronics guy set up two timers so the first one switches on the vacuum for a set time and then the lamp switches on.
Good luck, don’t try and save too much, to do good printing you need good screens.

Those light sources are interesting. Almost a hybrid design utilizing an array of LED's performing more like a single point light, rather than the arrays we see on most other units attampting to cover the entire area.

[Jhtitan]

I have taken my design from other builds I have seen that claimed to be getting good results. I can only go by what they have said in the various posts and such. I do appreciate your input as I have no real experiance other than some electro etching stencils I have burned and used in the past.

"

oh, and that LED layout will cause massive hot spots/variations in the light field. The distance between the LEDs has to be the same in all directions. You will need to get strips closer if the spacing on them is represented correctly.

pierre

Don't read this the wrong way, I am asking. Do you have a formula for finding hot spots or was this more of a "It looks like you will have hotspots. I tried to have as much overlap of beam as possible and at this distance with 120° spread it looks like the light from one strip should be overlapping the 5 strips on either side of it. I originally had it figured at the same spacing as the LED's are spaced on the strips, 16.67mm. That just looked like overkill. That and I am not 100% sure of the LED spacing, I am going of a print I found elsewhere on the interwebs. I am still trying to find some info on calculating that optimal distance. My current spacing is actually double the spacing on the strips. This is to allow me to easily add more if needed.

We have priced a few and while I am not against buying a new unit the used units I have found so far are all pretty rough looking, they look more "used up" than "used."

Thanks
Jon 

[Northland]

Jon.... I have and use a home-built LED exposure unit (which is mounted inside of a converted Atlas Exposure unit frame). My build specs are:
--  30 meters of 5050 SMD strip light (to create a field that is 19"x31"). Powered by (two) 120vac to 12vdc power supplies.
--  5050 SMD has a spacing of about 5/8" (on the strip), to get uniform spacing my rows are 5/8" apart.
--  the field sits 2.5" below the glass
-- it works fine with Chromalime PP emulsion (35 second exposure time, when coating 1/1 sharp edge on 180 s-mesh).
-- works great for spot color work, but has serious limitations for halftone work.

If you proceed with your plan, you may want to consider the following:
-- 20 meters probably won't get create a field large enough for exposing 2 screens
-- the controller shown has 200 watt max capability and 20 meter of 5050 strip would be 300 watts
-- don't use an LED chip smaller than 5050

I agree with the other members, who recommend purchasing a used MH unit (especially if you want to have good halftone capability).

[Sbrem]

Even though we have the Saati ProLite 450 on order, I had already started a project to build a light using a single source, a COB LED (Chip on Board). The smaller the light source the better is traditional theory. Mine is a 200w, 405 nanometer, which needs a driver to run it the proper voltage, and it needs to be mounted to a heat sink with a fan. So essentially, it's the LED, the driver, a 12v source for the fan on the heat sink, and a terminal block. I have maybe $200.00 into it. I'm having trouble locating a box to hack that will hold the parts, something around 9" x 7" x 12". When I finish, after testing, I'll share the results. I guess I'll keep it as an emergency back up lamp.

Steve

[Frog]

Read enough posts (including some here), and you will see that these units are far from being one size fits all miracle machines.  We saw many reports of folks using name brand units that just did not perform as well as their metal halide units. Many also discovered their favorite emulsions needed to be replaced with emulsions tailor-made for LED.
That said, there are plenty of happy LED adopters, and that's why there are plenty of used metal halide units out there.
So, not trying to discourage, but just sayin'. Two sides to this coin. Besides, many folks (sounds like you) enjoy DIY tinkering.

Curious as to what unit your loaner was.

[IntegrityShirts]

Do you have a verified quality LED source? That's the issue I ran into. While getting on Aliexpress and searching for the right wavelength worked ok, the quality of the LED's I received varied wildly. While it did work OK, my led board was exponentially unprepared for thick stencil exposures. I still have it sitting beside my 5k Olec so in the event that the olec goes down, I do have a backup source, but honestly a florescent tube exposure box would be just as good as a back up and cheap and would have saved me hours/days of testing. But it was FUN so I don't regret it. Often times there are much more productive ways to spend our time making MONEY, but really, money ain't the spice of life. Have fun!

[Jhtitan]

Curious as to what unit your loaner was.

It was a Lawson brand. Had 6 or 8 fluorescent bulbs. I meant to take some pictures so I could duplicate it however exhaustion induced haze had me forgetful. damn drying ovens are heavy!

If you proceed with your plan, you may want to consider the following:
-- 20 meters probably won't get create a field large enough for exposing 2 screens
-- the controller shown has 200 watt max capability and 20 meter of 5050 strip would be 300 watts
-- don't use an LED chip smaller than 5050

I agree with the other members, who recommend purchasing a used MH unit (especially if you want to have good halftone capability).

That controller will be tripping a relay to handle the load. 450W PC power supply will handle the supply.

I may just go ahead and cut it down a bit, I have most everything ordered already, and keep looking for a good used one. I do appreciate all the info!

Jon

[blue moon]

Do you have a verified quality LED source? That's the issue I ran into. While getting on Aliexpress and searching for the right wavelength worked ok, the quality of the LED's I received varied wildly. While it did work OK, my led board was exponentially unprepared for thick stencil exposures. I still have it sitting beside my 5k Olec so in the event that the olec goes down, I do have a backup source, but honestly a florescent tube exposure box would be just as good as a back up and cheap and would have saved me hours/days of testing. But it was FUN so I don't regret it. Often times there are much more productive ways to spend our time making MONEY, but really, money ain't the spice of life. Have fun!

I have heard this before from one of the major manufacturers. Samples they tested, the LEDs did not actually emit the correct wavelength and they found a lot of variance within the batches. For a home built unit running at 390 instead of 405 might not be that big of a deal, but for a manufacturer to put out  something sub par would not be good (even though many do).

If I was building one, I would measure the spectral output of the lights and then use the spectrometer to play with the distance and spacing of the LEDs. You can get pretty close, but the square layout has a physical limitation of how uniform the field can be (since the distance between the lights on the diagonal is significantly longer then the distance to the adjacent bulbs). As discussed previously, I think a triangular layout should produce the best results.

see image courtesy of Steve Harpold from Brown manufacturing. Those guys are pretty sharp when it comes to engineering (check out the FireFly dryers and their DTG software/production system).



You can see that there are hot spots in the field right above the bulbs and due to the distance on the diagonal, the intensity is lower where there is more room between the light sources.

pierre

[Pangea]

Here's my build:

http://blog.pangeaprintingco.com/

I have limited space so I built it to fit on top of my screen rack. It'll burn 2/1 200 mesh screens with Cryocoat to a solid 7 in about 20-30 seconds (I typically under expose at 20 seconds then post expose). I've been able to get pretty decent halftones as well (65 LPI on 350 mesh for die cut stickers ~3"x3"). The whole build was under $500 and I feel like I'm getting more than $500 worth of equipment out of it.

[Jhtitan]

Pangea,
Yours was the main one I was basing mine off of. I have shrunk mine down a bit to tighten up the LED spacing.

We are now talking about buying one, and saving this build for later. If I stop now I am really only out a little $$ and have a bunch of LED strip I can still use on my RC planes 

We are in the learning portion and I don't want to have a bad screen burn hamper us in the beginning. Kinda sounds like this may be one of the more important pieces of equipment.