LED Bulbs Already Failing

[DouglasGrigar]

I'll make sure that others don't make the same mistake. 

I wonder what this portends...

I'll say this again, we've managed to get detail that in the beginning we couldn't get, but if you have to throw all the cargo off the ship to get it to float it's not really a cargo ship is it?  Meaning, the screens aren't usable for anything over 100 pieces due to the need to underexpose to achieve the detail.

So the emulsion is not linking enough past the knuckles of the mesh (clearly you identify this as underexposure and I agree).

Detail resolution is always a combination of mesh choice, emulsion choice/EOM, exposure, and positives (CTS)

Multi-Bulb creates more light scatter (halation is a word you will see) the tubes, LED, and MH all have some, and they shake out in a very obvious step of this from - tubes - LED - MH.

The underexposure for detail and the subsequent problems are worrisome.

I of course am known for pushing a full and complete exposure as the starting point, making adjustments in mesh choice and emulsion to take up the rest - a no-fail policy.

Im just now finishing pressure washing out the last of the screens I have been helping my client here in the UK with all day - all with this LED unit. I use three French Emulsions here of three different types. I have not had a single fail of any screen with any of the emulsions I use with this unit. As above all by the starting point of complete linking as measured my an exposure calculator, all of them would even pass the Ulano pressure washer timed/measure abuse test and of course thousands upon thousands of trouble free impressions.

This client is a dealer in Europe and I brought this up:

Any manufacturers defects - no
Any units not preforming as the clients need... no
Any customers not happy with the speed and ability to expose all five of the basic emulsions - no

[Frog]

In fact "in the wild" the only problems I have run in to on this LED unit are all here on the boards.

That's why I was so happy to see you and Mark jump on board here again as well

[ScreenFoo]

We need to change the title here.

LED bulbs already WINNING!!!

 

[Orion]

I of course am known for pushing a full and complete exposure as the starting point, making adjustments in mesh choice and emulsion to take up the rest - a no-fail policy.

Possibly the most important bit of information for anyone following this meandering thread is the complete exposure of an emulsion prior to development. There are no shortcuts to the exposure process without compromise to other aspects or variables of said process.

[Rockers]

I know of faults in many other units on the market that he (and Vastex) are addressing in their design.

For starters they should address that the little light that indicates "exposure" really only comes on if the LED s are actually on.That would be a good place for improvement .

[ABuffington]

As with cars, we all chose different ones.  We adjust to how they 'drive'.  As an emulsion manufacturer we did not buy into LED in the beginning.  Metal Halide wasn't broke, but LED has created a new screen making dynamic that is different from Metal Halide.  Both work, but with some adjustments needed.  I now sell LED mesh!  OK that's just a marketing slogan, but what I have found is that LED likes higher mesh counts.  A 150S vs a 110T, a 225/40 Yellow vs a 160W.   Our analysis of the exposure is this.  LED's do undercut a bit, but by moving mesh counts up a little the details are still decent.  Are the halftones as good as a Metal Halide, well no, they are a bit soft on the inside, but with a fast SBQ emulsion, complete exposure, (Thanks Greg, this needs reinforcement in our industry.) and an alternative mesh count that is higher but with similar print quality, your results will be fine.  Again, like cars, a new one may require rethinking on how you drive it, but it will take you where you want to go.

I recently dialed in a 24 auto shop with 2 STEII's, auto develop and more importantly auto reclaim.  Speed?  So fast a Metal Halide user could work 7 days a week and not equal what an STEII with multiple heads and exposure can output.  What helps is post exposure on SBQ products.  Without post exposure on LED you risk break down on discharge and water base, and more importantly any screen opener or solvents used in plastisol printing will lock in the emulsion without post exposure.  Auto reclaming machines aren't very strong IMO.  But with the right emulsion, post exposure, it works beautifully.  Post Exposure sources by strength and light quality: Sun from 10-3, 8k Metal Halide, 5k Metal Halide, LED.  The reason for LED in last here?  Spectral output.  Emulsions like mulit-spectral to completely cross link. Will the LED work for Post Exposure?  Yes for the majority of our print runs.  For 90K runs? not sure yet, untested.

As Pierre mentioned, you can control your haltones with curves.  Choke the tonal spread, 5-95, or even 8-92% still looks great, add some contrast and the print is excellent.

Long Runs? - I still prefer Metal Halide here.  But for me that is 60-90k range, and we rarely see this in the states.  LED is so fast.  Great improvement on Vacuum drawdown speed, and instantaneous release of vacuum.  And lets face it $300-500 MH bulbs every year gets costly. 

Here is where I see some issues on LED exposure: Emulsion types: SBQ is the best, followed by diazo (which will undercut a bit more due to longer exposure time), and dual cures I don't recommend, although you can get an exposure and an image, its the time needed for both SBQ and diazo that are present in a dual cure with a short LED exposure that is underserved.  SBQ is exposed long before the diazo in a dual cure. So the exposure  is marginal for long runs, and post exposure doesn't help much.  I have a new emulsion.  I have buried this in the post on purpose.  PM me if you want to know more about the emulsion I used to dial in the 24 auto shop with completely clean screens through an auto reclaim using all ink systems.

[244]

As with cars, we all chose different ones.  We adjust to how they 'drive'.  As an emulsion manufacturer we did not buy into LED in the beginning.  Metal Halide wasn't broke, but LED has created a new screen making dynamic that is different from Metal Halide.  Both work, but with some adjustments needed.  I now sell LED mesh!  OK that's just a marketing slogan, but what I have found is that LED likes higher mesh counts.  A 150S vs a 110T, a 225/40 Yellow vs a 160W.   Our analysis of the exposure is this.  LED's do undercut a bit, but by moving mesh counts up a little the details are still decent.  Are the halftones as good as a Metal Halide, well no, they are a bit soft on the inside, but with a fast SBQ emulsion, complete exposure, (Thanks Greg, this needs reinforcement in our industry.) and an alternative mesh count that is higher but with similar print quality, your results will be fine.  Again, like cars, a new one may require rethinking on how you drive it, but it will take you where you want to go.

I recently dialed in a 24 auto shop with 2 STEII's, auto develop and more importantly auto reclaim.  Speed?  So fast a Metal Halide user could work 7 days a week and not equal what an STEII with multiple heads and exposure can output.  What helps is post exposure on SBQ products.  Without post exposure on LED you risk break down on discharge and water base, and more importantly any screen opener or solvents used in plastisol printing will lock in the emulsion without post exposure.  Auto reclaming machines aren't very strong IMO.  But with the right emulsion, post exposure, it works beautifully.  Post Exposure sources by strength and light quality: Sun from 10-3, 8k Metal Halide, 5k Metal Halide, LED.  The reason for LED in last here?  Spectral output.  Emulsions like mulit-spectral to completely cross link. Will the LED work for Post Exposure?  Yes for the majority of our print runs.  For 90K runs? not sure yet, untested.

As Pierre mentioned, you can control your haltones with curves.  Choke the tonal spread, 5-95, or even 8-92% still looks great, add some contrast and the print is excellent.

Long Runs? - I still prefer Metal Halide here.  But for me that is 60-90k range, and we rarely see this in the states.  LED is so fast.  Great improvement on Vacuum drawdown speed, and instantaneous release of vacuum.  And lets face it $300-500 MH bulbs every year gets costly. 

Here is where I see some issues on LED exposure: Emulsion types: SBQ is the best, followed by diazo (which will undercut a bit more due to longer exposure time), and dual cures I don't recommend, although you can get an exposure and an image, its the time needed for both SBQ and diazo that are present in a dual cure with a short LED exposure that is underserved.  SBQ is exposed long before the diazo in a dual cure. So the exposure  is marginal for long runs, and post exposure doesn't help much.  I have a new emulsion.  I have buried this in the post on purpose.  PM me if you want to know more about the emulsion I used to dial in the 24 auto shop with completely clean screens through an auto reclaim using all ink systems.

There is a new unit called the Starlight Gemini that exposes as well as post cures the emulsion for those super long runs. One of the largest printers in the country has tested it and ordered one for each of his plants. You will be seeing a lot of them in big shops shortly!!

[alan802]

I'll make sure that others don't make the same mistake. 

I wonder what this portends...

It means when someone asks me what I think of it I'll be honest and tell them the unit isn't near as good as our 15 year old Richmond Solarbeam with an 18 month old, aging bulb.  I know it's being subtly stated that there is probably something wrong with the people operating the unit and I'm sure there are some at Vastex, you included, that know more than I do about stencil development, but honestly, I've shot thousands of screens and take my education and knowledge of the processes very seriously and if I can't get a 10% dot from a 50lpi on a 280/34, 10%EOM, and numerous emulsions tested, bla bla bla, it's not the fault of the user in THIS SITUATION. 

I've thought about plugging the Richmond back in and even with the old bulb, crappy scratched up glass and awful control panel, and testing the units side by side and documenting it all for everyone on the forum to see.  But I'm sure there will be plenty of skepticism that the test was biased and most importantly the guy doing the testing is not near as smart as he thinks he is .  Have a good weekend folks.

[alan802]

I'll make sure that others don't make the same mistake. 

I wonder what this portends...

It means when someone asks me what I think of it I'll be honest and tell them the unit isn't near as good as our 15 year old Richmond Solarbeam with an 18 month old, aging bulb.  I know it's being subtly stated that there is probably something wrong with the people operating the unit and I'm sure there are some at Vastex, you included, that know more than I do about stencil development, but honestly, I've shot thousands of screens and take my education and knowledge of the processes very seriously and if I can't get a 10% dot from a 50lpi on a 280/34, 10%EOM, and numerous emulsions tested, bla bla bla, it's not the fault of the user in THIS SITUATION. 

I've thought about plugging the Richmond back in and even with the old bulb, crappy scratched up glass and awful control panel, and testing the units side by side and documenting it all for everyone on the forum to see.  But I'm sure there will be plenty of skepticism that the test was biased and most importantly the guy doing the testing is not near as smart as he thinks he is .  Have a good weekend folks.

You know what, I was pissed off when I wrote the original post a few days back and honestly I was mad at myself and spouted off something that really isn't fair to Vastex and not what I'm about.  I apologize for that, and I also got a little hot after reading Douglas's reply to that statement because quite frankly the whole situation has me miffed and at a few different things.  When I read the posts on the last few pages I couldn't help but feel like there were things being said that I feel were a bit disrespectful but really how does anyone that isn't in my circle know what I don't know or what my level of knowledge is with stencil development.  So I have to realize who I'm reading and why it is they are saying what they are saying and try to put it in proper perspective.  I'm not going to start a campaign to talk about how bad something is, I will give people my opinion if I"m asked and if I start a thread with a comparison I mentioned earlier I can promise everyone out there that it will be done as professionally as possible, as fair as possible and if am anything in this world from here to my end it will be honest. 

So with that out of the way, I'll state a few things that might help people understand that although there may be a few guys on the forum that know more than I do about this subject, I have been fortunate enough to have had guys, that some in the industry might recognize, in the shop that have witnessed testing screens with an exposure calculator and I'll mention a few of those names but the higher ups from a competitor I'll not mention.  Joe Clarke spent 4 days at our shop and I don't think anyone here would claim that he's lacking the knowledge and experience to not develop stencils properly.  Technical guys (not sales guys) from Kiwo and Chromaline and M&R have been in the shop.  John Deihl with Douthitt also spent some time visiting the shop and I also was fortunate to spend a lot of time at the ISS show last year.  And while I had the ears and eyes of these people we weren't spending all our time talking about the weather.  I realize it might be a bit douchie to name drop like that given the circumstances but I do feel like it's important to help some understand that a bunch of goons aren't farting around with an exposure unit and not getting good results.

[DouglasGrigar]

I'll make sure that others don't make the same mistake. 

I wonder what this portends...

It means when someone asks me what I think of it I'll be honest and tell them the unit isn't near as good as our 15 year old Richmond Solarbeam with an 18 month old, aging bulb.  I know it's being subtly stated that there is probably something wrong with the people operating the unit and I'm sure there are some at Vastex, you included, that know more than I do about stencil development, but honestly, I've shot thousands of screens and take my education and knowledge of the processes very seriously and if I can't get a 10% dot from a 50lpi on a 280/34, 10%EOM, and numerous emulsions tested, bla bla bla, it's not the fault of the user in THIS SITUATION. 

I've thought about plugging the Richmond back in and even with the old bulb, crappy scratched up glass and awful control panel, and testing the units side by side and documenting it all for everyone on the forum to see.  But I'm sure there will be plenty of skepticism that the test was biased and most importantly the guy doing the testing is not near as smart as he thinks he is .  Have a good weekend folks.

So let’s get down to it Alan - you and I have met face to face at least a few times so you have at least a bit of a read on what my personality is like and what I do in the industry.

In effect I get paid by my customers, shops like your’s, to tell them they are doing things wrong...

By that I improve efficiency, clear or adjust constraints, and educate - I actually do not like using the words “wrong” or “correct” or anything so open ended because it is not ACCURATE. To be straight with you accurate is all I am concerned with!

I apologize for that, and I also got a little hot after reading Douglas's reply to that statement because quite frankly the whole situation has me miffed and at a few different things.  When I read the posts on the last few pages I couldn't help but feel like there were things being said that I feel were a bit disrespectful

This is where problems start, there is disrespectful and then there is concern...

If I find things that do not fit in a shop because they are doing it “wrong” there is always a chance that pointing that out will be construed as offensive - this is where I have things above anyone in sales - I can just bypass that emotion and go right for the issue - logic and reason AND PROFIT over emotions. Emotions don’t make us money, they cost us money most of the time. This is why, and you and I have talked face to face about this, that I have a hard personal policy of FACT over even my own emotions - If I am doing something “wrong” then I want to know about it and change it ASAP. My classes are on version 32 now - in that as new things pop up and we in the industry learn more I change immediately with the new information.

What you have to know here is that I was concerned about “axe-grinding” and you know without a doubt - I want to HELP FIX any problems. Parading a list of luminaries to me is a grocery list, it’s not that I don’t know people it is that it does not get to the technical issue.

Just about anyone here (possibly you included) knows that I am most concerned about helping you FIX or identify your problem. There is a very good chance that you may well be going outside of the capabilities of the machine I have already pointed out that comparing a MH to LED is apples and oranges - of course they are different.

Don’t think that I am thinking or calling you an idiot - far from it - the problem is that we are somehow missing something - somewhere because 2+2=4 but somehow we are getting 3 or 5 - that means we are missing a number or two somewhere...

I want to find out what that is - and make sure we fix things and accurately identify problems.

I hope that helps you understand that I am NOT trying to be offensive, I want to help.

[Dottonedan]

All due respect, Danny Grunniger's printing is beating the competition and his screens are imaged with an M&R STE, which has a small strip of led's. So I don't see too much argument past that. If a starlight has even more juice, ummmmm, correct me if I'm wrong buuuut, wouldn't that be even more accuracy. The guys I know with Starlights do amazing work, won't have to buy bulbs and, none of them seem to be lacking detail. What am I missing?

Maybe it is the use of of CTS or good dmax films which allow for longer and more complete exposure of the emulsion.  Does LEd work, yes, but why do we see manufacturers formulating emulsion specifically for LED light. My belief is a multi-spectral light source makes for optimal exposure no matter the emulsion type.

I'm not sure really, why some are designing emulsions specifically for LED. It's not D-Max. It might be that if a current supplier only sells a specific brand...and that didn't fit exactly with their customer wants for exposure (times) on an LED device, then the emulsion manufacturer may want to create an emulsion brand with a faster time just for LED. Not sure.

I know I've not come across an emulsion that does not work at all. All work, but at different exposure times. That's the difference and may be what drives the emulsion Co's to create one for LED that they don't currently have. Some customers hear of these great exposure times and it's true (for another type of emulsion). They can be super fast exposures and many shops just want fast, and don't want or need  "the best" or the finest image quality.

I've tested the D-Max on our basic every day ink coverage on the I-Image and has been tested and ran at the 3.75 D-max  That's good enough and often better than many Epson films. In worse case scenario's, some of the printers out there have been using films where you can hold them up to the light and read a newspaper through them. You can even get heavier coverage than that with more passes, more heads, slower speed, high ink output resolution etc.

[alan802]

I'm a process guy (everyone here knows me well enough to know I like to break things down and know how each part makes up the whole), but I also understand that the goal/results are what matters when it's all said and done.  The end results are where I simply can't separate LED with MH because I'm putting some of the most finely coated screens anyone on this forum has ever seen (seriously, my guy makes better screens than I do, EOM is within 2 microns across the entire printing area), and regardless of the type of bulb is producing the UV I'm after, the RESULTS are poor with one unit and superb on the other.  So, I spend virtually an entire year, testing emulsions, moving our EOM all across the board, checking every part of the process to eliminate the exposure unit as the issue because I'm married to it at this point.  And how I come to the conclusion that it is the problem has more to do with me being able to roll up our old MH unit next to it, put the exact same screen on it and our issues are solved.  By saying that that is irrelevant and LED and MH are apples and oranges doesn't work for me.  I may very well be in the minority on that subject, and I'm not saying they aren't different, obviously they are, but the FINAL RESULT, the developed stencil doesn't give a damn about any of that. 

I'm not against making some adjustments here and there, obviously, we've been doing it for a year, but there wasn't any built in reasons or instructions that said anything about shortcomings or to put in different terms, a different animal with LED.  For example, every single screen that goes to press has an additional 36" of screen tape on the shirt side that never had to be done before.  The stencil begins to break down anywhere from 70-150 shirts but it eventually happens even though there are few shops that use less pressure than we do, few shops that use such a large variety of squeegee blade edges and durometer.  Because of the breakdown of the emulsion from the squeegee blade, that's one of the tricks we've had to put in place.  To get a more durable stencil (which you can see, burned 5 times longer didn't yield much better crosslinking on the squeegee side) we can post expose or just slap some tape on the screens, but post exposure hasn't worked as well as I would have thought.  I haven't done a lot of testing with post exposure and stencil durability due to the ease and speed of just putting some tape on a screen.

I just shot a halftone test of 50lpi, one side was shot for 12 seconds, the other 65 seconds.  The pic shows the white shirt that I rubbed on the squeegee side, very lightly, and only 3-4 swipes.  This particular screen is a 305, 5-6 micron EOM, under 10% EOMR.  The 12 sec exposure yielded decent halftones from 10%-90% but the 90 was sketchy and the 85% was solid.  The 65 sec side was good from 28%-90%.  I'm trying to get the pics from my microscope camera into my computer but the software isn't working properly so I can only post the phone pic right now.

So, I just don't understand where else I'm supposed to go with this.  Are we to jump through more hoops, continue to look for a different emulsion, increase the length of my tin foil hat antenna, even less EOM, be happy with what we're getting?  So if LED is the apple, MH is the orange, what more do I need to do to get what I think is a quality stencil? 

   

[screenprintguy]

I had a demo with the starlight 23x31 2 weeks ago with Ron Hopkins in our shop. I literally put my 2,000 psi power washer on the squeegee side of halftones and "tried" to blow them out and it just would not happen. We held detail in those screens on a job we have printed several times over the past few years that we were not getting with our tri light. I know there must be some "scientific" stuff that can be debated, but I think, depending on the unit in question, LED is the real deal. 4 second exposures on a I-Image CTS imaged screens that held the lowest % halftones that we've seen yet, and  haaaaaaaaad to be exposed and cross linked through and through to be able to take the beating of a pressure washer and not blow out those dots. Producing a final print better than we have in the past, arguing becomes a moot point. I have a couple friends in Orlando, primarily all water based printing, one uses murikami 7500 diazo, the other sp1400 diazo. Both use Starlights, neither have issues with emulsion break down, neither use hardeners. The guy using 7500 diazo does very large runs, 10,000 and above and never has issues. Loves his starlight and is NOT a bs'er. I mean just look at Danny G's work, not many can compete with his quality of printing and correct me if I'm wrong, but he images and exposes with an STE CTS LED, how much more could one want. The awards speak for themselves. Just my 2 cents, not trying to ruffle feathers but it just seems like the point is being missed that some of these other LED brands are just not up to the same standard as the Starlight.

[jvanick]

come on guys, somebody in texas has to have a Starlight that they can load in the back of their car and take to Alan's shop...  (or even let him bring some screens to image in their shop)

Anybody in my neck of the woods (chicago) is more than welcome to bring your screens, film, emulsion, etc to our shop to test with. 

[dirkdiggler]

agree totally Mike!