TSB
screen printing => Screen Making => Topic started by: Gilligan on January 07, 2013, 12:17:50 AM
-
accurip at 55 lpi 22.5 degrees on 150-S mesh.
this pattern is on the screen as well but not the film.
I do think the screen was over exposed as we lost the small dots.
anyway, looking to fix this if we can.
-
Try either going to a 260 mesh, or knock your line count down to 45.
In general, to best reduce your chances of moire interference patterns between halftones and mesh, you want a mesh count of at least 4.5 times your line count.
-
Since your screen has the pattern as well as the print I'd say yes this is moire. Put it on a 230/40 s mesh and you'll be good to go. We run 43 or 47 lpi on 180 and under mesh.
-
So basically dots are ending up being lined up on the thread itself at this lines per inch vs mesh count?
Just seems odd... but I guess that's because I haven't had to fight it before (never really used half-tones to their "full potential" yet.
-
55lpi on s thread 150/48 doesn't work well in my opinion.
45lpi is ok, but even at 45 you won't get super smooth halftones all the time, especially on large halftone areas.
-
55lpi=230 mesh
-
I don't think 150S will quite make it for you. Like the others said, go up to at least 230, and with a fade out to nothing, 260 sounds even better.
Steve
-
You can try to do 55 again on the 150s Murakami has certain angles that yeild no Moire on their mesh. Look on their website for them. Again don't count on it to work but they say it can be done
-
I think that experimentation with seemingly crazy mesh/line count, even with a whiz-bang product is great, but should be reserved for after basic halftones are mastered.
Gilly has no (or very little) frame of reference.
I encourage newbies to shoot for a ratio of 5:1 when possible, to further eliminate possible headaches. Heck, folks want to use higher-than-in-the-past mesh counts on solids nowadays. Halftones should be a no-brainer.
As for angles, sometimes the best laid plans of mice and men etc,
We try to stretch our mesh straight, but sometimes, even a prescribed angle on the film just isn't working. That's where the breathing room of the wider ratio differential helps save our butts.
Would mesh being cocked a half of a degree be visible?
Who out there with a hundred halftone screens under their belt has never needed to re-shoot a screen?
-
Frog is correct.
This is the highest count we have on hand and this is also a design just for us (not a real job).
We figured, "Why not".
Tell you what though... my guy is in love with the S mesh. It can definitely lay down some ink even though it's a higher count. That 150/40 throws down some ink! My guy was getting some one hit white and not even trying! I bid a job and my pricing sheet (excel) was glitching out so I could only see the basics of the formula and under bid it by a buck (still as high as most of you guys, maybe even higher) but I was also thinking it would be a PFP job... if we can do it in one hit then that makes up a bit of that buck, especially on the time side of things.
-
Gilly, you could have a bad wash out as well, some people don't really take the time to wash a screen out with halftones. After you wash a screen with halftones you can hold then up to a light and see the patterns and you can find that not all the dots washed out correct. A lot of people wash halftones soft to keep from blowing them out, but if the screen is burn correct you should not have a problem..just my 3 cents
Darryl
-
Can you tell if the threads are more or less fully blocking the smaller dots with a loupe? I'm surprised you got as many nice dots out as you did on that.
You gotta figure, if you fade to nothing, and you have sub-100 micron dots on a 48 micron thread, that's not going to work out quite right every time.
Can you limit the end of the fade to 5% instead of going all the way to 0%?
-
I figured all the way to 0% would be fine because whatever we got is what we got.
But the pattern is all around the design at about that percentage.
The thing is that the thread is WAY smaller than the "blocked" area. So something doesn't add up there.
I'm pretty sure I over exposed a bit... that is probably some of the problem. We exposed at our normal Light Units which is ~25% EOM on white 156.
I'd have all this dialed in but we are still struggling with coating consistently, so until then we just kind of wing it.
-
The real answer is you need to test a range of tones and using a loupe to examine exposed screens to find the tones that work on your mesh counts. Once you have a tonal range, try different angles (22.5 is NOT the correct angle) and again use the loupe to find the best range.
Once you've done all this, print a test shirt and give it to your wife, next time she makes tonal art, reference the test print and create tones that will fit inside that job on the mesh count you want to use.
-
Using an anti-halo mesh, yellow or orange will alleviate some of the undercutting, but I believe you are going to need a higher mesh count.
-
Gilly, just for fun and to get the moire thing figured out a little more take two films with 1/2 tones, stack them on you light table and slowly rotate the top film to see the moire come and go between the dots, it's more fun than spirograph! ;D. In some cases you can actually see moire between the film and mesh in the screen, although that's easier on an uncoated screen. Almost any 1/2 tone can be angled on almost any mesh to be moire free, but sticking to peoples suggestions above will be the best thing as there are other factors to consider too.
-
I figured all the way to 0% would be fine because whatever we got is what we got.
But the pattern is all around the design at about that percentage.
The thing is that the thread is WAY smaller than the "blocked" area. So something doesn't add up there.
I'm pretty sure I over exposed a bit... that is probably some of the problem. We exposed at our normal Light Units which is ~25% EOM on white 156. (???)
I'd have all this dialed in but we are still struggling with coating consistently, so until then we just kind of wing it.
Unfortunately, if you're having issues coating, that's what will need to change before consistent halftones will happen.
Even if the dot is twice or three times the size of the thread, and part of it resolves, it can and will still create the types of patterns you're seeing there. I mentioned a while back I thought of a thread and an opening as a smallish reasonable minimum dot size--on this mesh that would be about 170 microns--If my math isn't too addled, that's around 10% @ 55LPI.
I can only hope you're not trying to shoot 55LPI halftones on white 150 mesh. ;)
-
I find that shooting halftones on white mesh is trouble from jump street try using color mesh like yellow or orange...halftones burn great on those colors I,m guessing cuz the light does not bounce around as it does on white mesh.
Darryl
-
I was asked in a PM if 22.5 isn't right, then what is?
Due to the wide variations of mesh out there.. the angle that works is the one that fits your mesh, not mine.
You have to test for it plain and simple.
-
160 (or 156) are not as easy to find in dyed mesh, but they are out there.
Remember too, that there are many different colors and shades (yellow, orange, magenta that I have seen) so a new exposure calculation should be done on each!
-
I was asked in a PM if 22.5 isn't right, then what is?
Due to the wide variations of mesh out there.. the angle that works is the one that fits your mesh, not mine.
You have to test for it plain and simple.
26° and 61° work well too, but you do have to test. Sometimes, only part of the image produces moire, which we've usually solved with finer mesh counts.
Steve
-
Murakami's own chart places 22.5 degrees smack dab in the middle of "OK" as far as a 55LPI halftone is concerned. Although I'd be interested in a lesson as to why they're wrong, if they happen to be.
-
I think it has more to do with mesh count than angle in this case, throw that on a yellow 230 and he'd probably have no trouble.
-
For the record this was on 150-S thread. It ONLY comes in YELLOW.
I mentioned white mesh as in the only thing that we actually tested exposure on. We dialed in our exposure at about 25% EOM on white 156 mesh (static frame). I no longer use those frames.
I'm just looking to get my coating in step and then I'll get all of these exposure times dialed in as well as run some halftone dot test. I had printed out a sheet a long time ago with 25, 35 and 45 LPI... probably time to do it again at 55 LPI with all the percentages.
-
Dot Tone Dan had a test sheet on here that you could print out and burn with all types of halftones, check with him I,m sure he still might have it or I'll try and find mine.
Darryl
-
Thaks Darryl,
Yes, here it is. https://www.yousendit.com/download/UW13dFdhV3JRR2RqQThUQw (https://www.yousendit.com/download/UW13dFdhV3JRR2RqQThUQw)
download soon. It's good for 5 days.
It's a ripless test. meaning, the dots have been made for everyone already. I am not a big fan of it since it's not THE best way to test. It is set to 1200ppi for crisp clean dots, but still, using the dot tone bitmap conversions in photoshop as apposed to your actual rip is not going to be a 100% true test but will be very good. It may be the only test availabel so far for those who do not have a RIP.
I may have also included another set in grayscale for you to convert yourself using your own rip. I did this a few years ago and didn't open it again, so I'm not sure whats in it. You might find a cracker Jack toy.
Here is another good read from a post way back.
http://www.theshirtboard.com/index.php/topic,295.0.html (http://www.theshirtboard.com/index.php/topic,295.0.html)
-
I took a grey scale image set and just kept moving it physically on my document and printing it out at different LPI's on the same sheet of film.
Passing it through the printer multiple times.
This let me get the 3 different sets on one sheet for exposure.
-
If anyone has freehand 10 hit me up and ill share my test films.
-
I thought Murakami had a pdf about angles / LPI on different mesh counts.
Anyway we do 45LPI on 150S, if we need 50-55LPI we use 225S
The reason S thread can do higher LPI just fine is because of the much less thread in the way blocking the image/ink which means less moire. We did 50LPI on accident once sim process, the 150S had a bit of a pattern but the 4-5 colors over top got rid of it luckily and it worked fine.
I was asked in a PM if 22.5 isn't right, then what is?
Due to the wide variations of mesh out there.. the angle that works is the one that fits your mesh, not mine.
You have to test for it plain and simple.
I thought the 22.5 degree angle was the angle to choose because it is best at not causing a pattern with the mesh assuming the mesh is perpendicular to the frame (which is the goal, and hopefully most mesh is).
-
I thought the 22.5 degree angle was the angle to choose because it is best at not causing a pattern with the mesh assuming the mesh is perpendicular to the frame (which is the goal, and hopefully most mesh is).
According to Mr Coudray at halftonemastery.com, 22.5 or 67.5 are the angles, because at 0, 90 degrees we get interference from the thread. While at 45 deg we get interference from the knuckle of the two intersecting threads. So we must have an angle that is far away as possible from 0,45 or90 degree..... Thus this is where he came up with 22.5 or 67.5 for single angle print....
Good luck
Anthony.
P.S: I invested in the halftonemastery.com course. I believe this will save me several years of heartaches in experimentation and guess work.
-
It's right here:
http://murakamiscreen.com/wp-content/uploads/2012/07/HalftoneAnglestopreventmoire.pdf (http://murakamiscreen.com/wp-content/uploads/2012/07/HalftoneAnglestopreventmoire.pdf)
I can only hope I missed something, but it's amazing to me that it's being implied that a halftone angle change is capable of magically fixing the thread occlusion moire going on here...
-
It's right here:
[url]http://murakamiscreen.com/wp-content/uploads/2012/07/HalftoneAnglestopreventmoire.pdf[/url] ([url]http://murakamiscreen.com/wp-content/uploads/2012/07/HalftoneAnglestopreventmoire.pdf[/url])
I can only hope I missed something, but it's amazing to me that it's being implied that a halftone angle change is capable of magically fixing the thread occlusion moire going on here...
When you say "going on here", I'm assuming you're referring to Gilly's original issue of the moire.
The screen "angle" comments are 2ndary to his real issue (to me in my opinion) is obvious.
The main issue is the fact he's using 55lpi halftone on low mesh. The "general rule of thumb" as Frog pointed out is all math. Using between 4 and 5, (I prefer 5. will help guide you in a proper mesh to halftone selection.
150 mesh, divided by 4 = 37.5 line screen.
150 mesh, divided by 4.5 = 33 line screen.
150 mesh, divided by 5 = 30 line screen.
At the most, we should be using 38 line screen on a 150 mesh (when you want your fade to go all the way out without interference.
Can you use a higher line screen? Sure, many do, but many lose those small dots. At best, they get that jagged edge at the end where you see what looks saw toothed. That is the mesh blocking out the last few percentages of dots. This is so common that it's accepted as par for the course. I might not even fight it that much or re burn a screen at a lower line count in my own shop it's so common (if I saw that). It's not "bad" per say, just not optimum.
Here in Gilly's post, this is an extreme case and is why he's inquiring about it. He knows it aint right. Its a case where you do not want to be and leave your shop (in my opinion). This says to other printers, my shop made an error and let it go. Thats how I see this stuff (if I say it on a person). I think it says to the average buyer, (something looks wired on that but I can't put my finger on it).
So, with that said, someone might say, "so you would use that big of a dot?" No. I wouldn't. I would bump up my choice of screen mesh and use a finer dot that matches that. Then adjust how I print it so I still get the needed coverage.
My goal in a gradation (100% to 0%) is to maintain holding all dots as much as possible. Given that I can print most of my jobs on a 230 or higher, I would start there. If I wanted to use 55lpi for sure, it would go like this.
230 mesh, divided by 4 = 57.5 line screen.
230 mesh, divided by 4.5 = 51 line screen.
230 mesh, divided by 5 = 46 line screen.
I would not use anything lower than a 45 line screen on a 230 mesh. 55 is as high as I go and that is pushing it due to the fact that I also want my line screens to be the same on all, and I know I am only using the 230 as my base. Top colors are 300 mesh.
300 mesh, divided by 4 = 75 line screen.
300 mesh, divided by 4.5 = 66 line screen.
300 mesh, divided by 5 = 60 line screen
I would not use anything higher than a 60 line screen on a 300 mesh.
Being that I want all of my art to be the same line screen between a 230 and up, I know I want to use a lpi that will work on my lowest mesh and higher.
So, I round off either lower or higher and I choose 55 lpi on a 230 mesh and use it for my 300's as well. All of this increases my chances of maintaining as much of the smaller dots as I can.
This applies to S thread as well. The S thread does not jump you another 10 line screens but enables you to pick up another few more dots. You have to consider that you still have so many other variables that effect the dot quality as well such as "screen angle".
Once you know you're using the best choice of mesh (for the halftone used), you can still get moire from other variables such as the emulsion thickness, ink type, screen angles, etc.
22.5 is not (THE) correct one to use for (ALL) shops. Because lets say, your shop is a little more unique are using very thick ink on medium mesh, (thats one contributing factor), than add using a screen mesh that has a very low EOM Emulsion Over Mesh ratio, such as 1:1 of a thin fast pass of emulsion on a medium mesh or the opposite of that would be 2:2 of a very slow pass of emulsion on a medium mesh. Each of these can add to your chances of taking away from or adding to creating mesh opening to dot size/angle interference. Each shop is different but in general 22.5 does work for most.
-
Seemed like Gilly's initial issue of the moire was addressed with a number of people explaining you can't get a good fade to nothing on a 150 count mesh because of stencil/mesh parameters.
As far as what's going on here, I was surprised to read 'wrong angle' as an answer to the question, I was even more surprised to see people then discussing angle as if it could cause the issue as pictured, but I was by far the most interested in hearing about how Murakami's tech info is listing 'wrong' angle/LPI combos--although for some reason I doubt any proof of how or why is forthcoming.
I'm with you 100% on minimum dot size--the <5% tones aren't going to resolve on that mesh without the type of moire evidenced in the first photo--no matter what angle you put them at.
-
Seemed like Gilly's initial issue of the moire was addressed with a number of people explaining you can't get a good fade to nothing on a 150 count mesh because of stencil/mesh parameters.
More so, because of the halftones and not the stencil/mesh parameters. It can have nothing to do with the mesh itself, but a combination of the wrong mesh (with that specific halftone) of 55lpi used.
As far as what's going on here, I was surprised to read 'wrong angle' as an answer to the question, I was even more surprised to see people then discussing angle as if it could cause the issue as pictured, but I was by far the most interested in hearing about how Murakami's tech info is listing 'wrong' angle/LPI combos--although for some reason I doubt any proof of how or why is forthcoming.
I'm not so sure you read it as it is. It's saying (DO NOT USE THIS MESH) with 22.5 screen angles as THIS MESH does not work well at X angle. In some other areas, it shows that the same screen angle DOES work on a different mesh. The Yellow areas are what NOT to use. The Green area says OK to use. I think this is a very open tech sheet and very accurate. Not everyone goes this far to say "what angle" works on it and what does not but again, these are just the variables and test results from what (they) learned in there own shop. You may get better or worse results but somewhere close.
I'm with you 100% on minimum dot size--the <5% tones aren't going to resolve on that mesh without the type of moire evidenced in the first photo--no matter what angle you put them at.
True here, in this case with the 55lpi in the equation and no matter the screen angle used.
It's also true, that you CAN use this same 150 S thread mesh at 45 degrees (using the correct line screen that works with it like 30 line screen....and maybe still get moire, but switch it to 22.5 degrees and it's gone.
Just to be clear here, I'm not arguing your position. I am simply pointing out or pin pointing in more detail. Most everyone is right about not using that mesh, but I was explaining what line count should work best on that mesh.
-
Just so everyone knows... I'm still here and still reading every bit of it.
I never thought angles was the answer to my problem. I just wanted to make sure that it was in fact moire and at that point I assumed it was the mesh count. I did the math before... I still wanted to see what it looked like... it was something for us to play with and see what did what. I wanted to see some of the best shading I figured I would use and I didn't really care all that much if it didn't go below 20%, I just wanted to see it go.
I actually wanted to do a gray scale photo for my first test, but the wife had this design for some of our own shirts and wanted to see it printed... as we all know, they get what they want. ;)
-
Gilligan,
Sometimes it's best to 'just go for it' when experimenting.
I didn't read all the posts above, but you got some good advice.
Have some fun with it.
And it's much easier to get answers, if you know what questions to ask, and all that.
Cheers.
:)