Decided to take it up....Making an LED Exposure unit.

[Prōdigium]

Love threads like this and interested to see what you come up with on your project Prodigium.  Good luck. 

I as well, and I am very excited about my project obviously. Screen printing has traditionally been an affordable business opportunity to get into and over my prior 20+ years experience I have seen lots of entry level products that sadly not only fail to produce a quality product but are so poorly made they actually make the process harder for a newbie. My product is meant to alter reality in that it will by design simplify the process while at the same time producing a quality screen no matter your needs. Because the actual costs for LED units are so low, in reality there is no reason for some of the prices out there.

I don't begrudge anyone making a buck, heck its my intentions as well..that's why we all are in business but its like being charges $50 for a grilled cheese sandwich. However I will fully acknowledge that when buying a "name" product your also buying support and service and that is a very valid consideration for anyone and that is part of the purchase price. But that is also why I am designing my unit to be 100% user serviceable...so long as you know which end of a screwdriver to use.

Steve, to clarify. The Cree unit I spoke of was a single UV LED with a rated output of 10 watts. Not 10 Cree units. On my mini test rig they were intense, very hot even without the matter of being enclosed in a box or even having glass over them. This was the same for many units I tested. Ultimately I opted the path as M&R, which I will give credit too....More bulbs, lower wattage = less heat & more uniform light distribution. I just do not agree on how the lights are spaced, or mounted. Its layout is counter-intuitive, strips are cheap and require some sort of methodology to resolve. The effects from heat will have more of an effect on the bonding agents used to glue it down and keep it level which will adversely effect exposure quality.  However I will concede that its very possible that in the real world its simply not that big an issue for the vast majority of customers. Sometimes lab tests lead you to resolve problems that really never really existed to begin with. But keeping the LED's on a flat plain is a real problem.

If there is any heat issues to my design it will not be caused by the LED's themselves, but will be the result of its working environment. Obviously if you in a shop where temps are high that will have some effects, even with fan cooling drawing hot ambient air over the LED's will accomplish little to nothing. However I will concede that heat can significantly shorten the useful life of the LED. Higher operating temperatures lead to higher junction temperatures,  increasing the degradation rate of the LED element, causing lower wattage output of the LED and decreased life over the long term....as explained to me by my LED mfg.

I left the technical details of this to my mfg as they fully understand how to compensate for heat since they are the mfg. I merely provided the working environmental details and design parameters and they stated to me what I had to do to best counter these effects. This is why I opted for a copper backed aluminum heat sink mounting board in fact.

But as Steve stated, and as my mfg told me that 50c is not really an issue. And once again lets be realistic here...even if you half the life of a standard LED rated at a conservative 30,000 hours your still in the DECADES time span.  A replacement LED strip in my unit will be lifetime warrantied for the original purchaser, so even IF you had to replace all the strips it would be FREE...sans shipping, and it would at most require 20 minutes to swap out. This talk about LED degradation really is much ado about nothing.

Should you find yourself in such a hot ambient shop, your bigger fears should be on the effects to your coated screens. Most emulsions will start to harden or even expose at such temps. So talk about heat may be a problem for some designs but its not one that has troubled me.

[abchung]

In building these LED units, I believe the most important thing in preventing under cut is the beam angle. No matter how accurate we place the LEDs, it will never be in the perfect spot. So to reduce the errors, we need steeper beam angle.



The draw back of steep angle is:
1. Increase the number of LED.
or
2. Increase the distance of the glass from the glass.

[Prōdigium]

In building these LED units, I believe the most important thing in preventing under cut is the beam angle. No matter how accurate we place the LEDs, it will never be in the perfect spot. So to reduce the errors, we need steeper beam angle.



The draw back of steep angle is:
1. Increase the number of LED.
or
2. Increase the distance of the glass from the glass.

Your on the right path, but your diagram is only representative of a light that has a single angle of exposure. There are very inexpensive ways to fix that problem...all you need to do is buy a Maglight and you'll figure it out. Optics much like wearing glasses solve such problems when used properly. Its called collimation.

[jvanick]

I just do not agree on how the lights are spaced, or mounted. Its layout is counter-intuitive, strips are cheap and require some sort of methodology to resolve. The effects from heat will have more of an effect on the bonding agents used to glue it down and keep it level which will adversely effect exposure quality.  However I will concede that its very possible that in the real world its simply not that big an issue for the vast majority of customers. Sometimes lab tests lead you to resolve problems that really never really existed to begin with. But keeping the LED's on a flat plain is a real problem.

our Starlight is 2-1/2 years old now and the LED strips are still perfectly flat to the pcb.  I'm not sure what they did to mount them, or how they got them flat to begin with as usually they come on a roll, but they certainly haven't moved, bubbled or changed.

[Prōdigium]

our Starlight is 2-1/2 years old now and the LED strips are still perfectly flat to the pcb.  I'm not sure what they did to mount them, or how they got them flat to begin with as usually they come on a roll, but they certainly haven't moved, bubbled or changed.

Good to know, and to be truthful I have not heard of this matter with M$R's product. It may also just be the volume of screens your producing? If you only making 15-20 screen a day, you will most likely NEVER have a problem with the LED's.

But if you do, it would be interesting to know how difficult it would be to repair/replace them. Has anyone had to have one repaired, how long were you down and how difficult was it?

[Prōdigium]

After having some companies talk to me about possible distribution and listening to what they think, I have decided that the product dimensions will be altered somewhat from my original plans. Originally the plan was to max it at 1 - 23" x 31" but several people wanted to have it handle 2 - 20"x24" screens so were going to listen to what people want.

The unit will now be sized slightly larger to accommodate either 2 - 20"x24" / 1 - 23"x31" / 1 - 25"x36"

UPDATE: Actually I have decided to make 2 sizes, one for only 20x24 screens and the full size one above. I wanted to be sure that we can be the most competitive on pricing and for many dealers I spoke too they want them for new startup shops where 20x24 screens is the norm.

Just got in one of our testing sample boards....nice, clean and should prevent ANY issues with heat, that I have had any with my tiny test rigs, but 16 LED's is not the same as 750 or more. Now I get to practice doing some manual reflow soldering in my oven. When we get into production I am thinking that they will not be white, but when you get samples they do not offer many options.

[ZooCity]

Just to follow up we do indeed have a warped array panel.  (being replaced)

I can't attest to whether this was truly heat related or not but it's wise to plan to keep the array both cool and perfectly stable.  Even a small amount of warp or distortion to the mounting is going to mess with the collimation.  Your heat sink mounting looks solid, maybe mount it on honeycomb al?

[Gilligan]

Our frames are larger than 23x31.

[Prōdigium]

Just to follow up we do indeed have a warped array panel.  (being replaced)

I can't attest to whether this was truly heat related or not but it's wise to plan to keep the array both cool and perfectly stable.  Even a small amount of warp or distortion to the mounting is going to mess with the collimation.  Your heat sink mounting looks solid, maybe mount it on honeycomb al?

Because I am well aware of problems related to collimation it was one of the major reasons for NOT using those cheap strips. My boards are exceptionally rigid, and while there is no need to mount them to anything else for strength, they will be mounted to a rack that allows the whole LED panel to slide out like a desk drawer for easy service or cleaning. Because of careful design considerations heat as I have said is not a problem that I expect to deal with.

[Prōdigium]

Our frames are larger than 23x31.

Sorry for the unclear update. Larger screens will not be a problem. Originally my plan was to make one size that could handle up to a 23x31 screen. Then after some feedback from possible dealers it was decided that it should handle up to a 25x36. However after some extra thought on the matter and trying to keep the base model in a price range I have opted now to produce 2 models.

1) Base unit will handle 1-20x24 screen & will come in at an even lower price than originally stated.
2) Large unit will handle 2-20x24 / 1-23x31 / 1-25x36 will be priced at the higher side of my original price. (still under $2000 with crate)
3) (Still Undecided) but a model for graphics printers that will accommodate up to a 31x41 screen.

I do not want to get carried away with multiple models and building the 3rd option for graphics screens will greatly depend of my success with the first 2 models. One thing that I will add is that the case is being redesigned to accommodate a pre-press registration system which more or less involves making the top a little bigger. It will be a very affordable upgrade option.

[mimosatexas]

I would aim to be able to fit 2 23x31 or 23x33 in the largest size, with the added benefit that it will hold single larger poster/sign screens as well.

[Gilligan]

Our frames are larger than 23x31.

Sorry for the unclear update. Larger screens will not be a problem. Originally my plan was to make one size that could handle up to a 23x31 screen. Then after some feedback from possible dealers it was decided that it should handle up to a 25x36. However after some extra thought on the matter and trying to keep the base model in a price range I have opted now to produce 2 models.

1) Base unit will handle 1-20x24 screen & will come in at an even lower price than originally stated.
2) Large unit will handle 2-20x24 / 1-23x31 / 1-25x36 will be priced at the higher side of my original price. (still under $2000 with crate)
3) (Still Undecided) but a model for graphics printers that will accommodate up to a 31x41 screen.

I do not want to get carried away with multiple models and building the 3rd option for graphics screens will greatly depend of my success with the first 2 models. One thing that I will add is that the case is being redesigned to accommodate a pre-press registration system which more or less involves making the top a little bigger. It will be a very affordable upgrade option.

AWESOME!

[Prōdigium]

I would aim to be able to fit 2 23x31 or 23x33 in the largest size, with the added benefit that it will hold single larger poster/sign screens as well.

Like I said, I am not going to get carried away with multiple models. The plan was, and still is to make a powerful unit that is affordable. Most people will never need to burn 2-23x31 screens at the same time, especially with exposure times sub 2-minutes. A model to fit 31x41 would be for graphics users but as mentioned its merely an idea at this time, no commitments.

[ZooCity]

I think a large sized unit would be worth making.  31x41 or thereabouts would be ideal for us, it matches our max frame size for cts and dovetails well with max frame on presses like the S.roque XL machines and you could shoot 23x31 2up.

[Prōdigium]

I know its an old thread now, and I wish that I could report back with some pictures...but the reality of manufacturing in China is one of where if you want to make something, there is everything you can imagine. However like most things its about numbers, 1 of something = forget it, 100 = maybe and 10,000 is a "base" but still a price that makes it hard. Well that is kind of where I am at, debating on how much I want to invest in order to get the price point I want.

My LED mfg wants a (min) order of 50,000 in order to make my custom design and there is a 90 day lead time due to special tooling and fitting my "small" run into their production schedule. Whats nuts is that they can make them in just one day....its the scale of capacity here, its insane. Under 50,000 unit order and they are made in the prototype lab and the prices triple!! The budget for this is already tight and I am weighing my desire to make a professional unit right off the bat, or as I am considering making an "entry level" unit with off the shelf LED's and testing the waters.

However I am intrigued to see that Anatol has obviously been reading this thread and has even implemented my design into their own equipment. LED's surface mounted on metal boards and in a staggered position for even light distribution. Not really very surprising though, but I guess that in a way it does validate my own design concept so maybe I should be flattered?

Either way a sneak peek at Anatol's new design that completely implements what I have clearly stated here as my own design.