Before You Went LED, What Were You Using?

[dirkdiggler]

I was told by a CCI rep that they have done a scientific study that proves that GREEN emulsion makes better stencils and holds better detail!

[Orion]

Let's look at 2 things:

How many milli watts of light on the squeegee side during exposure?  This is the area that will break down first.  So what lamp or light is going to have the most power to reach this side of the emulsion?  5k/8k MH or any other lamp of your chosing?  No doubt in my mind and in my observations, the MH 5k and 8k systems expose the squeegee side better.  LED and fluo tubes use shorter distance from emulsion to light to overcome this.  But if you give the exposure enough time to expose the inside of the screen you can wind up with under cutting of details.  Getting an image on the screen is easy.  Getting it fully exposed is another issue and can vary by manufacturer with LED units.  Plastisol?  no problems, Discharge and HSA inks?  Takes a bit of fine tuning of the coating method, opacity of film/image, pushing the time as much as possible and using my emulsion of course.  I sell it cause I have used it with great success in my business.  Just the facts, try it, you'll like it, or I'll help you find the magic if you need it.

Second what is the spectral output of the light?  MH = multi spectral, has all possible wavelengths from white light to 405/420 UV.  Emulsions like multi-spectral light with lots of amplitude.  So a multi spectral LED?  usually this is spill over wavelengths, or spikes in just a few wavelengths, or a single spike with some wavelengths to either side, but so far it isn't the amplitude or multi spectral output that MH provides and emulsions like for complete exposure.  It's not like the sensitizer has changed, it's that the new light has changed and causing failure.  6-9 second exposures?  These are images, for plastisol only.   What I have found is that your old coating techiques can be part of the problem.  While you may have coated 2:2 or 2:3 and achieved good exposure on your MH unit, the same coating may be too thick for the weak LED or fluo light to penetrate the entire emulsion thickness.  The inside is soft.  Any post exposure or hardening is a band aid on a weak exposure.  I have found that coating less can help get stronger exposures.  Try a 1:2 sharp on these systems.  Should still yield 7-10% EOM.    DTS with LED uses stronger light, Film and LED uses a weaker light on the same system since the film and glass cut down the amplitude of the light.  I am still in the Metal Halide camp, especially for large shops.  LED's have significant through put advantages.  Just need to fine tune the process of coating, time.  In most LED exposures the print side is exposed well, the inside can be soft.  Strong film or CTS opaque imagery helps achieve longer times on LED to expose the emulsion.  Try using a little less emulsion, it seems to expose better than 15-20% Eoms.

Bake the Cake, no baker has ever sold bread under cooked.  So why is obtaining the fastest 'cook time' of emulsion so important and like a screen printing merit badge?  I can expose in 9, me? 8? another 6!!! to the point that we hit it with a flash bulb and get a strong screen?  Not possible.  Cook your emulsions up to the point of of overexposure.  It's not how fast you cook it, it's how well you cook it that will give the best results on press, with less breakdown, pinholes and also allows SBQ emulsions to reclaim easier and avoid lock in.  (And for these reasons an MH unit is like a commercial bakers oven and not a small household oven cooking for an entire bakery). Short cut the light energy with lamp choice or time and your exposure will be weaker.  Give emulsion the energy it needs for the right amount of time and the screen will be very durable.

Ta-da!!! 

[ABuffington]

The color of emulsion?  The resolution of our emulsions is built into the base components, we add the color for contrast.  Green, Red and Blue are chosen since they have the least halation issues.  Our highest resolving emulsion is actually a very transparent aqua, but it's nearest similar emulsion is red, almost identical resolution, identical components, just contrast needed for hand vs auto registration. 

Getting back to the thread:  I just had a chat with Dan.  The key point here is this:  LED has a narrow latitude, you need to control EOM more accurately, control the temp and humidity of your room better.  (A cold day with a short exposure on and LED will be different than a hot day short exposure, the molecules and senstizer work better on a warm day than a cold day.)  My Phoenix AZ printers can expose the same SBQ far faster than I can in LA due to low humidity and warmer rooms.  Exposure is completely printer dependent.  What works on an MH Tri Light in a shop by the beach can be longer than what works inland by only 20-30 miles due to humidity drop.  Age of the lamp? What voltage 208 or true 240?, distance from lamp, eom?, so many variables.  The same applies to LED in terms of shop conditions, and ambient conditions.

The recipe for LED is narrow.  Having breakdown issues?  Hit a hard 7 on less coats to see if that helps.  Thinner is better with weak light.  MH has lots of things LED doesn't have.  A pinpoint light source, with engineered reflectors for more collimated light.  LED's as we currently know them are chaotic light.  The rays are hardly parallel. Sometimes going for maximum exposure under cuts the image, so we find recipes that work, and find a balance.  (By the way I am old, like I started with carbon arcs and the sun.  I am a big fan of the sun, 93 million miles away, 23k or more lamp strength, very parallel rays, almost too fast, but I digress.) 

Most of us aren't printing over 20k print runs, so fine tune the process for your printing.  Optimize your coating.  How?  Coat 1:1 sharp.  Then coat a second coat 2/3 of the way up on the inside, then another 1/3 of the screen length. This will yield 1:1, 1:2, 1:3 coating on the same screen.  Attach a very detailed image suitable for the mesh count used. Use lots of tape to keep in place, or if you have DTG you are way ahead of the curve.

Now if you can block out the bottom of the screen, block out 2/3 of the screen so that 1/3 of the screen can be exposed in the 1:1, 1:2 and 1:3 coatings. A doubled up black trash bag works fairly well, or the hard to find ruby or amberlith.  Expose the visible 1/3 of the 3 coating types with 25% of the current time for this particular mesh.  Move block out medium  over to leave 2/3 of screen to be exposed.  Expose for 25% of current time again.  Then remove the block out material and expose the entire screen with current time for the mesh used.

This will yield 9 different variations on one screen
1:1 Coating
Current time, Current time + 25%, Current time +50%

1:2 Coating
Current time, Current time + 25%, Current time +50%

1:3 Coating
Current time, Current time + 25%, Current time +50%

This will help you determine where the squeegee side has been exposed properly, where details are holding and where the details get lost due to too much emulsion thickness.

PM me if you want more info.  This is a variation on a step test, but adding the different coats on the squeegee side helps see what coating works for you.  Thinner has been better in most LED tests I have done.  SP-1400 coated 2:3 dull has now where near the strength as 1:2 sharp.  Thickness of emulsion helps, as long as the lamp used can expose it completely.  MH is a little better at thicker coatings, but an LED is so fast and does produce good screens for all textile printing, it just requires a more careful recipe.

By the way I have a large printer who coats Aquasol HV once on the squeegee side.  Thats it.  Running 20K of discharge quite often.  I don't recommend it, but they get great results and are a premier contract printer.

Al

[Dottonedan]

Real good stuff Al,


Couple things.


1, I agree 99%.  the 1% is the word "weak".  LED is not weak.  Maybe in strength when comparing to a 5k MH or higher it is less than, but not weak at all. Strong enough for the majority. Weak is a bad word. lol.


2, The conversation does not separate or differentiate when using the term LED. Like it's been mentioned before, Not all LED's are the same.


3,  Much of what you are pointing (where LED lacks in total cross linking) to make a stencil durable enough... is pertaining to very high numbers where the vast majority of shops don't live. 10, 20 and 50k orders.


Another overlooked factor is (when doing those orders of 10, 20 and 50k  is
A, you can afford to stop the press for the 15 min and change out the old and in with a new screen.
B, These orders don't happen every day for the vast majority so it's not a common problem when it is a problem.


Just some extra notes.

[Orion]

Dan, I think Buffington was speaking in the terms of "frequency amplitude" when he used the term weak.

[TCT]

Every time Al lays down some knowledge like that a little part of me dies.... I realize I've "been in the business" for 18 years and don't know jack sh!t

[Dottonedan]

Dan, I think Buffington was speaking in the terms of "frequency amplitude" when he used the term weak.

That maybe so. I realize he's trying to make things more clear. It is clear more most of us. If using the term in multiple areas of a long thread (I like on long threads by the way) then the term weak in conjunction with LED may unintentionally miss guide the lesser minded people to assume an LED system is a weak light source. Not so.  Sure, Al knows that and many of us know he's not saying that, but for some people, they will be running off quoting Al as saying (after his own research" Al is quoted as calling the LED systems a weak light source. We've seen it happen right here on this very forum. Weak, shouldn't even be used in the same sentence to discuss the LED.  Less than something else maybe in different scenarios, but not weak.  That's just my preference and it's nothing really related to Al himself. Al is invited to say what ever, when ever he wants and I'm not debating if he's wrong or not. Just that this specific terminology can leave room for error. No ham intended to anyone.


TCT, LOL.  Yes, These are the post that I crave tho. When Al, Greg, Richard , Rick Roth or Coudrey comes in, I get a little giddy cus I know they are gonna say something juicy.

[GraphicDisorder]

If we all chipped in our hourly rate spent reading/typing on this thread we could have gone together and have bought Alan another brand LED to try. Haha

[Itsa Little CrOoked]

Every time Al lays down some knowledge like that a little part of me dies.... I realize I've "been in the business" for 18 years and don't know jack sh!t

^^^^^^THIS^^^^^^



BUT!... If there weren't anything left to learn, you'd be bored. Burned out even.

[blue moon]

here's my current (05/2015) thinking on this subject. It is constantly changing as more information is revealed and it is pretty significantly different from what it was just 6 or so months ago.

400nm light is the same no matter how it's produced. What am I trying to say? LED or MH or Arc light, it does not matter. When the photon is moving at 400nm it does not know what excited it and got it moving. As long as it is at 400nm, it has EXACTLY the same properties regardless of how it started. This in turn means that the unit that reads 1mW of power is the same 1mW regardless of where it started, MH or LED.
OK, going on. . . so if an MH light produces 5KW and by the inverse square low it is showing 1mW when it hits the stencil, it is EXACTLY the same as the 450W LED that is closer and reads the same 1mW at the stencil. Same wavelength, same power. They will both penetrate and crosslink EXACTLY the same.
So manufacturers use the weaker LEDs and place them closer to the stencil. This generates the same results as the more powerful, further away source FOR THAT PARTICULAR WAVELENGTH!

Now that we have that out of the way, so what is causing the difference? Well, what is different with the lights? MH is a multispectrum light and thus produces multiple frequencies whereas LEDs produce only one at a time. The wavelength produced by the LED might not be the OPTIMAL frequency for crosslinking. MH is a shotgun approach and it will crosslink regardless of the emulsion. For example, 400nm LED hitting an emulsion that exposes at 350nm can stay on the whole day and it will not crosslink. MH, while it might not produce much at that frequency it produces some, so it will eventually expose. Again, think shotgun vs sniper. MH will eventually get it done, while if the sniper aim is off if it will miss no matter how many times you fire.

So if your emulsion is not in tune with your exposure wavelength, there will be problems with achieving complete exposure! This might manifest itself in very long exposure times as the sensitivity at particular wavelength might be marginal or in some cases it will not expose at all.

There are additional factors that come into play here, but the correct matching will be of the utmost importance.

pierre

[ericheartsu]

Again, think shotgun vs sniper.

this about sums up this thread for me

[sqslabs]

So forgive me if this is a stupid question, but wouldn't an easy solution be to create groups of 3-4 LEDs, each producing a different wavelength, and lay those out in strips?  Would this not emulate the multi-spectral properties of a MH?

[blue moon]

so physics aside, lets look at the reality of our world. . .

manufacturers are placing the source closer to the stencil to get more power delivered. This in turn has potential to cause undercutting. Having proper spacing and lenses on the LEDs is part of proper design. There are compromises that have to be made in any design. This is what differentiates the units (where the compromises were made).  Design will impact the field strength, uniformity, wavelength and so on. . .

On the emulsion side, the actual wavelength is in some cases not set in stone (as I understand). When an organic compound is part of the emulsion, it is not exactly the same every time (since it is not produced in laboratory conditions, but rather by mother nature. Think oil, middle eastern crude is very different from the stuff pulled up in Texas, but they are both oil.) This can introduce variations we normally don't expect to see. MH with it's multispectrun approach should cope with it better. Further more, crosslinked emulsion has different light blocking properties than the stuff that has not be crosslinked so it will impact the penetrating power. This means that the wavelength with better penetrating properties MIGHT be more suitable for exposure. Now, we can not switch to something that penetrates without crosslinking, so again we are faced with selecting the right wavelength for the LEDs to get both the penetrating and exposing in balance.

more to come if I can find time. . .

pierre

[blue moon]

So forgive me if this is a stupid question, but wouldn't an easy solution be to create groups of 3-4 LEDs, each producing a different wavelength, and lay those out in strips?  Would this not emulate the multi-spectral properties of a MH?

it would, but at the expense of the total power at required frequency. If you split it across four different wavelengths, the usable power could be as low as one quarter of the rated power.

pierre

[Screen Dan]

So forgive me if this is a stupid question, but wouldn't an easy solution be to create groups of 3-4 LEDs, each producing a different wavelength, and lay those out in strips?  Would this not emulate the multi-spectral properties of a MH?

it would, but at the expense of the total power at required frequency. If you split it across four different wavelengths, the usable power could be as low as one quarter of the rated power.

pierre

...Unless we get the screen 4 times closer.